Issue #12451: pydoc.synopsis() now reads the encoding cookie if available, to
read the Python script from the right encoding.
diff --git a/Lib/pydoc.py b/Lib/pydoc.py
index 2391013e..1619446 100755
--- a/Lib/pydoc.py
+++ b/Lib/pydoc.py
@@ -57,16 +57,17 @@
# the current directory is changed with os.chdir(), an incorrect
# path will be displayed.
-import os
-import sys
import builtins
import imp
-import io
import inspect
+import io
+import os
import pkgutil
import platform
import re
+import sys
import time
+import tokenize
import warnings
from collections import deque
from reprlib import Repr
@@ -227,7 +228,7 @@
if lastupdate < mtime:
info = inspect.getmoduleinfo(filename)
try:
- file = open(filename)
+ file = tokenize.open(filename)
except IOError:
# module can't be opened, so skip it
return None
diff --git a/Lib/test/test_pydoc.py b/Lib/test/test_pydoc.py
index 08ba86e..a8f9fbfc 100644
--- a/Lib/test/test_pydoc.py
+++ b/Lib/test/test_pydoc.py
@@ -16,7 +16,7 @@
from collections import namedtuple
from contextlib import contextmanager
from test.support import TESTFN, forget, rmtree, EnvironmentVarGuard, \
- reap_children, captured_output, captured_stdout
+ reap_children, captured_output, captured_stdout, unlink
from test import pydoc_mod
@@ -389,6 +389,17 @@
self.assertIn('_replace', helptext)
self.assertIn('_asdict', helptext)
+ def test_synopsis(self):
+ self.addCleanup(unlink, TESTFN)
+ for encoding in ('ISO-8859-1', 'UTF-8'):
+ with open(TESTFN, 'w', encoding=encoding) as script:
+ if encoding != 'UTF-8':
+ print('#coding: {}'.format(encoding), file=script)
+ print('"""line 1: h\xe9', file=script)
+ print('line 2: hi"""', file=script)
+ synopsis = pydoc.synopsis(TESTFN, {})
+ self.assertEqual(synopsis, 'line 1: h\xe9')
+
class TestDescriptions(unittest.TestCase):