add 'text.parse_float()' + cleanup in text.py

gallery_dl/extractor/sankaku.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Copyright 2014-2018 Mike Fährmann
+# Copyright 2014-2019 Mike Fährmann
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 2 as
@@ -88,7 +88,7 @@ class SankakuExtractor(SharedConfigExtractor):
             "id": text.parse_int(post_id),
             "md5": file_url.rpartition("/")[2].partition(".")[0],
             "tags": text.unescape(tags),
-            "vote_average": float(vavg or 0),
+            "vote_average": text.parse_float(vavg),
             "vote_count": text.parse_int(vcnt),
             "created_at": created,
             "rating": (rating or "?")[0].lower(),

gallery_dl/text.py

@@ -1,12 +1,12 @@
 # -*- coding: utf-8 -*-
 
-# Copyright 2015-2018 Mike Fährmann
+# Copyright 2015-2019 Mike Fährmann
 #
 # This program is free software; you can redistribute it and/or modify
 # it under the terms of the GNU General Public License version 2 as
 # published by the Free Software Foundation.
 
-"""Collection of functions that work in strings/text"""
+"""Collection of functions that work on strings/text"""
 
 import re
 import html
@@ -47,7 +47,7 @@ def split_html(txt, sep=None):
     """Split input string by html-tags"""
     try:
         return [
-            x for x in re.split("<[^>]+>", txt)
+            x.strip() for x in re.split("<[^>]+>", txt)
             if x and not x.isspace()
         ]
     except TypeError:
@@ -165,6 +165,16 @@ def parse_int(value, default=0):
         return default
 
 
+def parse_float(value, default=0.0):
+    """Convert 'value' to float"""
+    if not value:
+        return default
+    try:
+        return float(value)
+    except (ValueError, TypeError):
+        return default
+
+
 def parse_query(qs):
     """Parse a query string into key-value pairs"""
     result = {}
@@ -182,12 +192,11 @@ if os.name == "nt":
 else:
     clean_path = clean_path_posix
 
-urljoin = urllib.parse.urljoin
-unquote = urllib.parse.unquote
-escape = html.escape
 
-try:
+urljoin = urllib.parse.urljoin
+
+quote = urllib.parse.quote
+unquote = urllib.parse.unquote
+
+escape = html.escape
 unescape = html.unescape
-except AttributeError:
-    import html.parser
-    unescape = html.parser.HTMLParser().unescape

test/test_text.py

@@ -73,6 +73,7 @@ class TestText(unittest.TestCase):
         self.assertEqual(f("Hello World."), ["Hello World."])
         self.assertEqual(f(" Hello  World.  "), ["Hello  World."])
         self.assertEqual(f("Hello<br/>World."), result)
+        self.assertEqual(f(" Hello <br/> World.  "), result)
         self.assertEqual(
             f("<div><b class='a'>Hello</b><i>World.</i></div>"), result)
 
@@ -260,6 +261,27 @@ class TestText(unittest.TestCase):
             self.assertEqual(f(value, default), default)
         self.assertEqual(f("zzz", default), default)
 
+    def test_parse_float(self, f=text.parse_float):
+        self.assertEqual(f(0), 0.0)
+        self.assertEqual(f("0"), 0.0)
+        self.assertEqual(f(123), 123.0)
+        self.assertEqual(f("123"), 123.0)
+        self.assertEqual(f(123.456), 123.456)
+        self.assertEqual(f("123.456"), 123.456)
+
+        # invalid arguments
+        for value in INVALID_ALT:
+            self.assertEqual(f(value), 0.0)
+        self.assertEqual(f("zzz"), 0.0)
+        self.assertEqual(f([1, 2, 3]), 0.0)
+        self.assertEqual(f({1: 2, 3: 4}), 0.0)
+
+        # 'default' argument
+        default = "default"
+        for value in INVALID_ALT:
+            self.assertEqual(f(value, default), default)
+        self.assertEqual(f("zzz", default), default)
+
     def test_parse_query(self, f=text.parse_query):
         # standard usage
         self.assertEqual(f(""), {})