Skip to content

Latest commit

 

History

History
73 lines (58 loc) · 2.47 KB

README.md

File metadata and controls

73 lines (58 loc) · 2.47 KB
Raw

regex_spm

Enable Structural Pattern Matching (PEP-634) for Python regular expressions

The string goes in the place of the subject, wrapped by a regex_spm function.
The various possible regex patterns go after the cases.

Simple example

import regex_spm

match regex_spm.fullmatch_in("abracadabra"):
  case r"\d+": print("It's all digits")
  case r"\D+": print("There are no digits in the search string")
  case _: print("It's something else")

regex_spm provides 3 functions, which correspond to re search functions:

  • re.search → regex_spm.search_in
  • re.match → regex_spm.match_in
  • re.fullmatch → regex_spm.fullmatch_in

Accessing the match object

To access the matched groups or the entire re.Match object, simply capture the pattern:

match regex_spm.fullmatch_in("123,45"):
  case r"(\d+),(?P<second>\d+)" as m:
    print("Notice the `as m` at the end of the line above")
    print(f"The first group is {m[1]}")
    print(f"The second group is {m['second']}")
    print(f"The full `re.Match` object is available as {m.match}")

Pattern caveats (workaround below)

#1: Due to the mechanics of Python SPM, it is not possible to use case blocks like case re.compile(r"my-regex"): .... Python would try to use re.compile as a class name, which does not exist.

#2: It is also not possible to save patterns to simple local variables like:

pattern1 = r"my-first-pattern"
pattern2 = re.compile(r"my-second-pattern")
match regex_spm.search_in(my_string):
  case pattern1: print("This does not work, it matches any string. Python interprets `pattern1` "
                       "as simply a new capture variable name, hiding its previous value.")
  case pattern2: print("This does not work either")

Using patterns from variables (caveat workaround)

Python requires "value patterns" to use a "dotted name", i.e. there must be a . in tha name used to reach the stored pattern.

Putting stored patterns in a class like below is sufficient.

class PageRegexes:
  index = re.compile(r"example\.com/index.html")
  shopping_cart = r"example\.com/shopping_cart.html"

match regex_spm.search_in(my_url):
  case PageRegexes.index: print("It's the index page")
  case PageRegexes.shopping_cart: print("It's the shopping cart page")
  case r"example\.com/about.html": print("You can even mix and match")

Install

Requires Python 3.10+ of course

pip install regex_spm