1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
|
import unittest
from sqlglot.dialects import BigQuery
from sqlglot.errors import TokenError
from sqlglot.tokens import Tokenizer, TokenType
class TestTokens(unittest.TestCase):
def test_space_keywords(self):
for string, length in (
("group bys", 2),
(" group bys", 2),
(" group bys ", 2),
("group by)", 2),
("group bys)", 3),
("group \r", 1),
):
tokens = Tokenizer().tokenize(string)
self.assertTrue("GROUP" in tokens[0].text.upper())
self.assertEqual(len(tokens), length)
def test_comment_attachment(self):
tokenizer = Tokenizer()
sql_comment = [
("/*comment*/ foo", ["comment"]),
("/*comment*/ foo --test", ["comment", "test"]),
("--comment\nfoo --test", ["comment", "test"]),
("foo --comment", ["comment"]),
("foo", []),
("foo /*comment 1*/ /*comment 2*/", ["comment 1", "comment 2"]),
("foo\n-- comment", [" comment"]),
("1 /*/2 */", ["/2 "]),
("1\n/*comment*/;", ["comment"]),
]
for sql, comment in sql_comment:
self.assertEqual(tokenizer.tokenize(sql)[0].comments, comment)
def test_token_line_col(self):
tokens = Tokenizer().tokenize(
"""SELECT /*
line break
*/
'x
y',
x"""
)
self.assertEqual(tokens[0].line, 1)
self.assertEqual(tokens[0].col, 6)
self.assertEqual(tokens[1].line, 5)
self.assertEqual(tokens[1].col, 3)
self.assertEqual(tokens[2].line, 5)
self.assertEqual(tokens[2].col, 4)
self.assertEqual(tokens[3].line, 6)
self.assertEqual(tokens[3].col, 1)
tokens = Tokenizer().tokenize("SELECT .")
self.assertEqual(tokens[1].line, 1)
self.assertEqual(tokens[1].col, 8)
self.assertEqual(Tokenizer().tokenize("'''abc'")[0].start, 0)
self.assertEqual(Tokenizer().tokenize("'''abc'")[0].end, 6)
self.assertEqual(Tokenizer().tokenize("'abc'")[0].start, 0)
tokens = Tokenizer().tokenize("SELECT\r\n 1,\r\n 2")
self.assertEqual(tokens[0].line, 1)
self.assertEqual(tokens[0].col, 6)
self.assertEqual(tokens[1].line, 2)
self.assertEqual(tokens[1].col, 3)
self.assertEqual(tokens[2].line, 2)
self.assertEqual(tokens[2].col, 4)
self.assertEqual(tokens[3].line, 3)
self.assertEqual(tokens[3].col, 3)
tokens = Tokenizer().tokenize(" SELECT\n 100")
self.assertEqual(tokens[0].line, 1)
self.assertEqual(tokens[0].col, 8)
self.assertEqual(tokens[1].line, 2)
self.assertEqual(tokens[1].col, 7)
def test_crlf(self):
tokens = Tokenizer().tokenize("SELECT a\r\nFROM b")
tokens = [(token.token_type, token.text) for token in tokens]
self.assertEqual(
tokens,
[
(TokenType.SELECT, "SELECT"),
(TokenType.VAR, "a"),
(TokenType.FROM, "FROM"),
(TokenType.VAR, "b"),
],
)
for simple_query in ("SELECT 1\r\n", "\r\nSELECT 1"):
tokens = Tokenizer().tokenize(simple_query)
tokens = [(token.token_type, token.text) for token in tokens]
self.assertEqual(
tokens,
[
(TokenType.SELECT, "SELECT"),
(TokenType.NUMBER, "1"),
],
)
def test_command(self):
tokens = Tokenizer().tokenize("SHOW;")
self.assertEqual(tokens[0].token_type, TokenType.SHOW)
self.assertEqual(tokens[1].token_type, TokenType.SEMICOLON)
tokens = Tokenizer().tokenize("EXECUTE")
self.assertEqual(tokens[0].token_type, TokenType.EXECUTE)
self.assertEqual(len(tokens), 1)
tokens = Tokenizer().tokenize("FETCH;SHOW;")
self.assertEqual(tokens[0].token_type, TokenType.FETCH)
self.assertEqual(tokens[1].token_type, TokenType.SEMICOLON)
self.assertEqual(tokens[2].token_type, TokenType.SHOW)
self.assertEqual(tokens[3].token_type, TokenType.SEMICOLON)
def test_error_msg(self):
with self.assertRaisesRegex(TokenError, "Error tokenizing 'select /'"):
Tokenizer().tokenize("select /*")
def test_jinja(self):
# Check that {#, #} are treated as token delimiters, even though BigQuery overrides COMMENTS
tokenizer = BigQuery.Tokenizer()
tokens = tokenizer.tokenize(
"""
SELECT
{{ x }},
{{- x -}},
{# it's a comment #}
{% for x in y -%}
a {{+ b }}
{% endfor %};
"""
)
tokens = [(token.token_type, token.text) for token in tokens]
self.assertEqual(
tokens,
[
(TokenType.SELECT, "SELECT"),
(TokenType.L_BRACE, "{"),
(TokenType.L_BRACE, "{"),
(TokenType.VAR, "x"),
(TokenType.R_BRACE, "}"),
(TokenType.R_BRACE, "}"),
(TokenType.COMMA, ","),
(TokenType.BLOCK_START, "{{-"),
(TokenType.VAR, "x"),
(TokenType.BLOCK_END, "-}}"),
(TokenType.COMMA, ","),
(TokenType.BLOCK_START, "{%"),
(TokenType.FOR, "for"),
(TokenType.VAR, "x"),
(TokenType.IN, "in"),
(TokenType.VAR, "y"),
(TokenType.BLOCK_END, "-%}"),
(TokenType.VAR, "a"),
(TokenType.BLOCK_START, "{{+"),
(TokenType.VAR, "b"),
(TokenType.R_BRACE, "}"),
(TokenType.R_BRACE, "}"),
(TokenType.BLOCK_START, "{%"),
(TokenType.VAR, "endfor"),
(TokenType.BLOCK_END, "%}"),
(TokenType.SEMICOLON, ";"),
],
)
tokens = tokenizer.tokenize("""'{{ var('x') }}'""")
tokens = [(token.token_type, token.text) for token in tokens]
self.assertEqual(
tokens,
[
(TokenType.STRING, "{{ var("),
(TokenType.VAR, "x"),
(TokenType.STRING, ") }}"),
],
)
def test_partial_token_list(self):
tokenizer = Tokenizer()
try:
# This is expected to fail due to the unbalanced string quotes
tokenizer.tokenize("foo 'bar")
except TokenError as e:
self.assertIn("Error tokenizing 'foo 'ba'", str(e))
partial_tokens = tokenizer.tokens
self.assertEqual(len(partial_tokens), 1)
self.assertEqual(partial_tokens[0].token_type, TokenType.VAR)
self.assertEqual(partial_tokens[0].text, "foo")
def test_token_repr(self):
# Ensures both the Python and the Rust tokenizer produce a human-friendly representation
self.assertEqual(
repr(Tokenizer().tokenize("foo")),
"[<Token token_type: TokenType.VAR, text: foo, line: 1, col: 3, start: 0, end: 2, comments: []>]",
)
|
