diff options
| -rw-r--r-- | CHANGES.rst | 9 | ||||
| -rw-r--r-- | coverage/phystokens.py | 2 | ||||
| -rw-r--r-- | tests/test_phystokens.py | 18 |
3 files changed, 23 insertions, 6 deletions
diff --git a/CHANGES.rst b/CHANGES.rst index 082fda6d..a599352a 100644 --- a/CHANGES.rst +++ b/CHANGES.rst @@ -6,6 +6,15 @@ Change history for Coverage.py ============================== +Version 4.0.2, in progress +-------------------------- + +- Fixed an unusual edge case of detecting source encodings, described in + `issue 443`_. + +.. _issue 443: https://bitbucket.org/ned/coveragepy/issues/443/coverage-gets-confused-when-encoding + + Version 4.0.2 --- 4 November 2015 --------------------------------- diff --git a/coverage/phystokens.py b/coverage/phystokens.py index f5bd0bc9..b34b1c3b 100644 --- a/coverage/phystokens.py +++ b/coverage/phystokens.py @@ -153,7 +153,7 @@ class CachedTokenizer(object): generate_tokens = CachedTokenizer().generate_tokens -COOKIE_RE = re.compile(r"^\s*#.*coding[:=]\s*([-\w.]+)", flags=re.MULTILINE) +COOKIE_RE = re.compile(r"^[ \t]*#.*coding[:=][ \t]*([-\w.]+)", flags=re.MULTILINE) @contract(source='bytes') def _source_encoding_py2(source): diff --git a/tests/test_phystokens.py b/tests/test_phystokens.py index 7bdece72..e28fb176 100644 --- a/tests/test_phystokens.py +++ b/tests/test_phystokens.py @@ -102,6 +102,7 @@ ENCODING_DECLARATION_SOURCES = [ b"#!/usr/bin/python\n# vim: set fileencoding=cp850:\n", b"# This Python file uses this encoding: cp850\n", b"# This file uses a different encoding:\n# coding: cp850\n", + b"\n# coding=cp850\n\n", ] class SourceEncodingTest(CoverageTest): @@ -126,11 +127,6 @@ class SourceEncodingTest(CoverageTest): source = b'def parse(src, encoding=None):\n pass' self.assertEqual(source_encoding(source), DEF_ENCODING) - def test_detect_source_encoding_on_second_line(self): - # A coding declaration should be found despite a first blank line. - source = b"\n# coding=cp850\n\n" - self.assertEqual(source_encoding(source), 'cp850') - def test_dont_detect_source_encoding_on_third_line(self): # A coding declaration doesn't count on the third line. source = b"\n\n# coding=cp850\n\n" @@ -160,6 +156,18 @@ class NeuterEncodingDeclarationTest(CoverageTest): for source in ENCODING_DECLARATION_SOURCES: neutered = neuter_encoding_declaration(source.decode("ascii")) neutered = neutered.encode("ascii") + + # The neutered source should have the same number of lines. + source_lines = source.splitlines() + neutered_lines = neutered.splitlines() + self.assertEqual(len(source_lines), len(neutered_lines)) + + # Only one of the lines should be different. + lines_different = sum( + int(nline != sline) for nline, sline in zip(neutered_lines, source_lines) + ) + self.assertEqual(lines_different, 1) + self.assertEqual( source_encoding(neutered), DEF_ENCODING, |