Fix parsing of names containing special chars (fixes 291).

andialbrecht · andialbrecht · commit 358ad4b97547 · 2016-09-14T14:45:44.000+02:00
diff --git a/CHANGELOG b/CHANGELOG
@@ -4,6 +4,7 @@ Development Version
 Bug Fixes
 
 * Fix parsing of incomplete AS (issue284, by vmuriart).
+* Fix parsing of Oracle names containing dollars (issue291).
 
 
 Internal Changes
diff --git a/sqlparse/keywords.py b/sqlparse/keywords.py
@@ -39,7 +39,7 @@ def is_keyword(value):
 
         (r'\?', tokens.Name.Placeholder),
         (r'%(\(\w+\))?s', tokens.Name.Placeholder),
-        (r'[$:?]\w+', tokens.Name.Placeholder),
+        (r'(?<!\w)[$:?]\w+', tokens.Name.Placeholder),
 
         # FIXME(andi): VALUES shouldn't be listed here
         # see https://github.com/andialbrecht/sqlparse/pull/64
@@ -76,7 +76,7 @@ def is_keyword(value):
         (r'CREATE(\s+OR\s+REPLACE)?\b', tokens.Keyword.DDL),
         (r'DOUBLE\s+PRECISION\b', tokens.Name.Builtin),
 
-        (r'[_A-Z]\w*', is_keyword),
+        (r'[_A-Z][_$#\w]*', is_keyword),
 
         (r'[;:()\[\],\.]', tokens.Punctuation),
         (r'[<>=~!]+', tokens.Operator.Comparison),
diff --git a/tests/test_tokenize.py b/tests/test_tokenize.py
@@ -163,3 +163,17 @@ def test_parse_endifloop(s):
     p = sqlparse.parse(s)[0]
     assert len(p.tokens) == 1
     assert p.tokens[0].ttype is T.Keyword
+
+
+@pytest.mark.parametrize('s', [
+    'foo',
+    'Foo',
+    'FOO',
+    'v$name',  # issue291
+])
+def test_parse_identifiers(s):
+    p = sqlparse.parse(s)[0]
+    assert len(p.tokens) == 1
+    token = p.tokens[0]
+    assert str(token) == s
+    assert isinstance(token, sql.Identifier)
