remove containers from identifiers

tests/test_identifier_parser.py

@@ -1,3 +1,5 @@
+import pytest
+
 from wikitools import identifier_parser
 
 
@@ -75,6 +77,17 @@ def test__link(self):
         ):
             assert identifier_parser.extract_identifier(heading) == (identifier, 0)
 
+    @pytest.mark.parametrize(
+        "payload",
+        [
+            {"heading": "### ::{ flag=TH }:: woah", "identifier": "woah"},
+            {"heading": "### woah ::{ flag=TH }::", "identifier": "woah"},
+            {"heading": "### ::{ flag=TH }:: Thailand vs. ::{ flag=NZ }:: New Zealand", "identifier": "thailand-vs.--new-zealand"},
+        ]
+    )
+    def test__flag(self, payload):
+        assert identifier_parser.extract_identifier(payload["heading"]) == (payload["identifier"], 0)
+
     def test__custom(self):
         for line, identifier, pos in (
             ('osu! is a free-to-win game.', None, 0),

wikitools/identifier_parser.py

@@ -1,11 +1,15 @@
 import typing
+import re
 
 from wikitools import link_parser
 
 ID_PREFIXES = {'#', 'id='}
 
 ESCAPEABLE_CHARS = { '\\', '(' , ')', '[', ']', '{', '}', '<', '>', '#', '*', '~', '`', '_', '-', '+', '|', '\'', '"', '@', '$', ';', ':', ',', '.' }
 
+CONTAINER_REGEX = re.compile(r"::{.*?}::")
+
+
 # backslashes are sometimes used (perhaps unnecessarily) to avoid remark errors
 def unescape(s: str) -> str:
     unescaped = ""
@@ -76,4 +80,10 @@ def extract_identifier(
             if k == len(links_on_line) - 1:
                 heading += s[start:]
 
-    return ("-".join(word.lower() for word in unescape(heading).strip().split()), 0)
+    identifier = "-".join(word.lower() for word in unescape(heading).strip().split())
+
+    # headings can contain custom containers, such as flags
+    # TODO: maybe do this in a smarter way
+    identifier = re.sub(CONTAINER_REGEX, "", identifier).strip("-")
+
+    return (identifier, 0)

wikitools_cli/VERSION.py

@@ -1 +1 @@
-VERSION = "2.2.0"
+VERSION = "2.2.1"