Commit Graph

3989 Commits (68b0fce62f684a4b491ebb36dea06bd4d9ff5219)

Author SHA1 Message Date
Henry Nelson 68b0fce62f Document explicitly ignoring expression values 2018-12-11 01:42:54 -05:00
Andrew Kelley 1b0f4d5976
implement compile error note for function parameter type mismatch 2018-12-05 15:32:25 -05:00
Jimmi Holst Christensen 518ff33e64 Allow packages in TestStep 2018-12-05 14:10:09 +01:00
Andrew Kelley 9ac838c8e3
LLD patch: allow non-allocated sections to go into allocated sections
Patch submitted upstream: https://reviews.llvm.org/D55276
2018-12-04 11:42:22 -05:00
Andrew Kelley 3a1612a0f5
std.debug: fix some issues with freestanding debug info 2018-12-02 23:46:45 -05:00
Andrew Kelley 5c3b8cb365
expose std.debug.DwarfInfo and delete unused function 2018-12-02 20:08:06 -05:00
Andrew Kelley a436d2ab8c
std.debug.printSourceAtAddressDwarf
which works in freestanding mode
2018-12-02 19:34:11 -05:00
Andrew Kelley 4292ed9571
std.debug.StackIterator 2018-12-02 18:54:04 -05:00
Andrew Kelley a40d160a5c
introduce std.io.SeekableStream
Relevant #764

dwarf debug info is modified to use this instead of std.os.File
directly to make it easier for bare metal projects to take advantage
of debug info parsing
2018-12-02 18:36:18 -05:00
Andrew Kelley 6f5e7ee09c
make std.unicode.Utf8Iterator public 2018-12-01 02:14:17 -05:00
Andrew Kelley a9eab69b58
zig build: addStaticExecutable 2018-12-01 01:21:59 -05:00
Jimmi Holst Christensen 823969a5a4
Implemented new more flexible readLineFrom (#1801) 2018-11-29 22:38:39 +01:00
Andrew Kelley b29769532b
ci: workaround azure quirk for windows too 2018-11-29 14:54:23 -05:00
Andrew Kelley bbdc12891b
ci: workaround azure quirk with set -x 2018-11-29 11:05:56 -05:00
dbandstra 53766e7a3a make parseUnsigned handle types <8 bits wide 2018-11-29 09:53:43 -05:00
daurnimator 4ecb3ceafb Add std.LinkedList.concat 2018-11-29 09:44:24 -05:00
Andrew Kelley 7005ec5efe
Revert "Add math min/max for Float and Value"
This reverts commit 078a0a6999.

On closer inspection, I'm not sure these values for float min/max
make sense. Why is it max instead of true_max? Why isn't it positive
and negative infinity?

Let's discuss further before committing to these changes.
2018-11-29 09:33:47 -05:00
Jimmi Holst Christensen f74320d56d Implemented getOrPutValue which wraps getOrPut 2018-11-29 09:10:15 +01:00
Wink Saville 078a0a6999 Add math min/max for Float and Value 2018-11-28 23:13:12 -05:00
Andrew Kelley c54fe0d3ae
work around to support debian's fork of llvm 7.0.1
it has a patch that adds an OS type, breaking the public API
this commit avoids depending on the last os type enum item,
but retains the safety assertion checks.

closes #1788
2018-11-28 22:58:52 -05:00
Josh Wolfe 11e8afb37c fix child_process piped streams not getting closed 2018-11-28 18:33:55 -05:00
Andrew Kelley 57f44eb2bd
Merge branch 'freebsd2'
Tier 2 support for FreeBSD
2018-11-27 21:06:35 -05:00
Andrew Kelley dd2450b1b2
tier 2 support for freebsd 2018-11-27 20:56:43 -05:00
Andrew Kelley e3bf40742d
Merge branch 'master' into freebsd2 2018-11-27 20:36:44 -05:00
Sahnvour 1fb15be05f stack traces: fix for windows 2018-11-27 20:34:55 -05:00
Andrew Kelley 3c4965a616
readme: update support table 2018-11-27 11:27:45 -05:00
daurnimator 2baf0e2ffc add std.math.IntFittingRange 2018-11-27 10:19:06 -05:00
Andrew Kelley 6bbe4cf9f7
Merge branch 'winksaville-Fix-pushToParent-to-work-for-arrays-of-Objects' 2018-11-27 00:53:09 -05:00
Andrew Kelley a5b79017a3
fix regression from d5648d26
thanks Wink Saville for the test case.
2018-11-27 00:52:32 -05:00
Andrew Kelley 7818550edc
Merge branch 'Fix-pushToParent-to-work-for-arrays-of-Objects' of https://github.com/winksaville/zig into winksaville-Fix-pushToParent-to-work-for-arrays-of-Objects 2018-11-27 00:45:46 -05:00
Andrew Kelley 0860be03b1
Merge branch 'kristate-zig-backport-os_file_read-EISDIR' 2018-11-26 20:07:19 -05:00
Andrew Kelley 67a39a4c99
stage1: better file path handling
* better message printed when cache hash fails
 * better handling of '/' as root source file
 * os_path_split parses '/' and '/a' correctly

closes #1693
closes #1746
2018-11-26 20:04:35 -05:00
Andrew Kelley d2ed4d3dd3
Merge branch 'zig-backport-os_file_read-EISDIR' of https://github.com/kristate/zig into kristate-zig-backport-os_file_read-EISDIR 2018-11-26 19:59:58 -05:00
daurnimator f6cd02be65 add std.meta.stringToEnum 2018-11-25 11:48:11 -05:00
Andrew Kelley a07490a4bc
Merge pull request #1783 from ziglang/rand-range
Use better rand range implementations
2018-11-25 11:44:08 -05:00
Josh Wolfe 9ae5200bd2 factor out and expose biased range limiting function 2018-11-24 22:25:21 -05:00
Josh Wolfe eed7b48fe3 test lots of types 2018-11-24 22:25:21 -05:00
Josh Wolfe 49b49618d2 add biased random range api 2018-11-24 22:25:21 -05:00
Josh Wolfe 1924ffa67d better debiased random range implementation 2018-11-24 22:25:21 -05:00
Andrew Kelley 4d747d452f
update zen os std lib for latest zig changes 2018-11-24 16:51:52 -05:00
Andrew Kelley 3d2752cc36
refactor type_requires_comptime to have possible error
fixes a compiler crash when building
https://github.com/AndreaOrru/zen
2018-11-24 16:15:58 -05:00
Andrew Kelley 56a8f2b018
fix @intCast not catching negative numbers to unsigned 2018-11-24 14:36:16 -05:00
Andrew Kelley b953e322a8
std.os.path.realC: make overflow more clearly impossible 2018-11-21 13:41:10 -05:00
Andrew Kelley bb3ac177a8
fix incorrect buf len 2018-11-20 12:37:42 -05:00
kristopher tate a69695a28c std/mem: writeIntLE: buf.* to buf; 2018-11-20 11:11:07 -05:00
Wink Saville a05716bd20 Allow json tests to be easily filtered 2018-11-19 18:08:38 -05:00
Andrew Kelley 4dafdc00d5
zig fmt 2018-11-19 17:28:18 -05:00
Andrew Kelley 9493738e54
Merge branch 'freebsd-up' of https://github.com/myfreeweb/zig into freebsd2 2018-11-19 17:24:41 -05:00
Sahnvour 703c6684d1 Crash fixes and small improvements to inline asm. (#1756)
* codegen: LLVMConstInlineAsm is deprecated.

* codegen: replace commas in asm constraint strings by pipes as required by LLVM.

* ir: enforce usage of '=' constraint modifier for inline assembly outputs.

Others are not currently supported and this was just asserted alter in `ir_render_asm`.

* asm: forbid comptime_int/floats as inputs in favor of explicitely sized constants.

Fixes a crash due to comptime_int/floats having no type_ref.

* asm: handle inputs with integers of <8 or non power of 2 bitsize.

We widen them to the next highest power of two.
2018-11-19 16:22:21 -05:00
Andrew Kelley 89e82281be
fix incorrect --help text 2018-11-19 13:26:35 -05:00