obs-studio/obs/window-basic-main.cpp

1909 lines
46 KiB
C++
Raw Normal View History

/******************************************************************************
Copyright (C) 2013-2014 by Hugh Bailey <obs.jim@gmail.com>
Zachary Lund <admin@computerquip.com>
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
******************************************************************************/
2013-12-22 16:42:02 -08:00
#include <obs.hpp>
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
#include <QMessageBox>
#include <QShowEvent>
#include <QFileDialog>
#include <QNetworkRequest>
#include <QNetworkReply>
#include <util/dstr.h>
#include <util/util.hpp>
#include <util/platform.h>
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
#include <graphics/math-defs.h>
#include "obs-app.hpp"
#include "platform.hpp"
#include "window-basic-settings.hpp"
#include "window-namedialog.hpp"
#include "window-basic-source-select.hpp"
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
#include "window-basic-main.hpp"
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
#include "window-basic-properties.hpp"
#include "window-log-reply.hpp"
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
#include "qt-wrappers.hpp"
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
#include "display-helpers.hpp"
#include "volume-control.hpp"
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
#include "ui_OBSBasic.h"
#include <fstream>
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
#include <sstream>
#include <QScreen>
#include <QWindow>
#define PREVIEW_EDGE_SIZE 10
using namespace std;
2013-11-22 15:20:52 -08:00
Q_DECLARE_METATYPE(OBSScene);
Q_DECLARE_METATYPE(OBSSceneItem);
2014-05-15 17:40:53 -07:00
Q_DECLARE_METATYPE(order_movement);
OBSBasic::OBSBasic(QWidget *parent)
: OBSMainWindow (parent),
ui (new Ui::OBSBasic)
{
ui->setupUi(this);
connect(windowHandle(), &QWindow::screenChanged, [this]() {
struct obs_video_info ovi;
if (obs_get_video_info(&ovi))
ResizePreview(ovi.base_width, ovi.base_height);
});
stringstream name;
name << "OBS " << App()->GetVersionString();
blog(LOG_INFO, "%s", name.str().c_str());
setWindowTitle(QT_UTF8(name.str().c_str()));
}
static void SaveAudioDevice(const char *name, int channel, obs_data_t parent)
{
obs_source_t source = obs_get_output_source(channel);
if (!source)
return;
obs_data_t data = obs_save_source(source);
obs_data_setobj(parent, name, data);
obs_data_release(data);
obs_source_release(source);
}
static obs_data_t GenerateSaveData()
{
obs_data_t saveData = obs_data_create();
obs_data_array_t sourcesArray = obs_save_sources();
obs_source_t currentScene = obs_get_output_source(0);
const char *sceneName = obs_source_getname(currentScene);
SaveAudioDevice(DESKTOP_AUDIO_1, 1, saveData);
SaveAudioDevice(DESKTOP_AUDIO_2, 2, saveData);
SaveAudioDevice(AUX_AUDIO_1, 3, saveData);
SaveAudioDevice(AUX_AUDIO_2, 4, saveData);
SaveAudioDevice(AUX_AUDIO_3, 5, saveData);
obs_data_setstring(saveData, "current_scene", sceneName);
obs_data_setarray(saveData, "sources", sourcesArray);
obs_data_array_release(sourcesArray);
obs_source_release(currentScene);
return saveData;
}
void OBSBasic::ClearVolumeControls()
{
VolControl *control;
for (size_t i = 0; i < volumes.size(); i++) {
control = volumes[i];
delete control;
}
volumes.clear();
}
void OBSBasic::Save(const char *file)
{
obs_data_t saveData = GenerateSaveData();
const char *jsonData = obs_data_getjson(saveData);
/* TODO maybe a message box here? */
if (!os_quick_write_utf8_file(file, jsonData, strlen(jsonData), false))
blog(LOG_ERROR, "Could not save scene data to %s", file);
obs_data_release(saveData);
}
static void LoadAudioDevice(const char *name, int channel, obs_data_t parent)
{
obs_data_t data = obs_data_getobj(parent, name);
if (!data)
return;
obs_source_t source = obs_load_source(data);
if (source) {
obs_set_output_source(channel, source);
obs_source_release(source);
}
obs_data_release(data);
}
void OBSBasic::CreateDefaultScene()
{
obs_scene_t scene = obs_scene_create(Str("Basic.Scene"));
obs_source_t source = obs_scene_getsource(scene);
obs_add_source(source);
#ifdef __APPLE__
source = obs_source_create(OBS_SOURCE_TYPE_INPUT, "display_capture",
Str("Basic.DisplayCapture"), NULL);
if (source) {
obs_scene_add(scene, source);
obs_add_source(source);
obs_source_release(source);
}
#endif
obs_set_output_source(0, obs_scene_getsource(scene));
obs_scene_release(scene);
}
void OBSBasic::Load(const char *file)
{
if (!file) {
blog(LOG_ERROR, "Could not find file %s", file);
return;
}
BPtr<char> jsonData = os_quick_read_utf8_file(file);
if (!jsonData) {
CreateDefaultScene();
return;
}
obs_data_t data = obs_data_create_from_json(jsonData);
obs_data_array_t sources = obs_data_getarray(data, "sources");
const char *sceneName = obs_data_getstring(data, "current_scene");
obs_source_t curScene;
LoadAudioDevice(DESKTOP_AUDIO_1, 1, data);
LoadAudioDevice(DESKTOP_AUDIO_2, 2, data);
LoadAudioDevice(AUX_AUDIO_1, 3, data);
LoadAudioDevice(AUX_AUDIO_2, 4, data);
LoadAudioDevice(AUX_AUDIO_3, 5, data);
obs_load_sources(sources);
curScene = obs_get_source_by_name(sceneName);
obs_set_output_source(0, curScene);
obs_source_release(curScene);
obs_data_array_release(sources);
obs_data_release(data);
}
static inline bool HasAudioDevices(const char *source_id)
{
const char *output_id = source_id;
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
obs_properties_t props = obs_get_source_properties(
OBS_SOURCE_TYPE_INPUT, output_id);
size_t count = 0;
if (!props)
return false;
obs_property_t devices = obs_properties_get(props, "device_id");
if (devices)
count = obs_property_list_item_count(devices);
obs_properties_destroy(props);
return count != 0;
}
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
static void OBSStartStreaming(void *data, calldata_t params)
{
UNUSED_PARAMETER(params);
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"StreamingStart");
}
static void OBSStopStreaming(void *data, calldata_t params)
{
int code = (int)calldata_int(params, "code");
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"StreamingStop", Q_ARG(int, code));
}
static void OBSStopRecording(void *data, calldata_t params)
{
UNUSED_PARAMETER(params);
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"RecordingStop");
}
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
#define SERVICE_PATH "obs-studio/basic/service.json"
void OBSBasic::SaveService()
{
if (!service)
return;
BPtr<char> serviceJsonPath(os_get_config_path(SERVICE_PATH));
if (!serviceJsonPath)
return;
obs_data_t data = obs_data_create();
obs_data_t settings = obs_service_get_settings(service);
obs_data_setstring(data, "type", obs_service_gettype(service));
obs_data_setobj(data, "settings", settings);
const char *json = obs_data_getjson(data);
os_quick_write_utf8_file(serviceJsonPath, json, strlen(json), false);
obs_data_release(settings);
obs_data_release(data);
}
bool OBSBasic::LoadService()
{
const char *type;
BPtr<char> serviceJsonPath(os_get_config_path(SERVICE_PATH));
if (!serviceJsonPath)
return false;
BPtr<char> jsonText = os_quick_read_utf8_file(serviceJsonPath);
if (!jsonText)
return false;
obs_data_t data = obs_data_create_from_json(jsonText);
obs_data_set_default_string(data, "type", "rtmp_common");
type = obs_data_getstring(data, "type");
obs_data_t settings = obs_data_getobj(data, "settings");
service = obs_service_create(type, "default", settings);
obs_data_release(settings);
obs_data_release(data);
return !!service;
}
bool OBSBasic::InitOutputs()
{
fileOutput = obs_output_create("flv_output", "default", nullptr);
if (!fileOutput)
return false;
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
streamOutput = obs_output_create("rtmp_output", "default", nullptr);
if (!streamOutput)
return false;
signal_handler_connect(obs_output_signalhandler(streamOutput),
"start", OBSStartStreaming, this);
signal_handler_connect(obs_output_signalhandler(streamOutput),
"stop", OBSStopStreaming, this);
signal_handler_connect(obs_output_signalhandler(fileOutput),
"stop", OBSStopRecording, this);
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
return true;
}
bool OBSBasic::InitEncoders()
{
x264 = obs_video_encoder_create("obs_x264", "h264", nullptr);
if (!x264)
return false;
2014-05-22 03:07:17 -07:00
aac = obs_audio_encoder_create("libfdk_aac", "aac", nullptr);
if(!aac)
aac = obs_audio_encoder_create("ffmpeg_aac", "aac", nullptr);
if (!aac)
return false;
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
return true;
}
bool OBSBasic::InitService()
{
if (LoadService())
return true;
service = obs_service_create("rtmp_common", nullptr, nullptr);
if (!service)
return false;
return true;
}
bool OBSBasic::InitBasicConfigDefaults()
{
bool hasDesktopAudio = HasAudioDevices(App()->OutputAudioSource());
bool hasInputAudio = HasAudioDevices(App()->InputAudioSource());
config_set_default_int(basicConfig, "Window", "PosX", -1);
config_set_default_int(basicConfig, "Window", "PosY", -1);
config_set_default_int(basicConfig, "Window", "SizeX", -1);
config_set_default_int(basicConfig, "Window", "SizeY", -1);
vector<MonitorInfo> monitors;
GetMonitors(monitors);
if (!monitors.size()) {
OBSErrorBox(NULL, "There appears to be no monitors. Er, this "
"technically shouldn't be possible.");
return false;
}
uint32_t cx = monitors[0].cx;
uint32_t cy = monitors[0].cy;
/* TODO: temporary */
config_set_default_string(basicConfig, "SimpleOutput", "FilePath",
GetDefaultVideoSavePath().c_str());
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
config_set_default_uint (basicConfig, "SimpleOutput", "VBitrate",
2500);
config_set_default_uint (basicConfig, "SimpleOutput", "ABitrate", 128);
config_set_default_uint (basicConfig, "Video", "BaseCX", cx);
config_set_default_uint (basicConfig, "Video", "BaseCY", cy);
cx = cx * 10 / 15;
cy = cy * 10 / 15;
config_set_default_uint (basicConfig, "Video", "OutputCX", cx);
config_set_default_uint (basicConfig, "Video", "OutputCY", cy);
config_set_default_uint (basicConfig, "Video", "FPSType", 0);
config_set_default_string(basicConfig, "Video", "FPSCommon", "30");
config_set_default_uint (basicConfig, "Video", "FPSInt", 30);
config_set_default_uint (basicConfig, "Video", "FPSNum", 30);
config_set_default_uint (basicConfig, "Video", "FPSDen", 1);
config_set_default_uint (basicConfig, "Audio", "SampleRate", 44100);
config_set_default_string(basicConfig, "Audio", "ChannelSetup",
"Stereo");
config_set_default_uint (basicConfig, "Audio", "BufferingTime", 1000);
config_set_default_string(basicConfig, "Audio", "DesktopDevice1",
hasDesktopAudio ? "default" : "disabled");
config_set_default_string(basicConfig, "Audio", "DesktopDevice2",
"disabled");
config_set_default_string(basicConfig, "Audio", "AuxDevice1",
hasInputAudio ? "default" : "disabled");
config_set_default_string(basicConfig, "Audio", "AuxDevice2",
"disabled");
config_set_default_string(basicConfig, "Audio", "AuxDevice3",
"disabled");
return true;
}
bool OBSBasic::InitBasicConfig()
{
BPtr<char> configPath(os_get_config_path("obs-studio/basic/basic.ini"));
int code = basicConfig.Open(configPath, CONFIG_OPEN_ALWAYS);
if (code != CONFIG_SUCCESS) {
OBSErrorBox(NULL, "Failed to open basic.ini: %d", code);
return false;
}
return InitBasicConfigDefaults();
}
void OBSBasic::InitOBSCallbacks()
{
signal_handler_connect(obs_signalhandler(), "source_add",
OBSBasic::SourceAdded, this);
signal_handler_connect(obs_signalhandler(), "source_remove",
OBSBasic::SourceRemoved, this);
signal_handler_connect(obs_signalhandler(), "channel_change",
OBSBasic::ChannelChanged, this);
signal_handler_connect(obs_signalhandler(), "source_activate",
OBSBasic::SourceActivated, this);
signal_handler_connect(obs_signalhandler(), "source_deactivate",
OBSBasic::SourceDeactivated, this);
}
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
void OBSBasic::InitPrimitives()
{
gs_entercontext(obs_graphics());
gs_renderstart(true);
gs_vertex2f(0.0f, 0.0f);
gs_vertex2f(0.0f, 1.0f);
gs_vertex2f(1.0f, 1.0f);
gs_vertex2f(1.0f, 0.0f);
gs_vertex2f(0.0f, 0.0f);
box = gs_rendersave();
gs_renderstart(true);
for (int i = 0; i <= 360; i += (360/20)) {
float pos = RAD(float(i));
gs_vertex2f(cosf(pos), sinf(pos));
}
circle = gs_rendersave();
gs_leavecontext();
}
void OBSBasic::OBSInit()
{
BPtr<char> savePath(os_get_config_path("obs-studio/basic/scenes.json"));
/* make sure it's fully displayed before doing any initialization */
show();
App()->processEvents();
if (!obs_startup(App()->GetLocale()))
throw "Failed to initialize libobs";
if (!InitBasicConfig())
throw "Failed to load basic.ini";
if (!ResetVideo())
throw "Failed to initialize video";
if (!ResetAudio())
throw "Failed to initialize audio";
InitOBSCallbacks();
/* TODO: this is a test, all modules will be searched for and loaded
* automatically later */
obs_load_module("test-input");
obs_load_module("obs-ffmpeg");
2014-05-22 03:07:17 -07:00
obs_load_module("obs-libfdk");
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
obs_load_module("obs-x264");
obs_load_module("obs-outputs");
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_load_module("rtmp-services");
#ifdef __APPLE__
obs_load_module("mac-avcapture");
obs_load_module("mac-capture");
#elif _WIN32
obs_load_module("win-wasapi");
obs_load_module("win-capture");
obs_load_module("win-dshow");
#else
obs_load_module("linux-xshm");
2014-04-28 17:59:53 -07:00
obs_load_module("linux-xcomposite");
obs_load_module("linux-pulseaudio");
obs_load_module("linux-v4l2");
#endif
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
if (!InitOutputs())
throw "Failed to initialize outputs";
if (!InitEncoders())
throw "Failed to initialize encoders";
if (!InitService())
throw "Failed to initialize service";
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
InitPrimitives();
Load(savePath);
ResetAudioDevices();
loaded = true;
}
OBSBasic::~OBSBasic()
{
BPtr<char> savePath(os_get_config_path("obs-studio/basic/scenes.json"));
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
SaveService();
Save(savePath);
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
/* XXX: any obs data must be released before calling obs_shutdown.
* currently, we can't automate this with C++ RAII because of the
* delicate nature of obs_shutdown needing to be freed before the UI
* can be freed, and we have no control over the destruction order of
* the Qt UI stuff, so we have to manually clear any references to
* libobs. */
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
if (properties)
delete properties;
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
if (transformWindow)
delete transformWindow;
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
ClearVolumeControls();
ui->sources->clear();
ui->scenes->clear();
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
gs_entercontext(obs_graphics());
vertexbuffer_destroy(box);
vertexbuffer_destroy(circle);
gs_leavecontext();
obs_shutdown();
}
OBSScene OBSBasic::GetCurrentScene()
{
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QListWidgetItem *item = ui->scenes->currentItem();
return item ? item->data(Qt::UserRole).value<OBSScene>() : nullptr;
}
OBSSceneItem OBSBasic::GetCurrentSceneItem()
{
QListWidgetItem *item = ui->sources->currentItem();
return item ? item->data(Qt::UserRole).value<OBSSceneItem>() : nullptr;
}
void OBSBasic::UpdateSources(OBSScene scene)
{
ui->sources->clear();
obs_scene_enum_items(scene,
[] (obs_scene_t scene, obs_sceneitem_t item, void *p)
{
OBSBasic *window = static_cast<OBSBasic*>(p);
window->InsertSceneItem(item);
UNUSED_PARAMETER(scene);
return true;
}, this);
}
void OBSBasic::InsertSceneItem(obs_sceneitem_t item)
{
obs_source_t source = obs_sceneitem_getsource(item);
const char *name = obs_source_getname(source);
QListWidgetItem *listItem = new QListWidgetItem(QT_UTF8(name));
listItem->setData(Qt::UserRole,
QVariant::fromValue(OBSSceneItem(item)));
ui->sources->insertItem(0, listItem);
ui->sources->setCurrentRow(0);
/* if the source was just created, open properties dialog */
if (sourceSceneRefs[source] == 0 && loaded) {
delete properties;
properties = new OBSBasicProperties(this, source);
properties->Init();
}
}
/* Qt callbacks for invokeMethod */
void OBSBasic::AddScene(OBSSource source)
{
const char *name = obs_source_getname(source);
obs_scene_t scene = obs_scene_fromsource(source);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QListWidgetItem *item = new QListWidgetItem(QT_UTF8(name));
item->setData(Qt::UserRole, QVariant::fromValue(OBSScene(scene)));
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
ui->scenes->addItem(item);
signal_handler_t handler = obs_source_signalhandler(source);
signal_handler_connect(handler, "item_add",
OBSBasic::SceneItemAdded, this);
signal_handler_connect(handler, "item_remove",
OBSBasic::SceneItemRemoved, this);
2014-05-15 17:40:53 -07:00
signal_handler_connect(handler, "item_move_up",
OBSBasic::SceneItemMoveUp, this);
signal_handler_connect(handler, "item_move_down",
OBSBasic::SceneItemMoveDown, this);
signal_handler_connect(handler, "item_move_top",
OBSBasic::SceneItemMoveTop, this);
signal_handler_connect(handler, "item_move_bottom",
OBSBasic::SceneItemMoveBottom, this);
}
void OBSBasic::RemoveScene(OBSSource source)
2013-12-28 21:29:13 -08:00
{
const char *name = obs_source_getname(source);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QListWidgetItem *sel = ui->scenes->currentItem();
QList<QListWidgetItem*> items = ui->scenes->findItems(QT_UTF8(name),
Qt::MatchExactly);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
if (sel != nullptr) {
if (items.contains(sel))
ui->sources->clear();
delete sel;
}
}
void OBSBasic::AddSceneItem(OBSSceneItem item)
{
obs_scene_t scene = obs_sceneitem_getscene(item);
obs_source_t source = obs_sceneitem_getsource(item);
if (GetCurrentScene() == scene)
InsertSceneItem(item);
sourceSceneRefs[source] = sourceSceneRefs[source] + 1;
}
void OBSBasic::RemoveSceneItem(OBSSceneItem item)
{
obs_scene_t scene = obs_sceneitem_getscene(item);
if (GetCurrentScene() == scene) {
for (int i = 0; i < ui->sources->count(); i++) {
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QListWidgetItem *listItem = ui->sources->item(i);
QVariant userData = listItem->data(Qt::UserRole);
if (userData.value<OBSSceneItem>() == item) {
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
delete listItem;
break;
}
}
}
obs_source_t source = obs_sceneitem_getsource(item);
int scenes = sourceSceneRefs[source] - 1;
sourceSceneRefs[source] = scenes;
if (scenes == 0) {
obs_source_remove(source);
sourceSceneRefs.erase(source);
}
}
void OBSBasic::UpdateSceneSelection(OBSSource source)
{
if (source) {
obs_source_type type;
obs_source_gettype(source, &type, NULL);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
obs_scene_t scene = obs_scene_fromsource(source);
const char *name = obs_source_getname(source);
if (!scene)
return;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QList<QListWidgetItem*> items =
ui->scenes->findItems(QT_UTF8(name), Qt::MatchExactly);
if (items.count()) {
sceneChanging = true;
ui->scenes->setCurrentItem(items.first());
sceneChanging = false;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
UpdateSources(scene);
}
2013-12-28 21:29:13 -08:00
}
}
2014-05-15 17:40:53 -07:00
void OBSBasic::MoveSceneItem(OBSSceneItem item, order_movement movement)
{
OBSScene scene = obs_sceneitem_getscene(item);
if (scene != GetCurrentScene())
return;
int curRow = ui->sources->currentRow();
if (curRow == -1)
return;
QListWidgetItem *listItem = ui->sources->takeItem(curRow);
switch (movement) {
case ORDER_MOVE_UP:
if (curRow > 0)
curRow--;
break;
case ORDER_MOVE_DOWN:
if (curRow < ui->sources->count())
curRow++;
break;
case ORDER_MOVE_TOP:
curRow = 0;
break;
case ORDER_MOVE_BOTTOM:
curRow = ui->sources->count();
break;
}
ui->sources->insertItem(curRow, listItem);
ui->sources->setCurrentRow(curRow);
}
void OBSBasic::ActivateAudioSource(OBSSource source)
{
VolControl *vol = new VolControl(source);
volumes.push_back(vol);
ui->volumeWidgets->layout()->addWidget(vol);
}
void OBSBasic::DeactivateAudioSource(OBSSource source)
{
for (size_t i = 0; i < volumes.size(); i++) {
if (volumes[i]->GetSource() == source) {
delete volumes[i];
volumes.erase(volumes.begin() + i);
break;
}
}
}
/* OBS Callbacks */
void OBSBasic::SceneItemAdded(void *data, calldata_t params)
{
OBSBasic *window = static_cast<OBSBasic*>(data);
obs_sceneitem_t item = (obs_sceneitem_t)calldata_ptr(params, "item");
2013-12-28 21:29:13 -08:00
QMetaObject::invokeMethod(window, "AddSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)));
2013-12-28 21:29:13 -08:00
}
void OBSBasic::SceneItemRemoved(void *data, calldata_t params)
{
OBSBasic *window = static_cast<OBSBasic*>(data);
obs_sceneitem_t item = (obs_sceneitem_t)calldata_ptr(params, "item");
QMetaObject::invokeMethod(window, "RemoveSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)));
}
void OBSBasic::SourceAdded(void *data, calldata_t params)
{
OBSBasic *window = static_cast<OBSBasic*>(data);
obs_source_t source = (obs_source_t)calldata_ptr(params, "source");
if (obs_scene_fromsource(source) != NULL)
QMetaObject::invokeMethod(window,
"AddScene",
Q_ARG(OBSSource, OBSSource(source)));
else
window->sourceSceneRefs[source] = 0;
}
void OBSBasic::SourceRemoved(void *data, calldata_t params)
{
obs_source_t source = (obs_source_t)calldata_ptr(params, "source");
if (obs_scene_fromsource(source) != NULL)
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"RemoveScene",
Q_ARG(OBSSource, OBSSource(source)));
}
void OBSBasic::SourceActivated(void *data, calldata_t params)
{
obs_source_t source = (obs_source_t)calldata_ptr(params, "source");
uint32_t flags = obs_source_get_output_flags(source);
if (flags & OBS_SOURCE_AUDIO)
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"ActivateAudioSource",
Q_ARG(OBSSource, OBSSource(source)));
}
void OBSBasic::SourceDeactivated(void *data, calldata_t params)
{
obs_source_t source = (obs_source_t)calldata_ptr(params, "source");
uint32_t flags = obs_source_get_output_flags(source);
if (flags & OBS_SOURCE_AUDIO)
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"DeactivateAudioSource",
Q_ARG(OBSSource, OBSSource(source)));
}
void OBSBasic::ChannelChanged(void *data, calldata_t params)
{
obs_source_t source = (obs_source_t)calldata_ptr(params, "source");
uint32_t channel = (uint32_t)calldata_int(params, "channel");
if (channel == 0)
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"UpdateSceneSelection",
Q_ARG(OBSSource, OBSSource(source)));
}
void OBSBasic::DrawBackdrop(float cx, float cy)
{
if (!box)
return;
effect_t solid = obs_get_solid_effect();
eparam_t color = effect_getparambyname(solid, "color");
technique_t tech = effect_gettechnique(solid, "Solid");
vec4 colorVal;
vec4_set(&colorVal, 0.0f, 0.0f, 0.0f, 1.0f);
effect_setvec4(color, &colorVal);
technique_begin(tech);
technique_beginpass(tech, 0);
gs_matrix_push();
gs_matrix_identity();
gs_matrix_scale3f(float(cx), float(cy), 1.0f);
gs_load_vertexbuffer(box);
gs_draw(GS_TRISTRIP, 0, 0);
gs_matrix_pop();
technique_endpass(tech);
technique_end(tech);
gs_load_vertexbuffer(nullptr);
}
void OBSBasic::RenderMain(void *data, uint32_t cx, uint32_t cy)
{
OBSBasic *window = static_cast<OBSBasic*>(data);
obs_video_info ovi;
obs_get_video_info(&ovi);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
window->previewCX = int(window->previewScale * float(ovi.base_width));
window->previewCY = int(window->previewScale * float(ovi.base_height));
gs_viewport_push();
gs_projection_push();
/* --------------------------------------- */
gs_ortho(0.0f, float(ovi.base_width), 0.0f, float(ovi.base_height),
-100.0f, 100.0f);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
gs_setviewport(window->previewX, window->previewY,
window->previewCX, window->previewCY);
window->DrawBackdrop(float(ovi.base_width), float(ovi.base_height));
obs_render_main_view();
gs_load_vertexbuffer(nullptr);
/* --------------------------------------- */
float right = float(window->ui->preview->width()) - window->previewX;
float bottom = float(window->ui->preview->height()) - window->previewY;
gs_ortho(-window->previewX, right,
-window->previewY, bottom,
-100.0f, 100.0f);
gs_resetviewport();
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
window->ui->preview->DrawSceneEditing();
/* --------------------------------------- */
gs_projection_pop();
gs_viewport_pop();
UNUSED_PARAMETER(cx);
UNUSED_PARAMETER(cy);
}
2014-05-15 17:40:53 -07:00
void OBSBasic::SceneItemMoveUp(void *data, calldata_t params)
{
OBSSceneItem item = (obs_sceneitem_t)calldata_ptr(params, "item");
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"MoveSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)),
Q_ARG(order_movement, ORDER_MOVE_UP));
}
void OBSBasic::SceneItemMoveDown(void *data, calldata_t params)
{
OBSSceneItem item = (obs_sceneitem_t)calldata_ptr(params, "item");
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"MoveSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)),
Q_ARG(order_movement, ORDER_MOVE_DOWN));
}
void OBSBasic::SceneItemMoveTop(void *data, calldata_t params)
{
OBSSceneItem item = (obs_sceneitem_t)calldata_ptr(params, "item");
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"MoveSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)),
Q_ARG(order_movement, ORDER_MOVE_TOP));
}
void OBSBasic::SceneItemMoveBottom(void *data, calldata_t params)
{
OBSSceneItem item = (obs_sceneitem_t)calldata_ptr(params, "item");
QMetaObject::invokeMethod(static_cast<OBSBasic*>(data),
"MoveSceneItem",
Q_ARG(OBSSceneItem, OBSSceneItem(item)),
Q_ARG(order_movement, ORDER_MOVE_BOTTOM));
}
/* Main class functions */
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_service_t OBSBasic::GetService()
{
if (!service)
service = obs_service_create("rtmp_common", NULL, NULL);
return service;
}
void OBSBasic::SetService(obs_service_t newService)
{
if (newService) {
if (service)
obs_service_destroy(service);
service = newService;
}
}
bool OBSBasic::ResetVideo()
{
struct obs_video_info ovi;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
GetConfigFPS(ovi.fps_num, ovi.fps_den);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
ovi.graphics_module = App()->GetRenderModule();
ovi.base_width = (uint32_t)config_get_uint(basicConfig,
"Video", "BaseCX");
ovi.base_height = (uint32_t)config_get_uint(basicConfig,
"Video", "BaseCY");
ovi.output_width = (uint32_t)config_get_uint(basicConfig,
"Video", "OutputCX");
ovi.output_height = (uint32_t)config_get_uint(basicConfig,
"Video", "OutputCY");
ovi.output_format = VIDEO_FORMAT_NV12;
ovi.adapter = 0;
ovi.gpu_conversion = true;
QTToGSWindow(ui->preview->winId(), ovi.window);
//required to make opengl display stuff on osx(?)
2013-12-31 03:02:07 -08:00
ResizePreview(ovi.base_width, ovi.base_height);
QSize size = GetPixelSize(ui->preview);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
ovi.window_width = size.width();
ovi.window_height = size.height();
2013-12-31 03:02:07 -08:00
if (!obs_reset_video(&ovi))
return false;
obs_add_draw_callback(OBSBasic::RenderMain, this);
return true;
}
2013-12-31 03:02:07 -08:00
bool OBSBasic::ResetAudio()
{
struct audio_output_info ai;
ai.name = "Main Audio Track";
ai.format = AUDIO_FORMAT_FLOAT;
ai.samples_per_sec = config_get_uint(basicConfig, "Audio",
"SampleRate");
const char *channelSetupStr = config_get_string(basicConfig,
"Audio", "ChannelSetup");
if (strcmp(channelSetupStr, "Mono") == 0)
ai.speakers = SPEAKERS_MONO;
else
ai.speakers = SPEAKERS_STEREO;
ai.buffer_ms = config_get_uint(basicConfig, "Audio", "BufferingTime");
return obs_reset_audio(&ai);
}
void OBSBasic::ResetAudioDevice(const char *sourceId, const char *deviceName,
const char *deviceDesc, int channel)
{
const char *deviceId = config_get_string(basicConfig, "Audio",
deviceName);
obs_source_t source;
obs_data_t settings;
bool same = false;
source = obs_get_output_source(channel);
if (source) {
settings = obs_source_getsettings(source);
const char *curId = obs_data_getstring(settings, "device_id");
same = (strcmp(curId, deviceId) == 0);
obs_data_release(settings);
obs_source_release(source);
}
if (!same)
obs_set_output_source(channel, nullptr);
if (!same && strcmp(deviceId, "disabled") != 0) {
obs_data_t settings = obs_data_create();
obs_data_setstring(settings, "device_id", deviceId);
source = obs_source_create(OBS_SOURCE_TYPE_INPUT,
sourceId, deviceDesc, settings);
obs_data_release(settings);
obs_set_output_source(channel, source);
obs_source_release(source);
}
}
void OBSBasic::ResetAudioDevices()
{
ResetAudioDevice(App()->OutputAudioSource(), "DesktopDevice1",
Str("Basic.DesktopDevice1"), 1);
ResetAudioDevice(App()->OutputAudioSource(), "DesktopDevice2",
Str("Basic.DesktopDevice2"), 2);
ResetAudioDevice(App()->InputAudioSource(), "AuxDevice1",
Str("Basic.AuxDevice1"), 3);
ResetAudioDevice(App()->InputAudioSource(), "AuxDevice2",
Str("Basic.AuxDevice2"), 4);
ResetAudioDevice(App()->InputAudioSource(), "AuxDevice3",
Str("Basic.AuxDevice3"), 5);
}
2013-12-31 03:02:07 -08:00
void OBSBasic::ResizePreview(uint32_t cx, uint32_t cy)
{
QSize targetSize;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
/* resize preview panel to fix to the top section of the window */
targetSize = GetPixelSize(ui->preview);
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
GetScaleAndCenterPos(int(cx), int(cy),
targetSize.width() - PREVIEW_EDGE_SIZE * 2,
targetSize.height() - PREVIEW_EDGE_SIZE * 2,
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
previewX, previewY, previewScale);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
previewX += float(PREVIEW_EDGE_SIZE);
previewY += float(PREVIEW_EDGE_SIZE);
if (isVisible()) {
if (resizeTimer)
killTimer(resizeTimer);
resizeTimer = startTimer(100);
}
2013-12-31 03:02:07 -08:00
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::closeEvent(QCloseEvent *event)
2013-12-31 03:02:07 -08:00
{
QWidget::closeEvent(event);
if (!event->isAccepted())
return;
// remove draw callback in case our drawable surfaces go away before
// the destructor gets called
obs_remove_draw_callback(OBSBasic::RenderMain, this);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::changeEvent(QEvent *event)
{
/* TODO */
UNUSED_PARAMETER(event);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::resizeEvent(QResizeEvent *event)
{
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
struct obs_video_info ovi;
if (obs_get_video_info(&ovi))
ResizePreview(ovi.base_width, ovi.base_height);
UNUSED_PARAMETER(event);
}
void OBSBasic::timerEvent(QTimerEvent *event)
{
if (event->timerId() == resizeTimer) {
killTimer(resizeTimer);
resizeTimer = 0;
QSize size = GetPixelSize(ui->preview);
obs_resize(size.width(), size.height());
}
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_action_New_triggered()
{
/* TODO */
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_action_Open_triggered()
{
/* TODO */
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_action_Save_triggered()
{
/* TODO */
}
void OBSBasic::on_action_Settings_triggered()
{
OBSBasicSettings settings(this);
settings.exec();
}
void OBSBasic::on_scenes_currentItemChanged(QListWidgetItem *current,
QListWidgetItem *prev)
{
obs_source_t source = NULL;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
if (sceneChanging)
return;
if (current) {
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
obs_scene_t scene;
scene = current->data(Qt::UserRole).value<OBSScene>();
source = obs_scene_getsource(scene);
}
/* TODO: allow transitions */
obs_set_output_source(0, source);
UNUSED_PARAMETER(prev);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_scenes_customContextMenuRequested(const QPoint &pos)
{
/* TODO */
UNUSED_PARAMETER(pos);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionAddScene_triggered()
{
string name;
QString format{QTStr("Basic.Main.DefaultSceneName.Text")};
int i = 1;
QString placeHolderText = format.arg(i);
2014-05-14 13:20:08 -07:00
obs_source_t source = nullptr;
while ((source = obs_get_source_by_name(QT_TO_UTF8(placeHolderText)))) {
obs_source_release(source);
placeHolderText = format.arg(++i);
2014-05-14 13:20:08 -07:00
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
bool accepted = NameDialog::AskForName(this,
QTStr("Basic.Main.AddSceneDlg.Title"),
QTStr("Basic.Main.AddSceneDlg.Text"),
name,
placeHolderText);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
if (accepted) {
if (name.empty()) {
QMessageBox::information(this,
QTStr("NoNameEntered"),
QTStr("NoNameEntered"));
on_actionAddScene_triggered();
return;
}
obs_source_t source = obs_get_source_by_name(name.c_str());
if (source) {
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QMessageBox::information(this,
QTStr("NameExists.Title"),
QTStr("NameExists.Text"));
obs_source_release(source);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
on_actionAddScene_triggered();
return;
}
obs_scene_t scene = obs_scene_create(name.c_str());
source = obs_scene_getsource(scene);
obs_add_source(source);
obs_scene_release(scene);
obs_set_output_source(0, source);
}
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionRemoveScene_triggered()
{
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QListWidgetItem *item = ui->scenes->currentItem();
if (!item)
2013-12-29 19:01:19 -08:00
return;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QVariant userData = item->data(Qt::UserRole);
obs_scene_t scene = userData.value<OBSScene>();
2013-12-29 19:01:19 -08:00
obs_source_t source = obs_scene_getsource(scene);
obs_source_remove(source);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSceneProperties_triggered()
{
/* TODO */
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSceneUp_triggered()
{
/* TODO */
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSceneDown_triggered()
{
/* TODO */
}
void OBSBasic::on_sources_currentItemChanged(QListWidgetItem *current,
QListWidgetItem *prev)
{
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
auto select_one = [] (obs_scene_t scene, obs_sceneitem_t item,
void *param)
{
obs_sceneitem_t selectedItem =
*reinterpret_cast<OBSSceneItem*>(param);
obs_sceneitem_select(item, (selectedItem == item));
UNUSED_PARAMETER(scene);
return true;
};
if (!current)
return;
OBSSceneItem item = current->data(Qt::UserRole).value<OBSSceneItem>();
obs_scene_enum_items(GetCurrentScene(), select_one, &item);
UNUSED_PARAMETER(prev);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_sources_customContextMenuRequested(const QPoint &pos)
{
/* TODO */
UNUSED_PARAMETER(pos);
}
void OBSBasic::AddSource(const char *id)
{
OBSBasicSourceSelect sourceSelect(this, id);
sourceSelect.exec();
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::AddSourcePopupMenu(const QPoint &pos)
{
const char *type;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
bool foundValues = false;
size_t idx = 0;
if (!GetCurrentScene()) {
// Tell the user he needs a scene first (help beginners).
QMessageBox::information(this,
QTStr("Basic.Main.AddSourceHelp.Title"),
QTStr("Basic.Main.AddSourceHelp.Text"));
return;
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QMenu popup;
while (obs_enum_input_types(idx++, &type)) {
Revamp API and start using doxygen The API used to be designed in such a way to where it would expect exports for each individual source/output/encoder/etc. You would export functions for each and it would automatically load those functions based on a specific naming scheme from the module. The idea behind this was that I wanted to limit the usage of structures in the API so only functions could be used. It was an interesting idea in theory, but this idea turned out to be flawed in a number of ways: 1.) Requiring exports to create sources/outputs/encoders/etc meant that you could not create them by any other means, which meant that things like faruton's .net plugin would become difficult. 2.) Export function declarations could not be checked, therefore if you created a function with the wrong parameters and parameter types, the compiler wouldn't know how to check for that. 3.) Required overly complex load functions in libobs just to handle it. It makes much more sense to just have a load function that you call manually. Complexity is the bane of all good programs. 4.) It required that you have functions of specific names, which looked and felt somewhat unsightly. So, to fix these issues, I replaced it with a more commonly used API scheme, seen commonly in places like kernels and typical C libraries with abstraction. You simply create a structure that contains the callback definitions, and you pass it to a function to register that definition (such as obs_register_source), which you call in the obs_module_load of the module. It will also automatically check the structure size and ensure that it only loads the required values if the structure happened to add new values in an API change. The "main" source file for each module must include obs-module.h, and must use OBS_DECLARE_MODULE() within that source file. Also, started writing some doxygen documentation in to the main library headers. Will add more detailed documentation as I go.
2014-02-12 07:04:50 -08:00
const char *name = obs_source_getdisplayname(
OBS_SOURCE_TYPE_INPUT, type);
if (strcmp(type, "scene") == 0)
continue;
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
QAction *popupItem = new QAction(QT_UTF8(name), this);
popupItem->setData(QT_UTF8(type));
popup.addAction(popupItem);
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
foundValues = true;
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
if (foundValues) {
QAction *ret = popup.exec(pos);
if (ret)
AddSource(ret->data().toString().toUtf8());
}
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionAddSource_triggered()
{
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
AddSourcePopupMenu(QCursor::pos());
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionRemoveSource_triggered()
{
OBSSceneItem item = GetCurrentSceneItem();
if (item)
obs_sceneitem_remove(item);
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSourceProperties_triggered()
{
Add source properties window (very preliminary) - Add a properties window for sources so that you can now actually edit the settings for sources. Also, display the source by itself in the window (Note: not working on mac, and possibly not working on linux). When changing the settings for a source, it will call obs_source_update on that source when you have modified any values automatically. - Add a properties 'widget', eventually I want to turn this in to a regular nice properties view like you'd see in the designer, but right now it just uses a form layout in a QScrollArea with regular controls to display the properties. It's clunky but works for the time being. - Make it so that swap chains and the main graphics subsystem will automatically use at least one backbuffer if none was specified - Fix bug where displays weren't added to the main display array - Make it so that you can get the properties of a source via the actual pointer of a source/encoder/output in addition to being able to look up properties via identifier. - When registering source types, check for required functions (wasn't doing it before). getheight/getwidth should not be optional if it's a video source as well. - Add an RAII OBSObj wrapper to obs.hpp for non-reference-counted libobs pointers - Add an RAII OBSSignal wrapper to obs.hpp for libobs signals to automatically disconnect them on destruction - Move the "scale and center" calculation in window-basic-main.cpp to its own function and in its own source file - Add an 'update' callback to WASAPI audio sources
2014-03-23 01:07:54 -07:00
OBSSceneItem item = GetCurrentSceneItem();
OBSSource source = obs_sceneitem_getsource(item);
if (source) {
delete properties;
properties = new OBSBasicProperties(this, source);
properties->Init();
}
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSourceUp_triggered()
{
2014-05-15 17:40:53 -07:00
OBSSceneItem item = GetCurrentSceneItem();
obs_sceneitem_setorder(item, ORDER_MOVE_UP);
}
2013-12-10 10:22:33 -08:00
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_actionSourceDown_triggered()
{
2014-05-15 17:40:53 -07:00
OBSSceneItem item = GetCurrentSceneItem();
obs_sceneitem_setorder(item, ORDER_MOVE_DOWN);
}
static char *ReadLogFile(const char *log)
{
BPtr<char> logDir(os_get_config_path("obs-studio/logs"));
string path = (char*)logDir;
path += "/";
path += log;
char *file = os_quick_read_utf8_file(path.c_str());
if (!file)
blog(LOG_WARNING, "Failed to read log file %s", path.c_str());
return file;
}
void OBSBasic::UploadLog(const char *file)
{
dstr fileString = {};
stringstream ss;
string jsonData;
dstr_move_array(&fileString, ReadLogFile(file));
if (!fileString.array)
return;
if (!*fileString.array) {
dstr_free(&fileString);
return;
}
ui->menuLogFiles->setEnabled(false);
dstr_replace(&fileString, "\\", "\\\\");
dstr_replace(&fileString, "\"", "\\\"");
dstr_replace(&fileString, "\n", "\\n");
dstr_replace(&fileString, "\r", "\\r");
dstr_replace(&fileString, "\t", "\\t");
dstr_replace(&fileString, "\t", "\\t");
dstr_replace(&fileString, "/", "\\/");
ss << "{ \"public\": false, \"description\": \"OBS " <<
App()->GetVersionString() << " log file uploaded at " <<
CurrentDateTimeString() << "\", \"files\": { \"" <<
file << "\": { \"content\": \"" <<
fileString.array << "\" } } }";
jsonData = std::move(ss.str());
2014-05-22 03:59:51 -07:00
logUploadPostData.setData(jsonData.c_str(), (int)jsonData.size());
QUrl url("https://api.github.com/gists");
logUploadReply = networkManager.post(QNetworkRequest(url),
&logUploadPostData);
connect(logUploadReply, SIGNAL(finished()),
this, SLOT(logUploadFinished()));
connect(logUploadReply, SIGNAL(readyRead()),
this, SLOT(logUploadRead()));
dstr_free(&fileString);
}
void OBSBasic::on_actionUploadCurrentLog_triggered()
{
UploadLog(App()->GetCurrentLog());
}
void OBSBasic::on_actionUploadLastLog_triggered()
{
UploadLog(App()->GetLastLog());
}
void OBSBasic::logUploadRead()
{
logUploadReturnData.push_back(logUploadReply->readAll());
}
void OBSBasic::logUploadFinished()
{
ui->menuLogFiles->setEnabled(true);
if (logUploadReply->error()) {
QMessageBox::information(this,
QTStr("LogReturnDialog.ErrorUploadingLog"),
logUploadReply->errorString());
return;
}
const char *jsonReply = logUploadReturnData.constData();
if (!jsonReply || !*jsonReply)
return;
obs_data_t returnData = obs_data_create_from_json(jsonReply);
QString logURL = obs_data_getstring(returnData, "html_url");
obs_data_release(returnData);
OBSLogReply logDialog(this, logURL);
logDialog.exec();
}
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
void OBSBasic::StreamingStart()
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
{
ui->streamButton->setText("Stop Streaming");
ui->streamButton->setEnabled(true);
}
void OBSBasic::StreamingStop(int code)
{
const char *errorMessage;
switch (code) {
case OBS_OUTPUT_BAD_PATH:
errorMessage = Str("Output.ConnectFail.BadPath");
break;
case OBS_OUTPUT_CONNECT_FAILED:
errorMessage = Str("Output.ConnectFail.ConnectFailed");
break;
case OBS_OUTPUT_INVALID_STREAM:
errorMessage = Str("Output.ConnectFail.InvalidStream");
break;
default:
case OBS_OUTPUT_ERROR:
errorMessage = Str("Output.ConnectFail.Error");
break;
case OBS_OUTPUT_DISCONNECTED:
/* doesn't happen if output is set to reconnect. note that
* reconnects are handled in the output, not in the UI */
errorMessage = Str("Output.ConnectFail.Disconnected");
}
activeRefs--;
ui->streamButton->setText(QTStr("Basic.Main.StartStreaming"));
ui->streamButton->setEnabled(true);
if (code != OBS_OUTPUT_SUCCESS)
QMessageBox::information(this,
QTStr("Output.ConnectFail.Title"),
QT_UTF8(errorMessage));
2014-04-14 02:22:09 -07:00
}
void OBSBasic::RecordingStop()
{
activeRefs--;
ui->recordButton->setText(QTStr("Basic.Main.StartRecording"));
}
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
void OBSBasic::SetupEncoders()
{
if (activeRefs == 0) {
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_data_t x264Settings = obs_data_create();
obs_data_t aacSettings = obs_data_create();
int videoBitrate = config_get_uint(basicConfig, "SimpleOutput",
"VBitrate");
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
int audioBitrate = config_get_uint(basicConfig, "SimpleOutput",
"ABitrate");
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_data_setint(x264Settings, "bitrate", videoBitrate);
obs_data_setbool(x264Settings, "cbr", true);
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_data_setint(aacSettings, "bitrate", audioBitrate);
Implement RTMP module (still needs drop code) - Implement the RTMP output module. This time around, we just use a simple FLV muxer, then just write to the stream with RTMP_Write. Easy and effective. - Fix the FLV muxer, the muxer now outputs proper FLV packets. - Output API: * When using encoders, automatically interleave encoded packets before sending it to the output. * Pair encoders and have them automatically wait for the other to start to ensure sync. * Change 'obs_output_signal_start_fail' to 'obs_output_signal_stop' because it was a bit confusing, and doing this makes a lot more sense for outputs that need to stop suddenly (disconnections/etc). - Encoder API: * Remove some unnecessary encoder functions from the actual API and make them internal. Most of the encoder functions are handled automatically by outputs anyway, so there's no real need to expose them and end up inadvertently confusing plugin writers. * Have audio encoders wait for the video encoder to get a frame, then start at the exact data point that the first video frame starts to ensure the most accrate sync of video/audio possible. * Add a required 'frame_size' callback for audio encoders that returns the expected number of frames desired to encode with. This way, the libobs encoder API can handle the circular buffering internally automatically for the encoder modules, so encoder writers don't have to do it themselves. - Fix a few bugs in the serializer interface. It was passing the wrong variable for the data in a few cases. - If a source has video, make obs_source_update defer the actual update callback until the tick function is called to prevent threading issues.
2014-04-07 22:00:10 -07:00
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_encoder_update(x264, x264Settings);
obs_encoder_update(aac, aacSettings);
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_data_release(x264Settings);
obs_data_release(aacSettings);
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_encoder_set_video(x264, obs_video());
obs_encoder_set_audio(aac, obs_audio());
}
}
void OBSBasic::on_streamButton_clicked()
{
if (obs_output_active(streamOutput)) {
obs_output_stop(streamOutput);
} else {
SaveService();
SetupEncoders();
obs-studio UI: Implement stream settings UI - Updated the services API so that it links up with an output and the output gets data from that service rather than via settings. This allows the service context to have control over how an output is used, and makes it so that the URL/key/etc isn't necessarily some static setting. Also, if the service is attached to an output, it will stick around until the output is destroyed. - The settings interface has been updated so that it can allow the usage of service plugins. What this means is that now you can create a service plugin that can control aspects of the stream, and it allows each service to create their own user interface if they create a service plugin module. - Testing out saving of current service information. Saves/loads from JSON in to obs_data_t, seems to be working quite nicely, and the service object information is saved/preserved on exit, and loaded again on startup. - I agonized over the settings user interface for days, and eventually I just decided that the only way that users weren't going to be fumbling over options was to split up the settings in to simple/basic output, pre-configured, and then advanced for advanced use (such as multiple outputs or services, which I'll implement later). This was particularly painful to really design right, I wanted more features and wanted to include everything in one interface but ultimately just realized from experience that users are just not technically knowledgable about it and will end up fumbling with the settings rather than getting things done. Basically, what this means is that casual users only have to enter in about 3 things to configure their stream: Stream key, audio bitrate, and video bitrate. I am really happy with this interface for those types of users, but it definitely won't be sufficient for advanced usage or for custom outputs, so that stuff will have to be separated. - Improved the JSON usage for the 'common streaming services' context, I realized that JSON arrays are there to ensure sorting, while forgetting that general items are optimized for hashing. So basically I'm just using arrays now to sort items in it.
2014-04-24 01:49:07 -07:00
obs_output_set_video_encoder(streamOutput, x264);
obs_output_set_audio_encoder(streamOutput, aac);
obs_output_set_service(streamOutput, service);
if (obs_output_start(streamOutput)) {
activeRefs++;
ui->streamButton->setEnabled(false);
ui->streamButton->setText(
QTStr("Basic.Main.Connecting"));
}
}
}
void OBSBasic::on_recordButton_clicked()
{
if (obs_output_active(fileOutput)) {
obs_output_stop(fileOutput);
} else {
const char *path = config_get_string(basicConfig,
"SimpleOutput", "FilePath");
os_dir_t dir = path ? os_opendir(path) : nullptr;
if (!dir) {
QMessageBox::information(this,
QTStr("Output.BadPath.Title"),
QTStr("Output.BadPath.Text"));
return;
}
os_closedir(dir);
string strPath;
strPath += path;
char lastChar = strPath.back();
if (lastChar != '/' && lastChar != '\\')
strPath += "/";
strPath += GenerateTimeDateFilename("flv");
SetupEncoders();
obs_output_set_video_encoder(fileOutput, x264);
obs_output_set_audio_encoder(fileOutput, aac);
obs_data_t settings = obs_data_create();
obs_data_setstring(settings, "path", strPath.c_str());
obs_output_update(fileOutput, settings);
obs_data_release(settings);
if (obs_output_start(fileOutput)) {
activeRefs++;
ui->recordButton->setText(
QTStr("Basic.Main.StopRecording"));
}
}
}
Change the UI to Qt (work in progress) -------------------------------------------------- Notes and details -------------------------------------------------- Why was this done? Because wxWidgets was just lacking in many areas. I know wxWidgets is designed to be used with native controls, and that's great, but wxWidgets just is not a feature-complete toolkit for multiplatform applications. It lacks in dialog editors, its code is archaic and outdated, and I just feel frustrated every time I try to do things with it. Qt on the other hand.. I had to actually try Qt to realize how much better it was as a toolkit. They've got everything from dialog editors, to an IDE, a debugger, build tools, just everything, and it's all top-notch and highly maintained. The focus of the toolkit is application development, and they spend their time trying to help people do exactly that: make programs. Great support, great tools, and because of that, great toolkit. I just didn't want to alienate any developers by being stubborn about native widgets. There *are* some things that are rather lackluster about it and design choices I disagree with though. For example, I realize that to have an easy to use toolkit you have to have some level of code generation. However, in my personal and humble opinion, moc just feels like a terrible way to approach the problem. Even now I feel like there are a variety of ways you could handle code generation and automatic management of things like that. I don't like the idea of circumventing the language itself like that. It feels like one giant massive hack. -------------------------------------------------- Things that aren't working properly: -------------------------------------------------- - Settings dialog is not implemented. The dialog is complete but the code to handle the dialog hasn't been constructed yet. - There is a problem with using Qt widgets as a device target on windows, with at least OpenGL: if I have the preview widget automatically resize itself, it seems to cause some sort of video card failure that I don't understand. - Because of the above, resizing the preview widget has been disabled until I can figure out what's going on, so it's currently only a 32x32 area. - Direct3D doesn't seem to render correctly either, seems that the viewport is messed up or something. I'm sort of confused about what's going on with it. - The new main window seems to be triggering more race conditions than the wxWidgets main window dialog did. I'm not entirely sure what's going on here, but this may just be existing race conditions within libobs itself that I just never spotted before (even though I tend to be very thorough with race conditions any time I use variables cross-thread)
2014-01-23 10:53:55 -08:00
void OBSBasic::on_settingsButton_clicked()
2013-12-10 10:22:33 -08:00
{
OBSBasicSettings settings(this);
settings.exec();
2013-12-10 10:22:33 -08:00
}
void OBSBasic::GetFPSCommon(uint32_t &num, uint32_t &den) const
{
const char *val = config_get_string(basicConfig, "Video", "FPSCommon");
if (strcmp(val, "10") == 0) {
num = 10;
den = 1;
} else if (strcmp(val, "20") == 0) {
num = 20;
den = 1;
} else if (strcmp(val, "25") == 0) {
num = 25;
den = 1;
} else if (strcmp(val, "29.97") == 0) {
num = 30000;
den = 1001;
} else if (strcmp(val, "48") == 0) {
num = 48;
den = 1;
} else if (strcmp(val, "59.94") == 0) {
num = 60000;
den = 1001;
} else if (strcmp(val, "60") == 0) {
num = 60;
den = 1;
} else {
num = 30;
den = 1;
}
}
void OBSBasic::GetFPSInteger(uint32_t &num, uint32_t &den) const
{
num = (uint32_t)config_get_uint(basicConfig, "Video", "FPSInt");
den = 1;
}
void OBSBasic::GetFPSFraction(uint32_t &num, uint32_t &den) const
{
num = (uint32_t)config_get_uint(basicConfig, "Video", "FPSNum");
den = (uint32_t)config_get_uint(basicConfig, "Video", "FPSDen");
}
void OBSBasic::GetFPSNanoseconds(uint32_t &num, uint32_t &den) const
{
num = 1000000000;
den = (uint32_t)config_get_uint(basicConfig, "Video", "FPSNS");
}
void OBSBasic::GetConfigFPS(uint32_t &num, uint32_t &den) const
{
uint32_t type = config_get_uint(basicConfig, "Video", "FPSType");
if (type == 1) //"Integer"
GetFPSInteger(num, den);
else if (type == 2) //"Fraction"
GetFPSFraction(num, den);
else if (false) //"Nanoseconds", currently not implemented
GetFPSNanoseconds(num, den);
else
GetFPSCommon(num, den);
}
config_t OBSBasic::Config() const
{
return basicConfig;
}
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
void OBSBasic::on_actionEditTransform_triggered()
{
delete transformWindow;
transformWindow = new OBSBasicTransform(this);
transformWindow->show();
}
void OBSBasic::on_actionResetTransform_triggered()
{
auto func = [] (obs_scene_t scene, obs_sceneitem_t item, void *param)
{
if (!obs_sceneitem_selected(item))
return true;
obs_sceneitem_info info;
vec2_set(&info.pos, 0.0f, 0.0f);
vec2_set(&info.scale, 1.0f, 1.0f);
info.rot = 0.0f;
info.alignment = OBS_ALIGN_TOP | OBS_ALIGN_LEFT;
info.bounds_type = OBS_BOUNDS_NONE;
info.bounds_alignment = OBS_ALIGN_CENTER;
vec2_set(&info.bounds, 0.0f, 0.0f);
obs_sceneitem_set_info(item, &info);
UNUSED_PARAMETER(scene);
UNUSED_PARAMETER(param);
return true;
};
obs_scene_enum_items(GetCurrentScene(), func, nullptr);
}
static void GetItemBox(obs_sceneitem_t item, vec3 &tl, vec3 &br)
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
{
matrix4 boxTransform;
obs_sceneitem_get_box_transform(item, &boxTransform);
vec3_set(&tl, M_INFINITE, M_INFINITE, 0.0f);
vec3_set(&br, -M_INFINITE, -M_INFINITE, 0.0f);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
auto GetMinPos = [&] (float x, float y)
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
{
vec3 pos;
vec3_set(&pos, x, y, 0.0f);
vec3_transform(&pos, &pos, &boxTransform);
vec3_min(&tl, &tl, &pos);
vec3_max(&br, &br, &pos);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
};
GetMinPos(0.0f, 0.0f);
GetMinPos(1.0f, 0.0f);
GetMinPos(0.0f, 1.0f);
GetMinPos(1.0f, 1.0f);
}
static vec3 GetItemTL(obs_sceneitem_t item)
{
vec3 tl, br;
GetItemBox(item, tl, br);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
return tl;
}
static void SetItemTL(obs_sceneitem_t item, const vec3 &tl)
{
vec3 newTL;
vec2 pos;
obs_sceneitem_getpos(item, &pos);
newTL = GetItemTL(item);
pos.x += tl.x - newTL.x;
pos.y += tl.y - newTL.y;
obs_sceneitem_setpos(item, &pos);
}
static bool RotateSelectedSources(obs_scene_t scene, obs_sceneitem_t item,
void *param)
{
if (!obs_sceneitem_selected(item))
return true;
float rot = *reinterpret_cast<float*>(param);
vec3 tl = GetItemTL(item);
rot += obs_sceneitem_getrot(item);
if (rot >= 360.0f) rot -= 360.0f;
else if (rot <= -360.0f) rot += 360.0f;
obs_sceneitem_setrot(item, rot);
SetItemTL(item, tl);
UNUSED_PARAMETER(scene);
UNUSED_PARAMETER(param);
return true;
};
void OBSBasic::on_actionRotate90CW_triggered()
{
float f90CW = 90.0f;
obs_scene_enum_items(GetCurrentScene(), RotateSelectedSources, &f90CW);
}
void OBSBasic::on_actionRotate90CCW_triggered()
{
float f90CCW = -90.0f;
obs_scene_enum_items(GetCurrentScene(), RotateSelectedSources, &f90CCW);
}
void OBSBasic::on_actionRotate180_triggered()
{
float f180 = 180.0f;
obs_scene_enum_items(GetCurrentScene(), RotateSelectedSources, &f180);
}
static bool MultiplySelectedItemScale(obs_scene_t scene, obs_sceneitem_t item,
void *param)
{
vec2 &mul = *reinterpret_cast<vec2*>(param);
if (!obs_sceneitem_selected(item))
return true;
vec3 tl = GetItemTL(item);
vec2 scale;
obs_sceneitem_getscale(item, &scale);
vec2_mul(&scale, &scale, &mul);
obs_sceneitem_setscale(item, &scale);
SetItemTL(item, tl);
UNUSED_PARAMETER(scene);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
return true;
}
void OBSBasic::on_actionFlipHorizontal_triggered()
{
vec2 scale;
vec2_set(&scale, -1.0f, 1.0f);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
obs_scene_enum_items(GetCurrentScene(), MultiplySelectedItemScale,
&scale);
}
void OBSBasic::on_actionFlipVertical_triggered()
{
vec2 scale;
vec2_set(&scale, 1.0f, -1.0f);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
obs_scene_enum_items(GetCurrentScene(), MultiplySelectedItemScale,
&scale);
}
static bool CenterAlignSelectedItems(obs_scene_t scene, obs_sceneitem_t item,
void *param)
{
obs_bounds_type boundsType = *reinterpret_cast<obs_bounds_type*>(param);
if (!obs_sceneitem_selected(item))
return true;
obs_video_info ovi;
obs_get_video_info(&ovi);
obs_sceneitem_info itemInfo;
vec2_set(&itemInfo.pos, 0.0f, 0.0f);
vec2_set(&itemInfo.scale, 1.0f, 1.0f);
itemInfo.alignment = OBS_ALIGN_LEFT | OBS_ALIGN_TOP;
itemInfo.rot = 0.0f;
vec2_set(&itemInfo.bounds,
float(ovi.base_width), float(ovi.base_height));
itemInfo.bounds_type = boundsType;
itemInfo.bounds_alignment = OBS_ALIGN_CENTER;
obs_sceneitem_set_info(item, &itemInfo);
UNUSED_PARAMETER(scene);
return true;
}
void OBSBasic::on_actionFitToScreen_triggered()
{
obs_bounds_type boundsType = OBS_BOUNDS_SCALE_INNER;
obs_scene_enum_items(GetCurrentScene(), CenterAlignSelectedItems,
&boundsType);
}
void OBSBasic::on_actionStretchToScreen_triggered()
{
obs_bounds_type boundsType = OBS_BOUNDS_STRETCH;
obs_scene_enum_items(GetCurrentScene(), CenterAlignSelectedItems,
&boundsType);
}
void OBSBasic::on_actionCenterToScreen_triggered()
{
auto func = [] (obs_scene_t scene, obs_sceneitem_t item, void *param)
{
vec3 tl, br, itemCenter, screenCenter, offset;
obs_video_info ovi;
if (!obs_sceneitem_selected(item))
return true;
obs_get_video_info(&ovi);
vec3_set(&screenCenter, float(ovi.base_width),
float(ovi.base_height), 0.0f);
vec3_mulf(&screenCenter, &screenCenter, 0.5f);
GetItemBox(item, tl, br);
vec3_sub(&itemCenter, &br, &tl);
vec3_mulf(&itemCenter, &itemCenter, 0.5f);
vec3_add(&itemCenter, &itemCenter, &tl);
vec3_sub(&offset, &screenCenter, &itemCenter);
vec3_add(&tl, &tl, &offset);
SetItemTL(item, tl);
UNUSED_PARAMETER(scene);
2014-06-25 01:40:39 -07:00
UNUSED_PARAMETER(param);
return true;
};
obs_scene_enum_items(GetCurrentScene(), func, nullptr);
UI: Add scene editing So, scene editing was interesting (and by interesting I mean excruciating). I almost implemented 'manipulator' visuals (ala 3dsmax for example), and used 3 modes for controlling position/rotation/size, but in a 2D editing, it felt clunky, so I defaulted back to simply click-and-drag for movement, and then took a similar though slightly different looking approach for handling scaling and reszing. I also added a number of menu item helpers related to positioning, scaling, rotating, flipping, and resetting the transform back to default. There is also a new 'transform' dialog (accessible via menu) which will allow you to manually edit every single transform variable of a scene item directly if desired. If a scene item does not have bounds active, pulling on the sides of a source will cause it to resize it via base scale rather than by the bounding box system (if the source resizes that scale will apply). If bounds are active, it will modify the bounding box only instead. How a source scales when a bounding box is active depends on the type of bounds being used. You can set it to scale to the inner bounds, the outer bounds, scale to bounds width only, scale to bounds height only, and a setting to stretch to bounds (which forces a source to always draw at the bounding box size rather than be affected by its internal size). You can also set it to be used as a 'maximum' size, so that the source doesn't necessarily get scaled unless it extends beyond the bounds. Like in OBS1, objects will snap to the edges unless the control key is pressed. However, this will now happen even if the object is rotated or oriented in any strange way. Snapping will also occur when stretching or changing the bounding box size.
2014-06-15 00:54:48 -07:00
}