[minhateca] Add extractor (Fixes #4094)

4 files changed, 78 insertions, 2 deletions

diff --git a/test/test_utils.py b/test/test_utils.py
index baa3a2156..04f1bf283 100644
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -376,6 +376,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_filesize('2 MiB'), 2097152)
         self.assertEqual(parse_filesize('5 GB'), 5000000000)
         self.assertEqual(parse_filesize('1.2Tb'), 1200000000000)
+        self.assertEqual(parse_filesize('1,24 KB'), 1240)
 
 if __name__ == '__main__':
     unittest.main()
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 647352b59..a56ec4fb5 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -217,6 +217,7 @@ from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
+from .minhateca import MinhatecaIE
 from .ministrygrid import MinistryGridIE
 from .mit import TechTVMITIE, MITIE, OCWMITIE
 from .mitele import MiTeleIE
diff --git a/youtube_dl/extractor/minhateca.py b/youtube_dl/extractor/minhateca.py
new file mode 100644
index 000000000..077c9b19d
--- /dev/null
+++ b/youtube_dl/extractor/minhateca.py
@@ -0,0 +1,71 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse,
+    compat_urllib_request,
+)
+from ..utils import (
+    int_or_none,
+    parse_filesize,
+)
+
+
+class MinhatecaIE(InfoExtractor):
+    _VALID_URL = r'https?://minhateca\.com\.br/[^?#]+,(?P<id>[0-9]+)\.'
+    _TEST = {
+        'url': 'http://minhateca.com.br/pereba/misc/youtube-dl+test+video,125848331.mp4(video)',
+        'info_dict': {
+            'id': '125848331',
+            'ext': 'mp4',
+            'title': 'youtube-dl test video',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'filesize_approx': 1530000,
+            'duration': 9,
+            'view_count': int,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        token = self._html_search_regex(
+            r'<input name="__RequestVerificationToken".*?value="([^"]+)"',
+            webpage, 'request token')
+        token_data = [
+            ('fileId', video_id),
+            ('__RequestVerificationToken', token),
+        ]
+        req = compat_urllib_request.Request(
+            'http://minhateca.com.br/action/License/Download',
+            data=compat_urllib_parse.urlencode(token_data))
+        req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        data = self._download_json(
+            req, video_id, note='Downloading metadata')
+
+        video_url = data['redirectUrl']
+        title_str = self._html_search_regex(
+            r'<h1.*?>(.*?)</h1>', webpage, 'title')
+        title, _, ext = title_str.rpartition('.')
+        filesize_approx = parse_filesize(self._html_search_regex(
+            r'<p class="fileSize">(.*?)</p>',
+            webpage, 'file size approximation', fatal=False))
+        duration = int_or_none(self._html_search_regex(
+            r'(?s)<p class="fileLeng[ht][th]">.*?([0-9]+)\s*s',
+            webpage, 'duration', fatal=False))
+        view_count = int_or_none(self._html_search_regex(
+            r'<p class="downloadsCounter">([0-9]+)</p>',
+            webpage, 'view count', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'ext': ext,
+            'filesize_approx': filesize_approx,
+            'duration': duration,
+            'view_count': view_count,
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 4d3cbac74..5e9ae7a42 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -1090,11 +1090,14 @@ def parse_filesize(s):
     }
 
     units_re = '|'.join(re.escape(u) for u in _UNIT_TABLE)
-    m = re.match(r'(?P<num>[0-9]+(?:\.[0-9]*)?)\s*(?P<unit>%s)' % units_re, s)
+    m = re.match(
+        r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)' % units_re, s)
     if not m:
         return None
 
-    return int(float(m.group('num')) * _UNIT_TABLE[m.group('unit')])
+    num_str = m.group('num').replace(',', '.')
+    mult = _UNIT_TABLE[m.group('unit')]
+    return int(float(num_str) * mult)
 
 
 def get_term_width():