Make test_tokenize pass again:
Add code to test_roundtrip() that figures out the encoding
from the first two lines of the file.
(We need to refactor this again to make it available to
all places that need this, e.g. linecache.py.)
diff --git a/Lib/test/test_tokenize.py b/Lib/test/test_tokenize.py
index 022b658..788a04b 100644
--- a/Lib/test/test_tokenize.py
+++ b/Lib/test/test_tokenize.py
@@ -80,7 +80,10 @@
"""
+# ' Emacs hint
+
import os, glob, random, time, sys
+import re
from io import StringIO
from test.test_support import (verbose, findfile, is_resource_enabled,
TestFailed)
@@ -96,7 +99,17 @@
# tokenization doesn't match the first.
def test_roundtrip(f):
## print 'Testing:', f
- fobj = open(f)
+ # Get the encoding first
+ fobj = open(f, encoding="latin-1")
+ first2lines = fobj.readline() + fobj.readline()
+ fobj.close()
+ m = re.search(r"coding:\s*(\S+)", first2lines)
+ if m:
+ encoding = m.group(1)
+ print(" coding:", encoding)
+ else:
+ encoding = "utf-8"
+ fobj = open(f, encoding=encoding)
try:
fulltok = list(generate_tokens(fobj.readline))
finally:
@@ -185,8 +198,6 @@
testdir = os.path.dirname(f) or os.curdir
testfiles = glob.glob(testdir + os.sep + 'test*.py')
- # Exclude test_pep263 which is encoded in KOI8-R
- testfiles = [t for t in testfiles if not t.endswith("pep263.py")]
if not is_resource_enabled('compiler'):
testfiles = random.sample(testfiles, 10)