Coverage for src / competitive_verifier / oj / languages / special_comments.py: 100%

1# Python Version: 3.x 

2import functools 

3import pathlib 

4import re 

5from collections.abc import Iterable, Mapping 

6from logging import getLogger 

7 

8from competitive_verifier.util import read_text_normalized 

9 

10logger = getLogger(__name__) 

11 

12SPECIAL_COMMENTS_PATTERN = re.compile( 

13 r"(?:verify-helper|verification-helper|competitive-verifier):\s*([0-9A-Za-z_]+)(?:\s(.*))?$" 

14) 

15 

16 

17# special comments like Vim and Python: see https://www.python.org/dev/peps/pep-0263/ 

18@functools.cache 

19def list_special_comments(path: pathlib.Path) -> Mapping[str, str]: 

20 attributes: dict[str, str] = {} 

21 for line in read_text_normalized(path).splitlines(): 

22 matched = SPECIAL_COMMENTS_PATTERN.search(line) 

23 if matched: 

24 key = matched.group(1) 

25 value = (matched.group(2) or "").strip() 

26 attributes[key] = value 

27 return attributes 

28 

29 

30def _unquote(s: str) -> str: 

31 if s.startswith(("'", '"', "`")): 

32 end_quote_pos = s.rfind(s[0]) 

33 if end_quote_pos == 0: 

34 # Remove opening quote from the URL like `"https://atcoder.jp/` 

35 return s[1:] 

36 # Remove quotes and trailing superfluous chars around the URL 

37 return s[1:end_quote_pos] 

38 return s 

39 

40 

41@functools.cache 

42def list_embedded_urls(path: pathlib.Path) -> Iterable[str]: 

43 pattern = re.compile( 

44 r"""['"`]?https?://\S*""" 

45 ) # use a broad pattern. There are no needs to make match strict. 

46 content = read_text_normalized(path) 

47 # The URL may be written like `"https://atcoder.jp/"`. In this case, we need to remove `"`s around the URL. 

48 # We also need to remove trailing superfluous chars in a case like `{"url":"https://atcoder.jp/"}`. 

49 urls = {_unquote(url) for url in pattern.findall(content)} 

50 return sorted(urls)