Bug 37484
| Summary: | new-run-webkit-tests: path_from_chromium_base Chromium hacks are sprinkled throughout win32 codepath | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Eric Seidel (no email) <eric> |
| Component: | Tools / Tests | Assignee: | Dirk Pranke <dpranke> |
| Status: | RESOLVED FIXED | ||
| Severity: | Normal | CC: | abarth, aroben, dpranke |
| Priority: | P2 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | PC | ||
| OS: | OS X 10.5 | ||
Eric Seidel (no email)
path_from_chromium_base Chromium hacks are sprinkled throughout new-run-webkit-tests win32 codepath
This blocks the WebKit Win port from working.
For now I think we should just hack around them with a few "if options.chromium" checks.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Eric Seidel (no email)
Example hack:
# Put the cygwin directory first in the path to find cygwin1.dll
env = os.environ
if sys.platform in ('cygwin', 'win32'):
env['PATH'] = '%s;%s' % (
self._port_obj.path_from_chromium_base('third_party',
'cygwin', 'bin'),
env['PATH'])
if sys.platform == 'win32' and self._register_cygwin:
setup_mount = self._port_obj.path_from_chromium_base('third_party',
'cygwin', 'setup_mount.bat')
subprocess.Popen(setup_mount).wait()
Eric Seidel (no email)
I think it's more that the http_server.py module doesn't really use the port infrastructure yet.
Dirk Pranke
This only calls to that method I see now are from the port/chromium*.py files, and it looks like they are all there legitimately to deal with downstream expectations, results, or builds.
I'm guessing someone (me?) must've cleaned this up at some point in the path. Please reopen or file a new bug if you see any other issues.