diff options
| -rw-r--r-- | youtube/comments.py | 2 | ||||
| -rw-r--r-- | youtube/yt_data_extract/common.py | 17 | ||||
| -rw-r--r-- | youtube/yt_data_extract/everything_else.py | 12 |
3 files changed, 18 insertions, 13 deletions
diff --git a/youtube/comments.py b/youtube/comments.py index a7a93cc..b3f1a90 100644 --- a/youtube/comments.py +++ b/youtube/comments.py @@ -90,7 +90,7 @@ def single_comment_ctoken(video_id, comment_id): def post_process_comments_info(comments_info): for comment in comments_info['comments']: comment['author_url'] = concat_or_none( - util.URL_ORIGIN, comment['author_url']) + '/', comment['author_url']) comment['author_avatar'] = concat_or_none( settings.img_prefix, comment['author_avatar']) diff --git a/youtube/yt_data_extract/common.py b/youtube/yt_data_extract/common.py index 2d3b637..9610479 100644 --- a/youtube/yt_data_extract/common.py +++ b/youtube/yt_data_extract/common.py @@ -90,15 +90,20 @@ def remove_redirect(url): return urllib.parse.parse_qs(query_string)['q'][0] return url -youtube_url_re = re.compile(r'^(?:(?:(?:https?:)?//)?(?:www\.)?youtube\.com)?(/.*)$') +norm_url_re = re.compile(r'^(?:(?:https?:)?//)?((?:[\w-]+\.)+[\w-]+)?(/.*)$') def normalize_url(url): + '''Insert https, resolve relative paths for youtube.com, and put www. infront of youtube.com''' if url is None: return None - match = youtube_url_re.fullmatch(url) + match = norm_url_re.fullmatch(url) if match is None: - raise Exception() + raise Exception(url) - return 'https://www.youtube.com' + match.group(1) + domain = match.group(1) or 'www.youtube.com' + if domain == 'youtube.com': + domain = 'www.youtube.com' + + return 'https://' + domain + match.group(2) def _recover_urls(runs): for run in runs: @@ -240,11 +245,11 @@ def extract_item_info(item, additional_info={}): )) info['author_url'] = ('https://www.youtube.com/channel/' + info['author_id']) if info['author_id'] else None info['description'] = extract_formatted_text(multi_get(item, 'descriptionSnippet', 'descriptionText')) - info['thumbnail'] = multi_deep_get(item, + info['thumbnail'] = normalize_url(multi_deep_get(item, ['thumbnail', 'thumbnails', 0, 'url'], # videos ['thumbnails', 0, 'thumbnails', 0, 'url'], # playlists ['thumbnailRenderer', 'showCustomThumbnailRenderer', 'thumbnail', 'thumbnails', 0, 'url'], # shows - ) + )) info['badges'] = [] for badge_node in multi_get(item, 'badges', 'ownerBadges', default=()): diff --git a/youtube/yt_data_extract/everything_else.py b/youtube/yt_data_extract/everything_else.py index 5bb8709..d91dad5 100644 --- a/youtube/yt_data_extract/everything_else.py +++ b/youtube/yt_data_extract/everything_else.py @@ -49,10 +49,10 @@ def extract_channel_info(polymer_json, tab): if info['short_description'] and len(info['short_description']) > 730: info['short_description'] = info['short_description'][0:730] + '...' info['channel_name'] = metadata.get('title') - info['avatar'] = multi_deep_get(metadata, + info['avatar'] = normalize_url(multi_deep_get(metadata, ['avatar', 'thumbnails', 0, 'url'], ['thumbnail', 'thumbnails', 0, 'url'], - ) + )) channel_url = multi_get(metadata, 'urlCanonical', 'channelUrl') if channel_url: channel_id = get(channel_url.rstrip('/').split('/'), -1) @@ -263,13 +263,13 @@ def extract_comments_info(polymer_json): # These 3 are sometimes absent, likely because the channel was deleted comment_info['author'] = extract_str(comment_renderer.get('authorText')) - comment_info['author_url'] = deep_get(comment_renderer, - 'authorEndpoint', 'commandMetadata', 'webCommandMetadata', 'url') + comment_info['author_url'] = normalize_url(deep_get(comment_renderer, + 'authorEndpoint', 'commandMetadata', 'webCommandMetadata', 'url')) comment_info['author_id'] = deep_get(comment_renderer, 'authorEndpoint', 'browseEndpoint', 'browseId') - comment_info['author_avatar'] = deep_get(comment_renderer, - 'authorThumbnail', 'thumbnails', 0, 'url') + comment_info['author_avatar'] = normalize_url(deep_get( + comment_renderer, 'authorThumbnail', 'thumbnails', 0, 'url')) comment_info['id'] = comment_renderer.get('commentId') comment_info['text'] = extract_formatted_text(comment_renderer.get('contentText')) comment_info['time_published'] = extract_str(comment_renderer.get('publishedTimeText')) |