From ccf6a90ba861e8664dcf296ee5b798a259b74a52 Mon Sep 17 00:00:00 2001
From: bato3 <bato3@bandyci.org>
Date: Sat, 14 Jul 2018 00:32:41 +0200
Subject: [PATCH] fixes after first review

---
 youtube_dl/extractor/mp4upload.py | 75 +++++++++++++++----------------
 1 file changed, 37 insertions(+), 38 deletions(-)

diff --git a/youtube_dl/extractor/mp4upload.py b/youtube_dl/extractor/mp4upload.py
index 2bb985214..d03b4cc78 100644
--- a/youtube_dl/extractor/mp4upload.py
+++ b/youtube_dl/extractor/mp4upload.py
@@ -9,6 +9,8 @@ from ..utils import (
     decode_packed_codes,
     get_element_by_class,
     get_element_by_id,
+    int_or_none,
+    float_or_none,
     parse_filesize,
     strip_or_none,
 )
@@ -20,25 +22,19 @@ class Mp4UploadIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://www.mp4upload.com/e52ycvdl4x29',
         'md5': '09780a74b0de79ada5f9a8955f0704fc',
-
         'info_dict': {
             'id': 'e52ycvdl4x29',
             'ext': 'mp4',
             'title': '橋本潮 - ロマンティックあげるよ.mp4',
             'timestamp': 1467471956,
             'thumbnail': r're:^https?://.*\.jpg$',
-
             'vcodec': 'ffh264',
             'width': 454,
             'height': 360,
             'fps': 29.970,
-
             'acodec': 'ffaac',
             'asr': 44100,
             'abr': 96,
-
-            # Something adds this to the _real_extract return value, and the test runner expects it present.
-            # Should probably be autocalculated from the timestamp instead, just like _real_extract.
             'upload_date': '20160702',
         },
     }, {
@@ -64,13 +60,6 @@ class Mp4UploadIE(InfoExtractor):
             'id': video_id,
         }
 
-        file_info = re.findall(
-            r'>(?P<label>[^<:]+):</span><span>(?P<value>[^<]+)</span></li>',
-            get_element_by_class('fileinfo', webpage)
-        )
-        if not file_info:
-            raise ExtractorError('I can\'t find file info', video_id=video_id)
-
         embedpage = self._download_webpage(embed_url, video_id, note='Downloading embed webpage')
 
         # It contains only `source url` and `thumbnail`
@@ -85,34 +74,41 @@ class Mp4UploadIE(InfoExtractor):
 
         info_dict['thumbnail'] = poor_info_dict.get('thumbnail')
         _f = {
-            'url': poor_info_dict.get('formats')[0].get('url'),
-            'ext': poor_info_dict.get('formats')[0].get('ext'),
+            'url': poor_info_dict.get('formats', [{}])[0].get('url'),
+            'ext': poor_info_dict.get('formats', [{}])[0].get('ext'),
             'format_id': '1',
         }
 
-        for info in file_info:
-            if info[0] == 'Codec':
-                _f['vcodec'] = info[1]
+        file_info = re.findall(
+            r'>(?P<label>[^<:]+):</span><span>(?P<value>[^<]+)</span></li>',
+            get_element_by_class('fileinfo', webpage)
+        )
+        if file_info:
+            for info in file_info:
+                if info[0] == 'Codec':
+                    _f['vcodec'] = info[1]
 
-            elif info[0] == 'Resolution':
-                _f['resolution'] = info[1].replace(' ', '')
-                resmatch = re.search(r'(?P<width>[\d]+) x (?P<height>[\d]+)', info[1])
-                if resmatch:
-                    _f['width'] = int(resmatch.group('width'))
-                    _f['height'] = int(resmatch.group('height'))
+                elif info[0] == 'Resolution':
+                    _f['resolution'] = info[1].replace(' ', '')
+                    resmatch = re.search(r'(?P<width>\d+)\s*x\s*(?P<height>\d+)', info[1])
+                    if resmatch:
+                        _f['width'] = int(resmatch.group('width'))
+                        _f['height'] = int(resmatch.group('height'))
 
-            elif info[0] == 'Framerate' and info[1] != '1000.000 fps':
-                _f['fps'] = float(info[1].replace(' fps', ''))
+                elif info[0] == 'Framerate':
+                    fps = float_or_none(re.sub(r'[^\d\.]+', '', info[1]))
+                    if fps < 100:
+                        _f['fps'] = fps
 
-            elif info[0] == 'Audio info':
-                audmatch = re.search(r'(?P<acodec>.+?), (?P<abr>[\d]+) kbps, (?P<asr>[\d]+) Hz', info[1])
-                if audmatch:
-                    _f['acodec'] = audmatch.group('acodec')
-                    _f['abr'] = int(audmatch.group('abr'))
-                    _f['asr'] = int(audmatch.group('asr'))
+                elif info[0] == 'Audio info':
+                    audmatch = re.search(r'(?P<acodec>.+?), (?P<abr>\d+) kbps, (?P<asr>\d+) Hz', info[1])
+                    if audmatch:
+                        _f['acodec'] = audmatch.group('acodec')
+                        _f['abr'] = int(audmatch.group('abr'))
+                        _f['asr'] = int(audmatch.group('asr'))
 
-            elif info[0] == 'Bitrate':
-                _f['vbr'] = int(info[1].replace(' Kbps', ''))
+                elif info[0] == 'Bitrate':
+                    _f['vbr'] = int_or_none(re.sub(r'\D+', '', info[1]))
 
         _f['filesize_approx'] = parse_filesize(
             self._html_search_regex(
@@ -122,11 +118,14 @@ class Mp4UploadIE(InfoExtractor):
         )
         info_dict['formats'] = [_f]
 
-        # can't use _html_search_regex, there's data both inside and outside a bold tag and I need it all
         date_raw = self._search_regex(r'Uploaded on(.+?)</div>', webpage, 'timestamp', fatal=False, flags=re.DOTALL)
         if date_raw:
-            date_raw = re.sub(r'<[^>]+>', '', date_raw)
-            date_raw = re.sub(r'[\s]+', ' ', date_raw)
-            info_dict['timestamp'] = time.mktime(time.strptime(date_raw, ' %Y-%m-%d %H:%M:%S '))
+            try:
+                info_dict['timestamp'] = time.mktime(time.strptime(
+                    re.sub(r'[^\d\-\:]+', '', date_raw),
+                    '%Y-%m-%d%H:%M:%S'
+                ))
+            except ValueError:
+                pass
 
         return info_dict