Coverage for src/local_deep_research/content_fetcher/__init__.py: 100%

1""" 

2Content Fetcher Module. 

3 

4Provides unified content fetching from various sources: 

5- Academic papers (arXiv, PubMed, Semantic Scholar, bioRxiv) 

6- Web pages (HTML) 

7- Direct PDF links 

8 

9Usage: 

10 from local_deep_research.content_fetcher import ContentFetcher 

11 

12 fetcher = ContentFetcher() 

13 result = fetcher.fetch("https://arxiv.org/abs/2301.12345") 

14 print(result["content"]) 

15""" 

16 

17from .url_classifier import URLClassifier, URLType 

18 

19__all__ = ["ContentFetcher", "URLClassifier", "URLType"] 

20 

21 

22def __getattr__(name): 

23 """Lazily resolve ``ContentFetcher`` on first attribute access. 

24 

25 ``fetcher`` transitively pulls in the full web/DB stack 

26 (requests, playwright, … via ``research_library.downloaders``). 

27 Deferring the import lets 

28 ``from ...content_fetcher.url_classifier import URLClassifier`` 

29 stay cheap and lets minimal test setups import the package 

30 without that dependency tree. See issue #4992. 

31 """ 

32 if name == "ContentFetcher": 

33 from .fetcher import ContentFetcher as _ContentFetcher 

34 

35 globals()["ContentFetcher"] = _ContentFetcher 

36 return _ContentFetcher 

37 raise AttributeError(f"module {__name__!r} has no attribute {name!r}") 

38 

39 

40def __dir__(): 

41 """Expose lazily-resolved names to ``dir()`` and tab completion.""" 

42 return sorted(set(globals()) | {"ContentFetcher"})