Merge pull request #11 from maksimstojkovic/main

Add support for wildcard regex tokenization
Sygil-Dev · Jan 3, 2024 · c663c61 · c663c61
2 parents 95b81f8 + 21d803a
commit c663c61
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 3 deletions.
diff --git a/src/whoosh/analysis/tokenizers.py b/src/whoosh/analysis/tokenizers.py
@@ -30,7 +30,7 @@
 from whoosh.util.text import rcompile
 
 
-default_pattern = rcompile(r"\w+(\.?\w+)*")
+default_pattern = rcompile(r"[\w\*]+(\.?[\w\*]+)*")
 
 
 # Tokenizers

diff --git a/src/whoosh/fields.py b/src/whoosh/fields.py
@@ -1224,8 +1224,12 @@ def self_parsing(self):
     def parse_query(self, fieldname, qstring, boost=1.0):
         from whoosh import query
 
-        terms = [query.Term(fieldname, g)
-                 for g in self.process_text(qstring, mode='query')]
+        terms = []
+        for g in self.process_text(qstring, mode='query'):
+            if g == "*":
+                terms.append(query.Wildcard(fieldname, g, boost=boost))
+            else:
+                terms.append(query.Term(fieldname, g, boost=boost))
         cls = query.Or if self.queryor else query.And
 
         return cls(terms, boost=boost)