Prevent syntax error on MySQL search when query includes symbols

Fixes #8614 and #12811

We match MySQL's behaviour of treating non-alphanumeric characters as token separators, equivalent to spaces. This way, the Lexeme instances consist of alphanumerics only, avoiding escaping issues that lead to these syntax errors.

(note: the \w class in regexp also counts underscores as "alphanumeric", but this is also consistent with the MySQL behaviour and presents no issues with escaping.)

wagtail/search/backends/database/mysql/mysql.py

@@ -1,3 +1,4 @@
+import re
 import warnings
 from collections import OrderedDict
 
@@ -344,7 +345,10 @@ class MySQLSearchQueryCompiler(BaseSearchQueryCompiler):
 
     def build_search_query_content(self, query, invert=False):
         if isinstance(query, PlainText):
-            terms = query.query_string.split()
+            # For Boolean full text search queries in MySQL,
+            # non-alphanumeric characters act as separators
+            terms = [term for term in re.split(r"\W+", query.query_string) if term]
+
             if not terms:
                 return SearchQuery("")
 

wagtail/search/tests/test_mysql_backend.py

@@ -87,6 +87,62 @@ class TestMySQLSearchBackend(BackendTests, TransactionTestCase):
             set(),
         )
 
+    def test_empty_autocomplete(self):
+        results = self.backend.autocomplete("", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            set(),
+        )
+
+        results = self.backend.autocomplete(" ", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            set(),
+        )
+
+        results = self.backend.autocomplete("*", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            set(),
+        )
+
+    def test_symbols_in_search_term(self):
+        # symbols as their own tokens should be ignored
+        results = self.backend.search("javascript @ parts", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            {"JavaScript: The good parts"},
+        )
+
+        results = self.backend.search("javascript parts @", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            {"JavaScript: The good parts"},
+        )
+
+        results = self.backend.search("@ javascript parts", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            {"JavaScript: The good parts"},
+        )
+
+        # tokens containing both symbols and alphanumerics should not be discarded
+        # or treated as equivalent to the same token without symbols
+        results = self.backend.search("java@script parts", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            set(),
+        )
+
+    def test_autocomplete_with_symbols(self):
+        # the * is not part of the autocomplete mechanism, but if someone includes it
+        # we want it to be gracefully ignored
+        results = self.backend.autocomplete("parts javasc*", models.Book.objects.all())
+        self.assertSetEqual(
+            {r.title for r in results},
+            {"JavaScript: The good parts"},
+        )
+
     @skip(
         "The MySQL backend doesn't support choosing individual fields for the search, only (body, title) or (autocomplete) fields may be searched."
     )