Refactor extractors: Pythonic improvements, docstrings, and optimizations

[Copilot]

Summary

Reviewed and refactored all files in newspaper/extractors/ for readability, Pythonic style, correctness, and minor performance improvements.

Changes

authors_extractor.py

• Moved inner functions (uniqify_list, parse_byline, get_authors, get_text_from_element) out of parse() and into proper class/static methods
• Pre-compiled regex patterns (_DIGITS_RE, _author_stopwords_re) at class level to avoid recompiling on every parse() call
• Replaced OrderedDict with plain dict (ordered since Python 3.7) — removes an unnecessary import
• Fixed variable shadowing (m was reused for both a regex match and a loop variable)
• Fixed return type annotation on content from str | list[str] to just str

pubdate_extractor.py

• Extracted inner parse_date_str closure to a proper _parse_date_str class method
• Eliminated redundant intermediate date_str local variables where the value was only used once

categories_extractor.py

• Used set.isdisjoint() instead of len(set.intersection()) == 0 (more readable and marginally faster)
• Replaced (x != "http" and x != "https") with idiomatic x not in ("http", "https")
• Used not _valid_categories instead of len(_valid_categories) == 0

content_extractor.py

• Simplified get_feed_urls from a multi-step loop + list + dedup pipeline into a clean set-comprehension
• Added missing module and __init__ docstrings

articlebody_extractor.py

• Used generator expressions in sum() instead of building intermediate lists (reduces memory allocation)
• Fixed add_siblings which was missing the res_node = copy.deepcopy(top_node) assignment
• Added missing docstrings (D100, D101, D102)

videos_extractor.py

• Used getattr(self, ..., None) for safe method dispatch — avoids a potential AttributeError if an unknown video tag is encountered
• Added missing module and __init__ docstrings

image_extractor.py / metadata_extractor.py / title_extractor.py

• Added missing module-level, class-level, and __init__ docstrings
• Fixed parse docstring in image_extractor.py to document all arguments

defines.py

• Added module-level docstring
• Added MetaImageDict class docstring; restored accidentally missing tag: str field

Validation

• ruff check newspaper/extractors/ — ✅ All checks passed
• pytest tests/unit/ — ✅ 128 tests pass (10 pre-existing failures from missing optional dependencies: nltk, pythainlp, indic-nlp-library)
• CodeQL security scan — ✅ 0 alerts