Merge 2e7b3d38b9 into e8d49b1c7f

yt_dlp/extractor/rule34video.py

@@ -9,6 +9,7 @@ from ..utils import (
     get_element_html_by_class,
     get_elements_by_class,
     int_or_none,
+    merge_dicts,
     parse_count,
     parse_duration,
     unescapeHTML,
@@ -72,45 +73,58 @@ class Rule34VideoIE(InfoExtractor):
 
         formats = []
 
-        for mobj in re.finditer(r'<a[^>]+href="(?P<video_url>[^"]+download=true[^"]+)".*>(?P<ext>[^\s]+) (?P<quality>[^<]+)p</a>', webpage):
-            url, ext, quality = mobj.groups()
+        for mobj in re.finditer(r'<a[^>]+href="(?P<video_url>[^"]+download=true[^"]+)".*>(?P<ext>[^\s]+) (?P<height>[^<]+)p</a>', webpage):
+            url, ext, height = mobj.groups()
             formats.append({
                 'url': url,
                 'ext': ext.lower(),
-                'quality': quality,
+                'height': int_or_none(height),
             })
 
         categories, creators, uploader, uploader_url = [None] * 4
         for col in get_elements_by_class('col', webpage):
             label = clean_html(get_element_by_class('label', col))
-            if label == 'Categories:':
+            if label == 'Categories':
                 categories = list(map(clean_html, get_elements_by_class('item', col)))
-            elif label == 'Artist:':
+            elif label == 'Artist':
                 creators = list(map(clean_html, get_elements_by_class('item', col)))
-            elif label == 'Uploaded By:':
-                uploader = clean_html(get_element_by_class('name', col))
-                uploader_url = extract_attributes(get_element_html_by_class('name', col) or '').get('href')
+            elif label == 'Uploaded by':
+                uploader_link = get_element_html_by_class('btn_link', col)
+                uploader = clean_html(uploader_link)
+                uploader_url = extract_attributes(uploader_link or '').get('href')
 
-        return {
-            **traverse_obj(self._search_json_ld(webpage, video_id, default={}), ({
-                'title': 'title',
-                'view_count': 'view_count',
-                'like_count': 'like_count',
-                'duration': 'duration',
-                'timestamp': 'timestamp',
-                'description': 'description',
-                'thumbnail': ('thumbnails', 0, 'url'),
-            })),
+        view_count, duration = [None] * 2
+        for item_info in get_elements_by_class('item_info', webpage):
+            item_info_text = clean_html(item_info)
+            if get_element_by_class('custom-eye', item_info) is not None:
+                precise_view_count = re.search(r'\(([\d,]+)\)', item_info_text)
+                view_count = parse_count(precise_view_count.group(1) if precise_view_count is not None else item_info_text)
+            elif get_element_by_class('custom-time', item_info) is not None:
+                duration = parse_duration(clean_html(item_info))
+
+        like_count = None
+        if like_count_text := re.search(r'\(([\d,]+)\)', get_element_by_class('voters count', webpage)):
+            like_count = parse_count(like_count_text.group(1))
+
+        json_ld = traverse_obj(self._search_json_ld(webpage, video_id, default={}), ({
+            'title': 'title',
+            'view_count': 'view_count',
+            'like_count': 'like_count',
+            'duration': 'duration',
+            'timestamp': 'timestamp',
+            'description': 'description',
+            'thumbnail': ('thumbnails', 0, 'url'),
+        }))
+
+        return merge_dicts({
             'id': video_id,
             'formats': formats,
             'title': self._html_extract_title(webpage),
             'thumbnail': self._html_search_regex(
                 r'preview_url:\s+\'([^\']+)\'', webpage, 'thumbnail', default=None),
-            'duration': parse_duration(self._html_search_regex(
-                r'"icon-clock"></i>\s+<span>((?:\d+:?)+)', webpage, 'duration', default=None)),
-            'view_count': int_or_none(self._html_search_regex(
-                r'"icon-eye"></i>\s+<span>([ \d]+)', webpage, 'views', default='').replace(' ', '')),
-            'like_count': parse_count(get_element_by_class('voters count', webpage)),
+            'duration': duration,
+            'view_count': view_count,
+            'like_count': like_count,
             'comment_count': int_or_none(self._search_regex(
                 r'[^(]+\((\d+)\)', get_element_by_attribute('href', '#tab_comments', webpage), 'comment count', fatal=False)),
             'age_limit': 18,
@@ -120,4 +134,4 @@ class Rule34VideoIE(InfoExtractor):
             'categories': categories,
             'tags': list(map(unescapeHTML, re.findall(
                 r'<a class="tag_item"[^>]+\bhref="https://rule34video\.com/tags/\d+/"[^>]*>(?P<tag>[^>]*)</a>', webpage))),
-        }
+        }, json_ld)