grouptools currently acts like a time-bomb: spite the good intention, it
currently ignores when developers mistakedly use os.path.join with
grouptools on Posix; then it explodes when the same code is used on
Windows.
This makes grouptools behavior on Windows the same as Linux, ie.,
silently ignore when os.path.join is mixed.
Another solution would be to use a different separator character on
Linux, but that would be a much more complex change than I can afford to
make at this moment.
---
framework/grouptools.py | 25 +++++++++++++++++++++++++
1 file changed, 25 insertions(+)
diff --git a/framework/grouptools.py b/framework/grouptools.py
index 3d26bbc..1000a51 100644
--- a/framework/grouptools.py
+++ b/framework/grouptools.py
@@ -30,6 +30,7 @@ posix paths they may not start with a leading '/'.
"""
import posixpath
+import os.path
__all__ = [
'join',
@@ -42,6 +43,22 @@ __all__ = [
'from_path'
]
+
+def _normalize(group):
+ """Helper to normalize group paths on Windows.
+
+ Although grouptools' heart is in the right place, the fact is that it fails
+ to spot when developers mistakedly use os.path.join for groups on Posix
+ systems.
+
+ So until this is improved somehow, make grouptools behavior on Windows
+ match Linux, ie, just silently ignore mixed use of grouptools and os.path.
+ """
+ if os.path.sep != '/':
+ group = group.replace(os.path.sep, '/')
+ return group
+
+
def _assert_illegal(group):
"""Helper that checks for illegal characters in input."""
assert isinstance(group, (str, unicode)), 'Type must be string or unicode'
@@ -61,6 +78,7 @@ def testname(group):
Analogous to os.path.basename
"""
+ group = _normalize(group)
_assert_illegal(group)
return posixpath.basename(group)
@@ -76,6 +94,7 @@ def groupname(group):
Analogous to os.path.dirname
"""
+ group = _normalize(group)
_assert_illegal(group)
return posixpath.dirname(group)
@@ -83,6 +102,7 @@ def groupname(group):
def splitname(group):
"""Split a group name, Returns tuple "(group, test)"."""
+ group = _normalize(group)
_assert_illegal(group)
return posixpath.split(group)
@@ -90,6 +110,7 @@ def splitname(group):
def commonprefix(args):
"""Given a list of groups, returns the longest common leading
component."""
+ args = [_normalize(group) for group in args]
for group in args:
_assert_illegal(group)
@@ -103,6 +124,7 @@ def join(*args):
'\\' in them, as these are groups not paths.
"""
+ args = [_normalize(group) for group in args]
for group in args:
assert isinstance(group, (str, unicode)), \
'Type must be string or unicode'
@@ -121,6 +143,8 @@ def relgroup(large, small):
start is longer than the group then '' is returned.
"""
+ large = _normalize(large)
+ small = _normalize(small)
for element in {large, small}:
_assert_illegal(element)
@@ -138,6 +162,7 @@ def split(group):
If input is '' return an empty list
"""
+ group = _normalize(group)
_assert_illegal(group)
if group == '':
return []
--
2.1.0
_______________________________________________
Piglit mailing list
Piglit@lists.freedesktop.org
https://urldefense.proofpoint.com/v2/url?u=http-3A__lists.freedesktop.org_mailman_listinfo_piglit&d=AwIFAg&c=Sqcl0Ez6M0X8aeM67LKIiDJAXVeAw-YihVMNtXt-uEs&r=zfmBZnnVGHeYde45pMKNnVyzeaZbdIqVLprmZCM2zzE&m=SsqcjtAtzTdE_vBy1rhD_WiZrV0ZdU_dE5uIAzr47F4&s=wQGaXZ5nxvOyc6gdSsbDQr_POe2MSYH6VbLv17cBbQk&e=
