spaCy 3.x RegularExpressionMatcher: Patterns and on_match pattern handlers

bibliographer_patterns.py

"""
A quick excerpt demonstrating usage of a custom `RegularExpressionMatcher` for spaCy 3.
This is from one of my personal projects (HaleyNLP/Irnerius). Module-level imports and other code blocks have been elided.
"""

@registry.misc('haleynlp.common.extraction.handler.on_match.bibliography._european_union_ecli')
def _european_union_ecli(
    span: Span,
    match: re.Match,
) -> None:
    """
    An `on_match` pattern handler for extracting ECLI citations. 
    
    Args:
        span (Span): 
            The spaCy Span in which a regular expression match was found.
        match (re.Match):
            A found regular expression match.
    
    Returns:
        None
    """
    
    # The majority of this function's implementation has been omitted
    # ... In short, it assembles an `Annotation` object from the regular expression `Match` and stores it on the `Doc` object 
    # ... (the `_.annotations` extension is created by another custom component not shown in this example)
    
    annotation: Annotation = ...
    span.doc._.annotations.citations.add(annotation)


MATCHING_RULES: Tuple = (

  {
      'key': '_european_union_ecli',
      'patterns': [
          re.compile(
              pattern=r"""
                  (?P<ECLI>
                      (?P<IDENTIFIER>ECLI):
                      (?P<COUNTRY>[A-Z]{2}):
                      (?P<COURT>[A-Z][A-Z0-9]{0,6}):
                      (?P<YEAR>\d{4}):
                      (?P<CASE>(\d|\.){1,25})
                  )
              """,
              flags=re.VERBOSE,
          )
      ],
      'on_match': 'haleynlp.common.extraction.handler.on_match.bibliography._european_union_ecli',
  },
  
  # ...other elements for different citation types, like United States Acts, Code, Laws, and court cases...

)


@registry.misc('haleynlp.en.component.config.bibliography')
def create_patterns_bibliography() -> Tuple:
    return MATCHING_RULES