Celes Renata
4 lines
1.7 KiB
Plaintext
4 lines
1.7 KiB
Plaintext
# file: /home/celes/sources/celesrenata/stonks-oracle/services/parser/html_parser.py
|
|
# hypothesis_version: 6.151.14
|
|
|
|
[0.0, 0.05, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.5, 0.6, 0.65, 0.7, 0.8, 0.85, 0.9, 0.95, 1.0, 4.0, 150, 300, '!', '#', '#article-body', '#story-body', '(?i)^tags:\\s*$', '(?i)read more:?\\s*$', ',', '.', '.article-body', '.article-content', '.article-text', '.content-body', '.entry-content', '.post-body', '.post-content', '.story-body', '10-k', '10-q', '8-k', ':', '?', '[.!?]+(?:\\s|$)', "[role='main']", '\\b', '\\n\\s*\\n', '\\s*\\[.*?ad.*?\\]\\s*', 'a', 'ad-container', 'ad-slot', 'advert', 'alias', 'alias_type', 'application/ld+json', 'article', 'aside', 'author', 'banner', 'body', 'body_found', 'brand', 'breadcrumb', 'button', 'canonical', 'canonical_url', 'class', 'comment', 'company_id', 'confidence', 'content', 'cookie', 'datePublished', 'description', 'div', 'diversity', 'earnings-call', 'earnings_call', 'edgar', 'en', 'filing', 'follow-us', 'footer', 'form', 'header', 'high', 'href', 'html', 'html.parser', 'html_parser', 'http', 'https', 'id', 'iframe', 'javascript:', 'keywords', 'lang', 'language', 'legal_name', 'link', 'low', 'main', 'match_count', 'medium', 'mention_type', 'menu', 'meta', 'metadata', 'modal', 'name', 'nav', 'nav-', 'newsletter', 'newsroom', 'noscript', 'og:description', 'og:site_name', 'og:title', 'og:url', 'p', 'pagination', 'paragraph', 'popup', 'press-release', 'press_release', 'promo', 'published_at', 'publisher', 'related-posts', 'script', 'sec.gov', 'section', 'sentence', 'share-bar', 'share-button', 'short_text', 'sidebar', 'signup', 'social-media', 'social-share', 'sponsored', 'style', 'subscribe', 'svg', 'tags', 'td', 'ticker', 'title', 'toolbar', 'transcript', 'very_short_text', 'widget', 'word_count'] |