Merge branch 'master' into v0.2.0

CHANGES

@@ -1,6 +1,25 @@
 Development Version
 -------------------
 
+Nothing yet.
+
+
+Release 0.1.16 (Jul 26, 2015)
+-----------------------------
+
+Bug Fixes
+* Fix a regression in get_alias() introduced in 0.1.15 (issue185).
+* Fix a bug in the splitter regarding DECLARE (issue193).
+* sqlformat command line tool doesn't duplicat newlines anymore (issue191).
+* Don't mix up MySQL comments starting with hash and MSSQL
+  temp tables (issue192).
+* Statement.get_type() now ignores comments at the beginning of
+  a statement (issue186).
+
+
+Release 0.1.15 (Apr 15, 2015)
+-----------------------------
+
 Bug Fixes
 * Fix a regression for identifiers with square bracktes
   notation (issue153, by darikg).

bin/sqlformat

@@ -77,7 +77,7 @@ def main():
         data = sys.stdin.read()
     else:
         try:
-            data = '\n'.join(open(args[0]).readlines())
+            data = ''.join(open(args[0]).readlines())
         except OSError:
             err = sys.exc_info()[1]  # Python 2.5 compatibility
             _error('Failed to read %s: %s' % (args[0], err), exit_=1)

sqlparse/__init__.py

@@ -6,7 +6,7 @@
 """Parse SQL statements."""
 
 
-__version__ = '0.1.14'
+__version__ = '0.1.17-dev'
 
 # Setup namespace
 from sqlparse import compat

sqlparse/engine/filter.py

@@ -40,7 +40,7 @@ def _change_splitlevel(self, ttype, value):
 
         unified = value.upper()
 
-        if unified == 'DECLARE' and self._is_create:
+        if unified == 'DECLARE' and self._is_create and self._begin_depth == 0:
             self._in_declare = True
             return 1
 

sqlparse/lexer.py

@@ -141,10 +141,10 @@ class Lexer(compat.with_metaclass(LexerMeta)):
 
     tokens = {
         'root': [
-            (r'(--|#).*?(\r\n|\r|\n)', tokens.Comment.Single),
+            (r'(--|# ).*?(\r\n|\r|\n)', tokens.Comment.Single),
             # $ matches *before* newline, therefore we have two patterns
             # to match Comment.Single
-            (r'(--|#).*?$', tokens.Comment.Single),
+            (r'(--|# ).*?$', tokens.Comment.Single),
             (r'(\r\n|\r|\n)', tokens.Newline),
             (r'\s+', tokens.Whitespace),
             (r'/\*', tokens.Comment.Multiline, 'multiline-comments'),
@@ -162,7 +162,7 @@ class Lexer(compat.with_metaclass(LexerMeta)):
             # FIXME(andi): VALUES shouldn't be listed here
             # see https://github.com/andialbrecht/sqlparse/pull/64
             (r'VALUES', tokens.Keyword),
-            (r'@[^\W\d_]\w+', tokens.Name),
+            (r'(@|##|#)[^\W\d_]\w+', tokens.Name),
             # IN is special, it may be followed by a parenthesis, but
             # is never a functino, see issue183
             (r'in\b(?=[ (])?', tokens.Keyword),

sqlparse/sql.py

@@ -233,15 +233,20 @@ def get_sublists(self):
     def _groupable_tokens(self):
         return self.tokens
 
-    def token_first(self, ignore_whitespace=True):
+    def token_first(self, ignore_whitespace=True, ignore_comments=False):
         """Returns the first child token.
 
         If *ignore_whitespace* is ``True`` (the default), whitespace
         tokens are ignored.
+
+        if *ignore_comments* is ``True`` (default: ``False``), comments are
+        ignored too.
         """
         for token in self.tokens:
             if ignore_whitespace and token.is_whitespace():
                 continue
+            if ignore_comments and isinstance(token, Comment):
+                continue
             return token
 
     def token_next_by_instance(self, idx, clss):
@@ -390,7 +395,8 @@ def get_alias(self):
             return self._get_first_name(kw, keywords=True)
 
         # "name alias" or "complicated column expression alias"
-        if len(self.tokens) > 2:
+        if len(self.tokens) > 2 \
+           and self.token_next_by_type(0, T.Whitespace) is not None:
             return self._get_first_name(reverse=True)
 
         return None
@@ -460,8 +466,11 @@ def get_type(self):
         The returned value is a string holding an upper-cased reprint of
         the first DML or DDL keyword. If the first token in this group
         isn't a DML or DDL keyword "UNKNOWN" is returned.
+
+        Whitespaces and comments at the beginning of the statement
+        are ignored.
         """
-        first_token = self.token_first()
+        first_token = self.token_first(ignore_comments=True)
         if first_token is None:
             # An "empty" statement that either has not tokens at all
             # or only whitespace tokens.

tests/test_grouping.py

@@ -187,6 +187,11 @@ def test_alias_case(self):  # see issue46
         self.assertEqual(len(p.tokens), 1)
         self.assertEqual(p.tokens[0].get_alias(), 'foo')
 
+    def test_alias_returns_none(self):  # see issue185
+        p = sqlparse.parse('foo.bar')[0]
+        self.assertEqual(len(p.tokens), 1)
+        self.assertEqual(p.tokens[0].get_alias(), None)
+
     def test_idlist_function(self):  # see issue10 too
         p = sqlparse.parse('foo(1) x, bar')[0]
         self.assert_(isinstance(p.tokens[0], sql.IdentifierList))

tests/test_parse.py

@@ -294,3 +294,13 @@ def test_single_line_comments(sql):
     assert p.tokens[-1].ttype == T.Comment.Single
 
 
+@pytest.mark.parametrize('sql', [
+    'foo',
+    '@foo',
+    '#foo',  # see issue192
+    '##foo'
+])
+def test_names_and_special_names(sql):
+    p = sqlparse.parse(sql)[0]
+    assert len(p.tokens) == 1
+    assert isinstance(p.tokens[0], sqlparse.sql.Identifier)

tests/test_regressions.py

@@ -248,3 +248,20 @@ def test_null_with_as():
         'FROM t1'
     ])
     assert formatted == tformatted
+
+
+def test_issue193_splitting_function():
+    sql = """CREATE FUNCTION a(x VARCHAR(20)) RETURNS VARCHAR(20)
+BEGIN
+ DECLARE y VARCHAR(20);
+ RETURN x;
+END;
+SELECT * FROM a.b;"""
+    splitted = sqlparse.split(sql)
+    assert len(splitted) == 2
+
+
+def test_issue186_get_type():
+    sql = "-- comment\ninsert into foo"
+    p = sqlparse.parse(sql)[0]
+    assert p.get_type() == 'INSERT'