cordova-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Edna Y Morales <eymor...@us.ibm.com>
Subject Re: iOS mobile-spec failures (was: Re: [Testing] No content to speak of)
Date Thu, 28 Aug 2014 00:50:00 GMT

Running only the file plugin tests, updating the version of jasmine in the
assets brought the failures down from 16 to 0. Thanks for the suggestion
Michal!
I will see what the results are for the rest of the plugins with the
updated version of jasmine tomorrow.

Thanks,
Edna Morales



From:	Michal Mocny <mmocny@chromium.org>
To:	dev <dev@cordova.apache.org>
Date:	08/27/2014 04:48 PM
Subject:	Re: iOS mobile-spec failures (was: Re: [Testing] No content to
            speak of)
Sent by:	mmocny@google.com



One quick thing to try is to update the version of jasmine used by the
harness app.  2.0 is new and its been a while since the assets were
updated.

-Michal


On Wed, Aug 27, 2014 at 1:14 PM, Edna Y Morales <eymorale@us.ibm.com>
wrote:

>  Yeah I can confirm that behavior. I was seeing it this morning when I
> noticed that some specs were showing up as failing twice?
>
> Thanks,
> Edna Morales
>
> [image: Inactive hide details for Martin Gonzalez ---08/27/2014 01:06:06
> PM---During testing, sometimes a callback is reached more than]Martin
> Gonzalez ---08/27/2014 01:06:06 PM---During testing, sometimes a callback
> is reached more than one time, this strange behavior I've faced
>
> From: Martin Gonzalez <martin.c.glez.glez@gmail.com>
> To: dev@cordova.apache.org
> Date: 08/27/2014 01:06 PM
> Subject: Re: iOS mobile-spec failures (was: Re: [Testing] No content to
> speak of)
> ------------------------------
>
>
>
> During testing, sometimes a callback is reached more than one time, this
> strange behavior I've faced over Windows phone with the Media tests,
> creating a loop. Then Edna test it out over iOS a few weeks ago and I
> recall that some of the callbacks were reached more than once, which is
> unusual.
> Options: Fix that behavior over iOS, and also set some routine over the
> file.tests to detect that kind of behavior, fail the test and move on
with
> the next, this would avoid that kind of loop.
>
> Could you guys confirm that kind of behavior?
>
>
> 2014-08-27 11:47 GMT-05:00 Ian Clelland <iclelland@chromium.org>:
>
> > Something strange is definitely going on in the file tests; in the
> vicinity
> > of spec.109. Custom matchers are not being added after that spec, but
if
> I
> > remove it, then the tests simply stop at that point.
> >
> >
> >
> > On Wed, Aug 27, 2014 at 11:40 AM, Ian Clelland <iclelland@chromium.org>
> > wrote:
> >
> > > No, it appears that something is up with Jasmine on ios -- the custom
> > > matchers aren't being attached in all cases, so I am seeing errors in
> the
> > > console like:
> > >
> > > TypeError: 'undefined' is not a function (evaluating
> > > 'expect(fileEntry.name).toCanonicallyMatch(fileName)')
> > >
> > > because .toCanonicallyMatch is not present. It's odd, because it is
> > > present in previous tests, so I don't know why it would be failing
> after
> > a
> > > certain point.
> > >
> > >
> > > On Wed, Aug 27, 2014 at 11:03 AM, Ian Clelland
<iclelland@chromium.org
> >
> > > wrote:
> > >
> > >> I'll check that out; the file changes could be related to my commits
> > >> yesterday
> > >>
> > >>
> > >> On Wed, Aug 27, 2014 at 10:58 AM, Edna Y Morales
<eymorale@us.ibm.com
> >
> > >> wrote:
> > >>
> > >>> I'm seeing many many failing iOS tests. They are seen across file,
> > >>> file-transfer, globalization, geolocation, and media with timeout
and
> > >>> reference errors mostly. Can anyone else verify this or is it just
> me?
> > >>>
> > >>> Thanks,
> > >>> Edna Morales
> > >>>
> > >>> [image: Inactive hide details for "Sergey Grebnov (Akvelon)"
> > >>> ---08/27/2014 09:39:45 AM---Vladimir Kotikov and me will join this
> > testing]"Sergey
> > >>> Grebnov (Akvelon)" ---08/27/2014 09:39:45 AM---Vladimir Kotikov and
> me
> > will
> > >>> join this testing party soon and help with the testing and
necessary
> im
> > >>>
> > >>> From: "Sergey Grebnov (Akvelon)" <v-segreb@microsoft.com>
> > >>> To: "dev@cordova.apache.org" <dev@cordova.apache.org>
> > >>> Date: 08/27/2014 09:39 AM
> > >>> Subject: RE: [Testing] No content to speak of
> > >>> ------------------------------
> > >>>
> > >>>
> > >>>
> > >>> Vladimir Kotikov and me will join this testing party soon and help
> with
> > >>> the testing and necessary improvements(if any). Currently we see
some
> > >>> problem with dynamic manual tests generation (at least in Camera
> > tests) on
> > >>> Windows due to special unsafe html restrictions on this platform,
> > Vladimir
> > >>> will send details and workaround.
> > >>>
> > >>> Thx!
> > >>> Sergey
> > >>> -----Original Message-----
> > >>> From: Martin Gonzalez [mailto:martin.c.glez.glez@gmail.com
> <martin.c.glez.glez@gmail.com>
> > >>> <martin.c.glez.glez@gmail.com>]
> > >>> Sent: Wednesday, August 27, 2014 5:12 PM
> > >>> To: dev@cordova.apache.org
> > >>> Subject: Re: [Testing] No content to speak of
> > >>>
> > >>> Agree, over android it looks good, however running the tests over
> > >>> Windows Phone, the final count is 13 new fails.
> > >>> Specs failing: 9, 9.5, 10, 11, 15, 24, 82, 110, 116, 121, 122, 123,
> > 124.
> > >>>
> > >>>
> > >>>
> > >>> 2014-08-26 11:20 GMT-05:00 Marcel Kinard <cmarcelk@gmail.com>:
> > >>>
> > >>> > Thanks, Ian! On Android 4.4, I'm now getting no failures with
the
> all
> > >>> > the automatic tests (new-style plugin tests and old-style
> non-plugin
> > >>> tests).
> > >>> >
> > >>> > Now to start poking at the manual tests...
> > >>> >
> > >>> > On Aug 26, 2014, at 11:27 AM, Ian Clelland
<iclelland@chromium.org
> >
> > >>> wrote:
> > >>> >
> > >>> > > The other issue should be resolved now, but if someone with
a
WP8
> > >>> > > device can test it out as well, it'd be appreciated.
> > >>> >
> > >>> >
> > >>>
> > >>>
> > >>> --
> > >>> Regards,
> > >>> Martin Gonzalez
> > >>>
> > >>>
> > >>
> > >
> >
>
>
>
> --
> Regards,
> Martin Gonzalez
>
>

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