cordova-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Grieve <agri...@chromium.org>
Subject Re: IC: current state of IOS master
Date Thu, 09 Jan 2014 02:27:23 GMT
Zap! I've removed them all after ensuring no changes were made to them that
were not made also to mobile spec.

Only two exceptions:
One is the recent one made to file (which I guess sparked this conversion).
I applied the patch to mobile-spec via git log -p & patch -p1.

The other is a change made to file-transfer tests. This patch only applied
8/12 hunks cleanly and applied to WP, so I didn't go through with it. Jesse
- if you want to try it out, you can get the diff here:
https://dl.dropboxusercontent.com/u/6648754/filetransfer.tests.patch and
apply it with:
cd cordova-mobile-spec
patch -p1 < path/to/filetransfer.tests.patch


On Wed, Jan 8, 2014 at 2:05 PM, Ian Clelland <iclelland@google.com> wrote:

> We can go and delete them. MIchal's right -- they're actively harmful right
> now.
>
> As long as they stay in the history, then I can still use the stale
> versions as a reference when I update the Jasmine2 versions (hopefully
> really soon now). (Don't drop the cdvtest branch)
>
>
> On Wed, Jan 8, 2014 at 3:59 PM, Michal Mocny <mmocny@chromium.org> wrote:
>
> > On Wed, Jan 8, 2014 at 3:53 PM, Andrew Grieve <agrieve@chromium.org>
> > wrote:
> >
> > > Anyone object to me going through and deleting the stale copy of tests
> > > within plugins?
> > >
> >
> > +1. AFAIK they are not used for anything, are out of date, and are
> causing
> > confusion.
> >
> >
> > >
> > > On Wed, Jan 8, 2014 at 12:40 PM, Ian Clelland <iclelland@chromium.org
> > > >wrote:
> > >
> > > > They are out of sync, for sure. The mobile-spec tests should be
> > > considered
> > > > canonical, I think -- that's what is actually being run by the CI
> > server,
> > > > and by everyone who is doing manual testing, as far as I know.
> (Anyone
> > > who
> > > > knows different: now is the time to chime in :) )
> > > >
> > > > We're very close to being able to split all of the tests out of
> mobile
> > > > spec; we have a test runner which is close to being ready that will
> > pull
> > > in
> > > > tests from individual plugins. Once that's ready, we'll be able to
> move
> > > the
> > > > tests from MS back into plugins.
> > > >
> > > > Most plugins right now have a set of tests that was copied in circa
> > > Cordova
> > > > 2.9/3.0, and so they'll mostly need to be refreshed when we finally
> > move
> > > to
> > > > plugin-hosted-tests.
> > > >
> > > > (And I think the tests are back to passing now, from the look of the
> CI
> > > > dashboard -- thanks for doing that! :) )
> > > >
> > > > Ian
> > > >
> > > >
> > > >
> > > >
> > > > On Wed, Jan 8, 2014 at 3:23 PM, Jesse <purplecabbage@gmail.com>
> wrote:
> > > >
> > > > > I reverted the merge and pushed to master.
> > > > >
> > > > > So should the mobile-spec tests, or the tests in each plugin repo
> be
> > > > > considered the root?  I fear they are out of sync.
> > > > >
> > > > > @purplecabbage
> > > > > risingj.com
> > > > >
> > > > >
> > > > > On Wed, Jan 8, 2014 at 10:54 AM, Jesse <purplecabbage@gmail.com>
> > > wrote:
> > > > >
> > > > > > I did the premature merge for CB-5602, I'll revert it.
> > > > > >
> > > > > > @purplecabbage
> > > > > > risingj.com
> > > > > >
> > > > > >
> > > > > > On Wed, Jan 8, 2014 at 8:47 AM, Ian Clelland <
> > iclelland@chromium.org
> > > > > >wrote:
> > > > > >
> > > > > >> I think that this has to do with CB-5602. It looks like
the file
> > api
> > > > > tests
> > > > > >> were overwritten with a version from the cordova-plugin-file
> repo,
> > > > which
> > > > > >> would have been rather stale, and the merge caused the whole
> suite
> > > to
> > > > > >> contain errors.
> > > > > >>
> > > > > >> We appear to have a completely different issue in the CI
server,
> > > > because
> > > > > >> the tests should have failed on all platforms, not just
> > iOS/release.
> > > > > >>
> > > > > >> [I've replied to CB-5602 as well; not sure which is the
better
> > place
> > > > for
> > > > > >> discussion]
> > > > > >>
> > > > > >> Ian
> > > > > >>
> > > > > >>
> > > > > >>
> > > > > >> On Wed, Jan 8, 2014 at 11:08 AM, David Kemp <
> drkemp@chromium.org>
> > > > > wrote:
> > > > > >>
> > > > > >> > Unfortunately, our iOS deploy has been broken for a
while. I
> > just
> > > > > spent
> > > > > >> > some time getting it back on its feet and find that
master
> does
> > > not
> > > > > pass
> > > > > >> > tests. I do not have a simple 'blame list' because
many
> commits
> > > have
> > > > > >> passed
> > > > > >> > since the test deployed correctly.
> > > > > >> >
> > > > > >> > All tests pass on iOS 3.3.x
> > > > > >> >
> > > > > >> > All tests pass on Android Master and 3.3.x
> > > > > >> >
> > > > > >> > On iOS master 7 file tests are failing on iOS 6.1.3
and 7.04
> > > > > >> >
> > > > > >> >
> > > > > >> >                "*spec*": "File API DirectoryEntry file.spec.25
> > > > > >> > DirectoryEntry.getDirectory: create new dir with space
> > > > > >> resolveFileSystemURI
> > > > > >> > with encoded URI.",
> > > > > >> >
> > > > > >> >                        "*exception*": "timeout: timed
out
> after
> > > 7500
> > > > > >> msec
> > > > > >> > waiting for win never called",
> > > > > >> >
> > > > > >> >                "*spec*": "File API File file.spec.40
should be
> > > > define
> > > > > >> File
> > > > > >> > attributes.",
> > > > > >> >
> > > > > >> >                        "*exception*": "Expected undefined
to
> be
> > > > > >> defined.",
> > > > > >> >
> > > > > >> >                "*spec*": "File API Entry file.spec.63
copyTo:
> > > > > directory
> > > > > >> > that does not exist.",
> > > > > >> >
> > > > > >> >                        "*exception*": "TypeError: 'null'
is
> not
> > an
> > > > > >> object
> > > > > >> > (evaluating 'parent.filesystem.__format__') in
> > > > > >> >
> > > > > >> >
> > > > > >>
> > > > >
> > > >
> > >
> >
> file:///var/mobile/Applications/836872AD-CFED-4BE7-A316-45353151C586/mobilespec.app/www/plugins/org.apache.cordova.file/www/Entry.js
> > > > > >> > (line 165)",
> > > > > >> >
> > > > > >> >                        "*exception*": "timeout: timed
out
> after
> > > 7500
> > > > > >> msec
> > > > > >> > waiting for itCopy never called",
> > > > > >> >
> > > > > >> >                "*spec*": "File API Entry file.spec.79
moveTo:
> > > > > directory
> > > > > >> > that does not exist.",
> > > > > >> >
> > > > > >> >                        "*exception*": "timeout: timed
out
> after
> > > 7500
> > > > > >> msec
> > > > > >> > waiting for itMove never called",
> > > > > >> >
> > > > > >> >                "*spec*": "File API read method file.spec.82
> > should
> > > > > error
> > > > > >> > out on non-existent file.",
> > > > > >> >
> > > > > >> >                        "*exception*": "timeout: timed
out
> after
> > > 7500
> > > > > >> msec
> > > > > >> > waiting for verifier never called",
> > > > > >> >
> > > > > >> >                "*spec*": "File API FileWriter file.spec.97
> > should
> > > be
> > > > > >> able
> > > > > >> > to write and append to file, File object.",
> > > > > >> >
> > > > > >> >                        "*exception*": "timeout: timed
out
> after
> > > 7500
> > > > > >> msec
> > > > > >> > waiting for verifier",
> > > > > >> >
> > > > > >> >                "*spec*": "File API Backwards compatibility
> > > > > file.spec.109
> > > > > >> > should be able to resolve a file:/// URL.",
> > > > > >> >
> > > > > >> >                        "*exception*": "ReferenceError:
Can't
> > find
> > > > > >> variable:
> > > > > >> > joinURL in
> > > > > >> >
> > > > > >> >
> > > > > >>
> > > > >
> > > >
> > >
> >
> file:///var/mobile/Applications/836872AD-CFED-4BE7-A316-45353151C586/mobilespec.app/www/autotest/tests/file.tests.js
> > > > > >> > (line 3599)",
> > > > > >> >
> > > > > >>
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message