- Issue #719888: Updated tokenize to use a bytes API. generate_tokens has been
renamed tokenize and now works with bytes rather than strings. A new
detect_encoding function has been added for determining source file encoding
according to PEP-0263. Token sequences returned by tokenize always start
with an ENCODING token which specifies the encoding used to decode the file.
This token is used to encode the output of untokenize back to bytes.
Credit goes to Michael "I'm-going-to-name-my-first-child-unittest" Foord from Resolver Systems for this work.
diff --git a/Lib/idlelib/EditorWindow.py b/Lib/idlelib/EditorWindow.py
index e1d9ba3..2b5c527 100644
--- a/Lib/idlelib/EditorWindow.py
+++ b/Lib/idlelib/EditorWindow.py
@@ -1437,7 +1437,9 @@
_tokenize.tabsize = self.tabwidth
try:
try:
- _tokenize.tokenize(self.readline, self.tokeneater)
+ tokens = _tokenize.generate_tokens(self.readline)
+ for token in tokens:
+ self.tokeneater(*token)
except _tokenize.TokenError:
# since we cut off the tokenizer early, we can trigger
# spurious errors
diff --git a/Lib/inspect.py b/Lib/inspect.py
index ceaea5a..5cb958b 100644
--- a/Lib/inspect.py
+++ b/Lib/inspect.py
@@ -657,7 +657,9 @@
"""Extract the block of code at the top of the given list of lines."""
blockfinder = BlockFinder()
try:
- tokenize.tokenize(iter(lines).__next__, blockfinder.tokeneater)
+ tokens = tokenize.generate_tokens(iter(lines).__next__)
+ for _token in tokens:
+ blockfinder.tokeneater(*_token)
except (EndOfBlock, IndentationError):
pass
return lines[:blockfinder.last]
diff --git a/Lib/test/test_tokenize.py b/Lib/test/test_tokenize.py
index 2938520..308158f 100644
--- a/Lib/test/test_tokenize.py
+++ b/Lib/test/test_tokenize.py
@@ -1,13 +1,14 @@
+# -*- coding: utf-8 -*-
+
doctests = """
Tests for the tokenize module.
- >>> import glob, random, sys
-
The tests can be really simple. Given a small fragment of source
code, print out a table with tokens. The ENDMARK is omitted for
brevity.
>>> dump_tokens("1 + 1")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '1' (1, 0) (1, 1)
OP '+' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
@@ -15,6 +16,7 @@
>>> dump_tokens("if False:\\n"
... " # NL\\n"
... " True = False # NEWLINE\\n")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'if' (1, 0) (1, 2)
NAME 'False' (1, 3) (1, 8)
OP ':' (1, 8) (1, 9)
@@ -34,27 +36,12 @@
... x += 2
... x += 5
... \"""
- >>> for tok in generate_tokens(StringIO(indent_error_file).readline): pass
+ >>> readline = BytesIO(indent_error_file.encode('utf-8')).readline
+ >>> for tok in tokenize(readline): pass
Traceback (most recent call last):
...
IndentationError: unindent does not match any outer indentation level
-Test roundtrip for `untokenize`. `f` is an open file or a string. The source
-code in f is tokenized, converted back to source code via tokenize.untokenize(),
-and tokenized again from the latter. The test fails if the second tokenization
-doesn't match the first.
-
- >>> def roundtrip(f):
- ... if isinstance(f, str): f = StringIO(f)
- ... token_list = list(generate_tokens(f.readline))
- ... f.close()
- ... tokens1 = [tok[:2] for tok in token_list]
- ... new_text = untokenize(tokens1)
- ... readline = iter(new_text.splitlines(1)).__next__
- ... tokens2 = [tok[:2] for tok in generate_tokens(readline)]
- ... return tokens1 == tokens2
- ...
-
There are some standard formattig practises that are easy to get right.
>>> roundtrip("if x == 1:\\n"
@@ -67,14 +54,14 @@
Some people use different formatting conventions, which makes
untokenize a little trickier. Note that this test involves trailing
whitespace after the colon. Note that we use hex escapes to make the
-two trailing blanks apperant in the expected output.
+two trailing blanks apparent in the expected output.
>>> roundtrip("if x == 1 : \\n"
... " print(x)\\n")
True
>>> f = test_support.findfile("tokenize_tests.txt")
- >>> roundtrip(open(f))
+ >>> roundtrip(open(f, 'rb'))
True
>>> roundtrip("if x == 1:\\n"
@@ -122,27 +109,33 @@
Ordinary integers and binary operators
>>> dump_tokens("0xff <= 255")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0xff' (1, 0) (1, 4)
OP '<=' (1, 5) (1, 7)
NUMBER '255' (1, 8) (1, 11)
>>> dump_tokens("0b10 <= 255")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0b10' (1, 0) (1, 4)
OP '<=' (1, 5) (1, 7)
NUMBER '255' (1, 8) (1, 11)
>>> dump_tokens("0o123 <= 0O123")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0o123' (1, 0) (1, 5)
OP '<=' (1, 6) (1, 8)
NUMBER '0O123' (1, 9) (1, 14)
>>> dump_tokens("1234567 > ~0x15")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '1234567' (1, 0) (1, 7)
OP '>' (1, 8) (1, 9)
OP '~' (1, 10) (1, 11)
NUMBER '0x15' (1, 11) (1, 15)
>>> dump_tokens("2134568 != 1231515")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '2134568' (1, 0) (1, 7)
OP '!=' (1, 8) (1, 10)
NUMBER '1231515' (1, 11) (1, 18)
>>> dump_tokens("(-124561-1) & 200000000")
+ ENCODING 'utf-8' (0, 0) (0, 0)
OP '(' (1, 0) (1, 1)
OP '-' (1, 1) (1, 2)
NUMBER '124561' (1, 2) (1, 8)
@@ -152,15 +145,18 @@
OP '&' (1, 12) (1, 13)
NUMBER '200000000' (1, 14) (1, 23)
>>> dump_tokens("0xdeadbeef != -1")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0xdeadbeef' (1, 0) (1, 10)
OP '!=' (1, 11) (1, 13)
OP '-' (1, 14) (1, 15)
NUMBER '1' (1, 15) (1, 16)
>>> dump_tokens("0xdeadc0de & 12345")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0xdeadc0de' (1, 0) (1, 10)
OP '&' (1, 11) (1, 12)
NUMBER '12345' (1, 13) (1, 18)
>>> dump_tokens("0xFF & 0x15 | 1234")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NUMBER '0xFF' (1, 0) (1, 4)
OP '&' (1, 5) (1, 6)
NUMBER '0x15' (1, 7) (1, 11)
@@ -170,18 +166,22 @@
Long integers
>>> dump_tokens("x = 0")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '0' (1, 4) (1, 5)
>>> dump_tokens("x = 0xfffffffffff")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '0xffffffffff (1, 4) (1, 17)
>>> dump_tokens("x = 123141242151251616110")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '123141242151 (1, 4) (1, 25)
>>> dump_tokens("x = -15921590215012591")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
OP '-' (1, 4) (1, 5)
@@ -190,32 +190,39 @@
Floating point numbers
>>> dump_tokens("x = 3.14159")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3.14159' (1, 4) (1, 11)
>>> dump_tokens("x = 314159.")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '314159.' (1, 4) (1, 11)
>>> dump_tokens("x = .314159")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '.314159' (1, 4) (1, 11)
>>> dump_tokens("x = 3e14159")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3e14159' (1, 4) (1, 11)
>>> dump_tokens("x = 3E123")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3E123' (1, 4) (1, 9)
>>> dump_tokens("x+y = 3e-1230")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '+' (1, 1) (1, 2)
NAME 'y' (1, 2) (1, 3)
OP '=' (1, 4) (1, 5)
NUMBER '3e-1230' (1, 6) (1, 13)
>>> dump_tokens("x = 3.14e159")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '3.14e159' (1, 4) (1, 12)
@@ -223,6 +230,7 @@
String literals
>>> dump_tokens("x = ''; y = \\\"\\\"")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "''" (1, 4) (1, 6)
@@ -231,6 +239,7 @@
OP '=' (1, 10) (1, 11)
STRING '""' (1, 12) (1, 14)
>>> dump_tokens("x = '\\\"'; y = \\\"'\\\"")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '\\'"\\'' (1, 4) (1, 7)
@@ -239,24 +248,28 @@
OP '=' (1, 11) (1, 12)
STRING '"\\'"' (1, 13) (1, 16)
>>> dump_tokens("x = \\\"doesn't \\\"shrink\\\", does it\\\"")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '"doesn\\'t "' (1, 4) (1, 14)
NAME 'shrink' (1, 14) (1, 20)
STRING '", does it"' (1, 20) (1, 31)
>>> dump_tokens("x = 'abc' + 'ABC'")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "'abc'" (1, 4) (1, 9)
OP '+' (1, 10) (1, 11)
STRING "'ABC'" (1, 12) (1, 17)
>>> dump_tokens('y = "ABC" + "ABC"')
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'y' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING '"ABC"' (1, 4) (1, 9)
OP '+' (1, 10) (1, 11)
STRING '"ABC"' (1, 12) (1, 17)
>>> dump_tokens("x = r'abc' + r'ABC' + R'ABC' + R'ABC'")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING "r'abc'" (1, 4) (1, 10)
@@ -267,6 +280,7 @@
OP '+' (1, 29) (1, 30)
STRING "R'ABC'" (1, 31) (1, 37)
>>> dump_tokens('y = r"abc" + r"ABC" + R"ABC" + R"ABC"')
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'y' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
STRING 'r"abc"' (1, 4) (1, 10)
@@ -280,6 +294,7 @@
Operators
>>> dump_tokens("def d22(a, b, c=2, d=2, *k): pass")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'def' (1, 0) (1, 3)
NAME 'd22' (1, 4) (1, 7)
OP '(' (1, 7) (1, 8)
@@ -301,6 +316,7 @@
OP ':' (1, 27) (1, 28)
NAME 'pass' (1, 29) (1, 33)
>>> dump_tokens("def d01v_(a=1, *k, **w): pass")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'def' (1, 0) (1, 3)
NAME 'd01v_' (1, 4) (1, 9)
OP '(' (1, 9) (1, 10)
@@ -321,6 +337,7 @@
>>> dump_tokens("if 1 < 1 > 1 == 1 >= 5 <= 0x15 <= 0x12 != " +
... "1 and 5 in 1 not in 1 is 1 or 5 is not 1: pass")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'if' (1, 0) (1, 2)
NUMBER '1' (1, 3) (1, 4)
OP '<' (1, 5) (1, 6)
@@ -357,6 +374,7 @@
Shift
>>> dump_tokens("x = 1 << 1 >> 5")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
@@ -368,6 +386,7 @@
Additive
>>> dump_tokens("x = 1 - y + 15 - 1 + 0x124 + z + a[5]")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
@@ -390,6 +409,7 @@
Multiplicative
>>> dump_tokens("x = 1//1*1/5*12%0x12")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'x' (1, 0) (1, 1)
OP '=' (1, 2) (1, 3)
NUMBER '1' (1, 4) (1, 5)
@@ -407,6 +427,7 @@
Unary
>>> dump_tokens("~1 ^ 1 & 1 |1 ^ -1")
+ ENCODING 'utf-8' (0, 0) (0, 0)
OP '~' (1, 0) (1, 1)
NUMBER '1' (1, 1) (1, 2)
OP '^' (1, 3) (1, 4)
@@ -419,6 +440,7 @@
OP '-' (1, 16) (1, 17)
NUMBER '1' (1, 17) (1, 18)
>>> dump_tokens("-1*1/1+1*1//1 - ---1**1")
+ ENCODING 'utf-8' (0, 0) (0, 0)
OP '-' (1, 0) (1, 1)
NUMBER '1' (1, 1) (1, 2)
OP '*' (1, 2) (1, 3)
@@ -442,6 +464,7 @@
Selector
>>> dump_tokens("import sys, time\\nx = sys.modules['time'].time()")
+ ENCODING 'utf-8' (0, 0) (0, 0)
NAME 'import' (1, 0) (1, 6)
NAME 'sys' (1, 7) (1, 10)
OP ',' (1, 10) (1, 11)
@@ -463,6 +486,7 @@
Methods
>>> dump_tokens("@staticmethod\\ndef foo(x,y): pass")
+ ENCODING 'utf-8' (0, 0) (0, 0)
OP '@' (1, 0) (1, 1)
NAME 'staticmethod (1, 1) (1, 13)
NEWLINE '\\n' (1, 13) (1, 14)
@@ -485,42 +509,43 @@
True
>>> roundtrip("# Comment \\\\nx = 0")
True
-
- >>>
- >>> tempdir = os.path.dirname(f) or os.curdir
- >>> testfiles = glob.glob(os.path.join(tempdir, "test*.py"))
- >>> if not test_support.is_resource_enabled("compiler"):
- ... testfiles = random.sample(testfiles, 10)
- ...
- >>> for testfile in testfiles:
- ... if not roundtrip(open(testfile)): break
- ... else: True
- True
"""
-
from test import test_support
-from tokenize import (tokenize, untokenize, generate_tokens, NUMBER, NAME, OP,
- STRING, ENDMARKER, tok_name)
-from io import StringIO
-import os
+from tokenize import (tokenize, _tokenize, untokenize, NUMBER, NAME, OP,
+ STRING, ENDMARKER, tok_name, detect_encoding)
+from io import BytesIO
+from unittest import TestCase
+import os, sys, glob
def dump_tokens(s):
"""Print out the tokens in s in a table format.
The ENDMARKER is omitted.
"""
- f = StringIO(s)
- for type, token, start, end, line in generate_tokens(f.readline):
+ f = BytesIO(s.encode('utf-8'))
+ for type, token, start, end, line in tokenize(f.readline):
if type == ENDMARKER:
break
type = tok_name[type]
print("%(type)-10.10s %(token)-13.13r %(start)s %(end)s" % locals())
-def roundtrip(s):
- f = StringIO(s)
- source = untokenize(generate_tokens(f.readline))
- print(source, end="")
+def roundtrip(f):
+ """
+ Test roundtrip for `untokenize`. `f` is an open file or a string.
+ The source code in f is tokenized, converted back to source code via
+ tokenize.untokenize(), and tokenized again from the latter. The test
+ fails if the second tokenization doesn't match the first.
+ """
+ if isinstance(f, str):
+ f = BytesIO(f.encode('utf-8'))
+ token_list = list(tokenize(f.readline))
+ f.close()
+ tokens1 = [tok[:2] for tok in token_list]
+ new_bytes = untokenize(tokens1)
+ readline = (line for line in new_bytes.splitlines(1)).__next__
+ tokens2 = [tok[:2] for tok in tokenize(readline)]
+ return tokens1 == tokens2
# This is an example from the docs, set up as a doctest.
def decistmt(s):
@@ -545,9 +570,8 @@
>>> exec(decistmt(s))
-3.217160342717258261933904529E-7
"""
-
result = []
- g = generate_tokens(StringIO(s).readline) # tokenize the string
+ g = tokenize(BytesIO(s.encode('utf-8')).readline) # tokenize the string
for toknum, tokval, _, _, _ in g:
if toknum == NUMBER and '.' in tokval: # replace NUMBER tokens
result.extend([
@@ -558,7 +582,238 @@
])
else:
result.append((toknum, tokval))
- return untokenize(result)
+ return untokenize(result).decode('utf-8')
+
+
+class TestTokenizerAdheresToPep0263(TestCase):
+ """
+ Test that tokenizer adheres to the coding behaviour stipulated in PEP 0263.
+ """
+
+ def _testFile(self, filename):
+ path = os.path.join(os.path.dirname(__file__), filename)
+ return roundtrip(open(path, 'rb'))
+
+ def test_utf8_coding_cookie_and_no_utf8_bom(self):
+ f = 'tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt'
+ self.assertTrue(self._testFile(f))
+
+ def test_latin1_coding_cookie_and_utf8_bom(self):
+ """
+ As per PEP 0263, if a file starts with a utf-8 BOM signature, the only
+ allowed encoding for the comment is 'utf-8'. The text file used in
+ this test starts with a BOM signature, but specifies latin1 as the
+ coding, so verify that a SyntaxError is raised, which matches the
+ behaviour of the interpreter when it encounters a similar condition.
+ """
+ f = 'tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt'
+ self.failUnlessRaises(SyntaxError, self._testFile, f)
+
+ def test_no_coding_cookie_and_utf8_bom(self):
+ f = 'tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt'
+ self.assertTrue(self._testFile(f))
+
+ def test_utf8_coding_cookie_and_utf8_bom(self):
+ f = 'tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt'
+ self.assertTrue(self._testFile(f))
+
+
+class Test_Tokenize(TestCase):
+
+ def test__tokenize_decodes_with_specified_encoding(self):
+ literal = '"ЉЊЈЁЂ"'
+ line = literal.encode('utf-8')
+ first = False
+ def readline():
+ nonlocal first
+ if not first:
+ first = True
+ return line
+ else:
+ return b''
+
+ # skip the initial encoding token and the end token
+ tokens = list(_tokenize(readline, encoding='utf-8'))[1:-1]
+ expected_tokens = [(3, '"ЉЊЈЁЂ"', (1, 0), (1, 7), '"ЉЊЈЁЂ"')]
+ self.assertEquals(tokens, expected_tokens,
+ "bytes not decoded with encoding")
+
+ def test__tokenize_does_not_decode_with_encoding_none(self):
+ literal = '"ЉЊЈЁЂ"'
+ first = False
+ def readline():
+ nonlocal first
+ if not first:
+ first = True
+ return literal
+ else:
+ return b''
+
+ # skip the end token
+ tokens = list(_tokenize(readline, encoding=None))[:-1]
+ expected_tokens = [(3, '"ЉЊЈЁЂ"', (1, 0), (1, 7), '"ЉЊЈЁЂ"')]
+ self.assertEquals(tokens, expected_tokens,
+ "string not tokenized when encoding is None")
+
+
+class TestDetectEncoding(TestCase):
+
+ def get_readline(self, lines):
+ index = 0
+ def readline():
+ nonlocal index
+ if index == len(lines):
+ raise StopIteration
+ line = lines[index]
+ index += 1
+ return line
+ return readline
+
+ def test_no_bom_no_encoding_cookie(self):
+ lines = (
+ b'# something\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, list(lines[:2]))
+
+ def test_bom_no_cookie(self):
+ lines = (
+ b'\xef\xbb\xbf# something\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines,
+ [b'# something\n', b'print(something)\n'])
+
+ def test_cookie_first_line_no_bom(self):
+ lines = (
+ b'# -*- coding: latin-1 -*-\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'latin-1')
+ self.assertEquals(consumed_lines, [b'# -*- coding: latin-1 -*-\n'])
+
+ def test_matched_bom_and_cookie_first_line(self):
+ lines = (
+ b'\xef\xbb\xbf# coding=utf-8\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, [b'# coding=utf-8\n'])
+
+ def test_mismatched_bom_and_cookie_first_line_raises_syntaxerror(self):
+ lines = (
+ b'\xef\xbb\xbf# vim: set fileencoding=ascii :\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ readline = self.get_readline(lines)
+ self.assertRaises(SyntaxError, detect_encoding, readline)
+
+ def test_cookie_second_line_no_bom(self):
+ lines = (
+ b'#! something\n',
+ b'# vim: set fileencoding=ascii :\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'ascii')
+ expected = [b'#! something\n', b'# vim: set fileencoding=ascii :\n']
+ self.assertEquals(consumed_lines, expected)
+
+ def test_matched_bom_and_cookie_second_line(self):
+ lines = (
+ b'\xef\xbb\xbf#! something\n',
+ b'f# coding=utf-8\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ encoding, consumed_lines = detect_encoding(self.get_readline(lines))
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines,
+ [b'#! something\n', b'f# coding=utf-8\n'])
+
+ def test_mismatched_bom_and_cookie_second_line_raises_syntaxerror(self):
+ lines = (
+ b'\xef\xbb\xbf#! something\n',
+ b'# vim: set fileencoding=ascii :\n',
+ b'print(something)\n',
+ b'do_something(else)\n'
+ )
+ readline = self.get_readline(lines)
+ self.assertRaises(SyntaxError, detect_encoding, readline)
+
+ def test_short_files(self):
+ readline = self.get_readline((b'print(something)\n',))
+ encoding, consumed_lines = detect_encoding(readline)
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, [b'print(something)\n'])
+
+ encoding, consumed_lines = detect_encoding(self.get_readline(()))
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, [])
+
+ readline = self.get_readline((b'\xef\xbb\xbfprint(something)\n',))
+ encoding, consumed_lines = detect_encoding(readline)
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, [b'print(something)\n'])
+
+ readline = self.get_readline((b'\xef\xbb\xbf',))
+ encoding, consumed_lines = detect_encoding(readline)
+ self.assertEquals(encoding, 'utf-8')
+ self.assertEquals(consumed_lines, [])
+
+
+class TestTokenize(TestCase):
+
+ def test_tokenize(self):
+ import tokenize as tokenize_module
+ encoding = object()
+ encoding_used = None
+ def mock_detect_encoding(readline):
+ return encoding, ['first', 'second']
+
+ def mock__tokenize(readline, encoding):
+ nonlocal encoding_used
+ encoding_used = encoding
+ out = []
+ while True:
+ next_line = readline()
+ if next_line:
+ out.append(next_line)
+ continue
+ return out
+
+ counter = 0
+ def mock_readline():
+ nonlocal counter
+ counter += 1
+ if counter == 5:
+ return b''
+ return counter
+
+ orig_detect_encoding = tokenize_module.detect_encoding
+ orig__tokenize = tokenize_module._tokenize
+ tokenize_module.detect_encoding = mock_detect_encoding
+ tokenize_module._tokenize = mock__tokenize
+ try:
+ results = tokenize(mock_readline)
+ self.assertEquals(list(results), ['first', 'second', 1, 2, 3, 4])
+ finally:
+ tokenize_module.detect_encoding = orig_detect_encoding
+ tokenize_module._tokenize = orig__tokenize
+
+ self.assertTrue(encoding_used, encoding)
__test__ = {"doctests" : doctests, 'decistmt': decistmt}
@@ -566,6 +821,10 @@
def test_main():
from test import test_tokenize
test_support.run_doctest(test_tokenize, True)
+ test_support.run_unittest(TestTokenizerAdheresToPep0263)
+ test_support.run_unittest(Test_Tokenize)
+ test_support.run_unittest(TestDetectEncoding)
+ test_support.run_unittest(TestTokenize)
if __name__ == "__main__":
test_main()
diff --git a/Lib/test/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt b/Lib/test/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt
new file mode 100644
index 0000000..4a7582a
--- /dev/null
+++ b/Lib/test/tokenize_tests-latin1-coding-cookie-and-utf8-bom-sig.txt
@@ -0,0 +1,13 @@
+# -*- coding: latin1 -*-
+# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf'
+# at the start of it. Make sure this is preserved if any changes
+# are made! Also note that the coding cookie above conflicts with
+# the presense of a utf-8 BOM signature -- this is intended.
+
+# Arbitrary encoded utf-8 text (stolen from test_doctest2.py).
+x = 'ЉЊЈЁЂ'
+def y():
+ """
+ And again in a comment. ЉЊЈЁЂ
+ """
+ pass
diff --git a/Lib/test/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt b/Lib/test/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt
new file mode 100644
index 0000000..d6d8d49
--- /dev/null
+++ b/Lib/test/tokenize_tests-no-coding-cookie-and-utf8-bom-sig-only.txt
@@ -0,0 +1,11 @@
+# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf'
+# at the start of it. Make sure this is preserved if any changes
+# are made!
+
+# Arbitrary encoded utf-8 text (stolen from test_doctest2.py).
+x = 'ЉЊЈЁЂ'
+def y():
+ """
+ And again in a comment. ЉЊЈЁЂ
+ """
+ pass
diff --git a/Lib/test/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt b/Lib/test/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt
new file mode 100644
index 0000000..7aeff6b
--- /dev/null
+++ b/Lib/test/tokenize_tests-utf8-coding-cookie-and-no-utf8-bom-sig.txt
@@ -0,0 +1,13 @@
+# -*- coding: utf-8 -*-
+# IMPORTANT: unlike the other test_tokenize-*.txt files, this file
+# does NOT have the utf-8 BOM signature '\xef\xbb\xbf' at the start
+# of it. Make sure this is not added inadvertently by your editor
+# if any changes are made to this file!
+
+# Arbitrary encoded utf-8 text (stolen from test_doctest2.py).
+x = 'ЉЊЈЁЂ'
+def y():
+ """
+ And again in a comment. ЉЊЈЁЂ
+ """
+ pass
diff --git a/Lib/test/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt b/Lib/test/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt
new file mode 100644
index 0000000..99b1399
--- /dev/null
+++ b/Lib/test/tokenize_tests-utf8-coding-cookie-and-utf8-bom-sig.txt
@@ -0,0 +1,12 @@
+# -*- coding: utf-8 -*-
+# IMPORTANT: this file has the utf-8 BOM signature '\xef\xbb\xbf'
+# at the start of it. Make sure this is preserved if any changes
+# are made!
+
+# Arbitrary encoded utf-8 text (stolen from test_doctest2.py).
+x = 'ЉЊЈЁЂ'
+def y():
+ """
+ And again in a comment. ЉЊЈЁЂ
+ """
+ pass
diff --git a/Lib/tokenize.py b/Lib/tokenize.py
index 797b6e0..d7043b1 100644
--- a/Lib/tokenize.py
+++ b/Lib/tokenize.py
@@ -1,8 +1,11 @@
"""Tokenization help for Python programs.
-generate_tokens(readline) is a generator that breaks a stream of
-text into Python tokens. It accepts a readline-like method which is called
-repeatedly to get the next line of input (or "" for EOF). It generates
+tokenize(readline) is a generator that breaks a stream of
+bytes into Python tokens. It decodes the bytes according to
+PEP-0263 for determining source file encoding.
+
+It accepts a readline-like method which is called
+repeatedly to get the next line of input (or b"" for EOF). It generates
5-tuples with these members:
the token type (see token.py)
@@ -13,32 +16,32 @@
It is designed to match the working of the Python tokenizer exactly, except
that it produces COMMENT tokens for comments and gives type OP for all
-operators
-
-Older entry points
- tokenize_loop(readline, tokeneater)
- tokenize(readline, tokeneater=printtoken)
-are the same, except instead of generating tokens, tokeneater is a callback
-function to which the 5 fields described above are passed as 5 arguments,
-each time a new token is found."""
+operators. Aditionally, all token lists start with an ENCODING token
+which tells you which encoding was used to decode the bytes stream."""
__author__ = 'Ka-Ping Yee <ping@lfw.org>'
-__credits__ = \
- 'GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, Skip Montanaro, Raymond Hettinger'
+__credits__ = ('GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, '
+ 'Skip Montanaro, Raymond Hettinger, Trent Nelson, '
+ 'Michael Foord')
-import string, re
+import re, string, sys
from token import *
+from codecs import lookup
+from itertools import chain, repeat
+cookie_re = re.compile("coding[:=]\s*([-\w.]+)")
import token
__all__ = [x for x in dir(token) if x[0] != '_'] + ["COMMENT", "tokenize",
- "generate_tokens", "NL", "untokenize"]
+ "detect_encoding", "NL", "untokenize", "ENCODING"]
del token
COMMENT = N_TOKENS
tok_name[COMMENT] = 'COMMENT'
NL = N_TOKENS + 1
tok_name[NL] = 'NL'
-N_TOKENS += 2
+ENCODING = N_TOKENS + 2
+tok_name[ENCODING] = 'ENCODING'
+N_TOKENS += 3
def group(*choices): return '(' + '|'.join(choices) + ')'
def any(*choices): return group(*choices) + '*'
@@ -132,33 +135,6 @@
class StopTokenizing(Exception): pass
-def printtoken(type, token, startrowcol, endrowcol, line): # for testing
- (srow, scol), (erow, ecol) = startrowcol, endrowcol
- print("%d,%d-%d,%d:\t%s\t%s" % \
- (srow, scol, erow, ecol, tok_name[type], repr(token)))
-
-def tokenize(readline, tokeneater=printtoken):
- """
- The tokenize() function accepts two parameters: one representing the
- input stream, and one providing an output mechanism for tokenize().
-
- The first parameter, readline, must be a callable object which provides
- the same interface as the readline() method of built-in file objects.
- Each call to the function should return one line of input as a string.
-
- The second parameter, tokeneater, must also be a callable object. It is
- called once for each token, with five arguments, corresponding to the
- tuples generated by generate_tokens().
- """
- try:
- tokenize_loop(readline, tokeneater)
- except StopTokenizing:
- pass
-
-# backwards compatible interface
-def tokenize_loop(readline, tokeneater):
- for token_info in generate_tokens(readline):
- tokeneater(*token_info)
class Untokenizer:
@@ -166,6 +142,7 @@
self.tokens = []
self.prev_row = 1
self.prev_col = 0
+ self.encoding = None
def add_whitespace(self, start):
row, col = start
@@ -180,6 +157,9 @@
self.compat(t, iterable)
break
tok_type, token, start, end, line = t
+ if tok_type == ENCODING:
+ self.encoding = token
+ continue
self.add_whitespace(start)
self.tokens.append(token)
self.prev_row, self.prev_col = end
@@ -193,12 +173,16 @@
indents = []
toks_append = self.tokens.append
toknum, tokval = token
+
if toknum in (NAME, NUMBER):
tokval += ' '
if toknum in (NEWLINE, NL):
startline = True
for tok in iterable:
toknum, tokval = tok[:2]
+ if toknum == ENCODING:
+ self.encoding = tokval
+ continue
if toknum in (NAME, NUMBER):
tokval += ' '
@@ -216,8 +200,11 @@
startline = False
toks_append(tokval)
+
def untokenize(iterable):
"""Transform tokens back into Python source code.
+ It returns a bytes object, encoded using the ENCODING
+ token, which is the first token sequence output by tokenize.
Each element returned by the iterable must be a token sequence
with at least two elements, a token number and token value. If
@@ -227,24 +214,89 @@
Untokenized source will match input source exactly
Round-trip invariant for limited intput:
- # Output text will tokenize the back to the input
- t1 = [tok[:2] for tok in generate_tokens(f.readline)]
+ # Output bytes will tokenize the back to the input
+ t1 = [tok[:2] for tok in tokenize(f.readline)]
newcode = untokenize(t1)
- readline = iter(newcode.splitlines(1)).__next__
- t2 = [tok[:2] for tokin generate_tokens(readline)]
+ readline = BytesIO(newcode).readline
+ t2 = [tok[:2] for tok in tokenize(readline)]
assert t1 == t2
"""
ut = Untokenizer()
- return ut.untokenize(iterable)
+ out = ut.untokenize(iterable)
+ if ut.encoding is not None:
+ out = out.encode(ut.encoding)
+ return out
-def generate_tokens(readline):
+
+def detect_encoding(readline):
"""
- The generate_tokens() generator requires one argment, readline, which
+ The detect_encoding() function is used to detect the encoding that should
+ be used to decode a Python source file. It requires one argment, readline,
+ in the same way as the tokenize() generator.
+
+ It will call readline a maximum of twice, and return the encoding used
+ (as a string) and a list of any lines (left as bytes) it has read
+ in.
+
+ It detects the encoding from the presence of a utf-8 bom or an encoding
+ cookie as specified in pep-0263. If both a bom and a cookie are present,
+ but disagree, a SyntaxError will be raised.
+
+ If no encoding is specified, then the default of 'utf-8' will be returned.
+ """
+ utf8_bom = b'\xef\xbb\xbf'
+ bom_found = False
+ encoding = None
+ def read_or_stop():
+ try:
+ return readline()
+ except StopIteration:
+ return b''
+
+ def find_cookie(line):
+ try:
+ line_string = line.decode('ascii')
+ except UnicodeDecodeError:
+ pass
+ else:
+ matches = cookie_re.findall(line_string)
+ if matches:
+ encoding = matches[0]
+ if bom_found and lookup(encoding).name != 'utf-8':
+ # This behaviour mimics the Python interpreter
+ raise SyntaxError('encoding problem: utf-8')
+ return encoding
+
+ first = read_or_stop()
+ if first.startswith(utf8_bom):
+ bom_found = True
+ first = first[3:]
+ if not first:
+ return 'utf-8', []
+
+ encoding = find_cookie(first)
+ if encoding:
+ return encoding, [first]
+
+ second = read_or_stop()
+ if not second:
+ return 'utf-8', [first]
+
+ encoding = find_cookie(second)
+ if encoding:
+ return encoding, [first, second]
+
+ return 'utf-8', [first, second]
+
+
+def tokenize(readline):
+ """
+ The tokenize() generator requires one argment, readline, which
must be a callable object which provides the same interface as the
readline() method of built-in file objects. Each call to the function
- should return one line of input as a string. Alternately, readline
+ should return one line of input as bytes. Alternately, readline
can be a callable function terminating with StopIteration:
- readline = open(myfile).__next__ # Example of alternate readline
+ readline = open(myfile, 'rb').__next__ # Example of alternate readline
The generator produces 5-tuples with these members: the token type; the
token string; a 2-tuple (srow, scol) of ints specifying the row and
@@ -252,18 +304,38 @@
ints specifying the row and column where the token ends in the source;
and the line on which the token was found. The line passed is the
logical line; continuation lines are included.
+
+ The first token sequence will always be an ENCODING token
+ which tells you which encoding was used to decode the bytes stream.
"""
+ encoding, consumed = detect_encoding(readline)
+ def readline_generator():
+ while True:
+ try:
+ yield readline()
+ except StopIteration:
+ return
+ chained = chain(consumed, readline_generator())
+ return _tokenize(chained.__next__, encoding)
+
+
+def _tokenize(readline, encoding):
lnum = parenlev = continued = 0
namechars, numchars = string.ascii_letters + '_', '0123456789'
contstr, needcont = '', 0
contline = None
indents = [0]
+ if encoding is not None:
+ yield (ENCODING, encoding, (0, 0), (0, 0), '')
while 1: # loop over lines in stream
try:
line = readline()
except StopIteration:
- line = ''
+ line = b''
+
+ if encoding is not None:
+ line = line.decode(encoding)
lnum = lnum + 1
pos, max = 0, len(line)
@@ -385,7 +457,8 @@
yield (DEDENT, '', (lnum, 0), (lnum, 0), '')
yield (ENDMARKER, '', (lnum, 0), (lnum, 0), '')
-if __name__ == '__main__': # testing
- import sys
- if len(sys.argv) > 1: tokenize(open(sys.argv[1]).readline)
- else: tokenize(sys.stdin.readline)
+
+# An undocumented, backwards compatible, API for all the places in the standard
+# library that expect to be able to use tokenize with strings
+def generate_tokens(readline):
+ return _tokenize(readline, None)