To make things simpler when upstreaming, lets start by just requiring users have simplejson installed to use run-chromium-webkit-tests. I can fix the autoinstall stuff later.
On Thu, Jan 21, 2010 at 6:01 PM, Adam Barth <[email protected]> wrote: > According to that site, the package is available under: > > License :: OSI Approved :: MIT License > > Is that compatible with being landed in the svn.webkit.org tree? > > Adam > > > On Thu, Jan 21, 2010 at 5:58 PM, Dirk Pranke <[email protected]> wrote: >> Autoinstall appears to be pretty flaky for me at the moment, and these >> aren't the sort of scripts that can be flaky :) I'll install >> simplejson into WebKitTools/simplejson as part of the patch and we can >> clean it up once things are stable. >> >> -- Dirk >> >> On Thu, Jan 21, 2010 at 2:45 PM, Eric Seidel <[email protected]> wrote: >>> Be aware, there are a couple known issues with our current autoinstall >>> setup: >>> https://bugs.webkit.org/show_bug.cgi?id=33632 >>> https://bugs.webkit.org/show_bug.cgi?id=33365 >>> >>> On Thu, Jan 21, 2010 at 2:35 PM, Adam Barth <[email protected]> wrote: >>>> This page makes it looks like we can just autoinstall simplejson: >>>> >>>> http://pypi.python.org/pypi/simplejson/ >>>> >>>> IMHO, that's better than checking it in. >>>> >>>> Adam >>>> >>>> >>>> On Thu, Jan 21, 2010 at 1:58 PM, Eric Seidel <[email protected]> wrote: >>>>> We also have webkitpy/autoinstall.py which knows how to download >>>>> modules on-demand. This is useful in the case that you're using code >>>>> with an incompatible license. >>>>> >>>>> see webkitpy/__init__.py for an example of how we pull in mechanize >>>>> (which is a HUGE module, with a compatible license for most files, but >>>>> includes a bit of code which is not compatible). >>>>> >>>>> On Thu, Jan 21, 2010 at 1:39 PM, David Levin <[email protected]> wrote: >>>>>> So far, things have been put in place with other files (like the python >>>>>> websockets code) but I think that is confusing for a number of reasons: >>>>>> 1. The level of review needed varies imo between 3rd party code that is >>>>>> being used vs new code added to wk. >>>>>> 2. The style never seems to match what is done for the rest of WebKit. >>>>>> 3. It becomes unclear how to update it because there aren't always good >>>>>> concise instructions about this. >>>>>> Personally, I'd much prefer a ThirdParty directory to check these things >>>>>> into which we could use to help address these things. >>>>>> dave >>>>>> >>>>>> On Thu, Jan 21, 2010 at 1:30 PM, Dirk Pranke <[email protected]> >>>>>> wrote: >>>>>>> >>>>>>> Hi all, >>>>>>> >>>>>>> I'm about to upload a patch that depends on third-party python code >>>>>>> (simplejson). The patch is a bunch of scripts that'll live under >>>>>>> WebKitTools/Scripts . Is there an appropriate place for the simplejson >>>>>>> code? In the absence of a better location, I'll probably check it in >>>>>>> under WebKitTools. >>>>>>> >>>>>>> -- Dirk >>>>>>> _______________________________________________ >>>>>>> webkit-dev mailing list >>>>>>> [email protected] >>>>>>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> webkit-dev mailing list >>>>>> [email protected] >>>>>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev >>>>>> >>>>>> >>>>> _______________________________________________ >>>>> webkit-dev mailing list >>>>> [email protected] >>>>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev >>>>> >>>> >>> _______________________________________________ >>> webkit-dev mailing list >>> [email protected] >>> http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev >>> >> > _______________________________________________ webkit-dev mailing list [email protected] http://lists.webkit.org/mailman/listinfo.cgi/webkit-dev
