RE: Windows8 and 8.1, to merge or to branch

2013-11-27 Thread Maxime LUCE
I agree in some ways. Correct me if I'm wrong but if we have a separate platform for Windows 8.1, it would be the tool in cli you want to convert an app to Windows 8.1 in near term, doesn't it ? In long term I agree that we should merge both platforms but I think that will not be in 6 month li

Platforms in subdirectory into repository

2013-11-27 Thread Maxime LUCE
In order to improve support of platforms in subdirectory, I created an improvement issue on JIRA : https://issues.apache.org/jira/browse/CB-5392 Then I created a fix available here : https://github.com/apache/cordova-cli/pull/91 Do you agree with that way ? Cordialement.

Re: Platforms in subdirectory into repository

2013-11-27 Thread Bryan Higgins
I like the idea of standardizing it this way. Was this patch tested on bb10? On Wed, Nov 27, 2013 at 6:21 AM, Maxime LUCE wrote: > In order to improve support of platforms in subdirectory, > > I created an improvement issue on JIRA : > https://issues.apache.org/jira/browse/CB-5392 > Then I crea

RE: Platforms in subdirectory into repository

2013-11-27 Thread Maxime LUCE
I didn't test it specifically but I take it into account in my development. I removed the specific test for blackberry as it is standardized in lazy_load.js, you can see it at : https://github.com/apache/cordova-cli/pull/91/files#diff-a97c0300b0c3e72467e54b438343f607L102 Cordialement. --

Intro email

2013-11-27 Thread Kyle Nitzsche
Hi, I am Kyle Nitzsche, and I work for Canonical on Ubuntu. I will be providing the docs for the Ubuntu platform support recently submitted. Cheers, Kyle

Re: Intro email

2013-11-27 Thread James Jong
Welcome Kyle! -James Jong On Nov 27, 2013, at 10:41 AM, Kyle Nitzsche wrote: > Hi, > > I am Kyle Nitzsche, and I work for Canonical on Ubuntu. > > I will be providing the docs for the Ubuntu platform support recently > submitted. > > Cheers, > Kyle

Re: plugins.cordova.io

2013-11-27 Thread James Jong
+1 -James Jong On Nov 26, 2013, at 5:29 PM, Steven Gill wrote: > yeah I don't see why not. The core plugins all already have them, they are > just not exposed yet. I will see if I can get some of these showing before > US thanksgiving > > > On Tue, Nov 26, 2013 at 2:10 PM, Michal Mocny wrote:

[FxOS] Special needs for contact plugin

2013-11-27 Thread Piotr Zalewa
Hi, I've almost finished the Contacts proxy for Firefox OS. [1] Access to contacts API in FxOS requires webapp to be "privileged" and special permissions needs to be set in app's manifest. The privileged app enforces Content Security Policy [2] which doesn't allow for inline javascript. Becau

Re: Platforms in subdirectory into repository

2013-11-27 Thread purplecabbage
I am currently testing it. Tests are currently f'd on windows, so it is very difficult to both develop and test as you require both os's. I am looking at fixing the way the tests are run so we can simplify this. Sent from my iPhone > On Nov 27, 2013, at 7:31 AM, Maxime LUCE wrote: > > I did

RE: Platforms in subdirectory into repository

2013-11-27 Thread Maxime LUCE
You can run each test separately by doing in console (in your cordova-cli root) node node_modules\jasmine-node\bin\jasmine-node spec\build.spec.js Tests which fail on Windows are located in hook.spec.js if I remember correctly. Cordialement. Maxime LUCE max...@touchi

Re: Intro email

2013-11-27 Thread Lorin Beer
Hi Kyle, welcome to the list! On Wed, Nov 27, 2013 at 9:01 AM, James Jong wrote: > Welcome Kyle! > -James Jong > > On Nov 27, 2013, at 10:41 AM, Kyle Nitzsche > wrote: > > > Hi, > > > > I am Kyle Nitzsche, and I work for Canonical on Ubuntu. > > > > I will be providing the docs for the Ubuntu

Re: [FxOS] Special needs for contact plugin

2013-11-27 Thread Lorin Beer
traditionally, we have had a per platform quirks section in the relevant api's documentation. Search for quirks in the contacts api and you should find sections for Android, BlackBerry, iOS and WP 7 and 8. Just add

Re: [For discussion] windows8 and run/emulate commands meaning

2013-11-27 Thread Lorin Beer
+1, we want the cli commands to mean the exact same thing regardless of platform targeted. 'Conjugating' the command based on platform breaks this design. Better to fail noisily with the commands that are not supported, and document everything along the way. On Tue, Nov 26, 2013 at 5:13 PM, Jess

Re: Intro email

2013-11-27 Thread Brian LeRoux
Welcome aboard Kyle! On Wed, Nov 27, 2013 at 9:37 AM, Lorin Beer wrote: > Hi Kyle, welcome to the list! > > > On Wed, Nov 27, 2013 at 9:01 AM, James Jong wrote: > > > Welcome Kyle! > > -James Jong > > > > On Nov 27, 2013, at 10:41 AM, Kyle Nitzsche > > > wrote: > > > > > Hi, > > > > > > I am K

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Naik, Archana
Hi, Steven Do you know when will amazon-fireos be part of cordova's official release? I see 3.2 was releases last week. When is the next release planned? Also, we wanted to support 3.0 and 3.1 as well. There are pull requests for these too. Is there any chance of merging those? Thanks Archana O

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Brian LeRoux
Aiming for a Dec release of 3.3 which hopefully can include both FireOS and Ubuntu platforms but, it is Dec, so we might have to bump to Jan. On Wed, Nov 27, 2013 at 9:58 AM, Naik, Archana wrote: > Hi, Steven > > Do you know when will amazon-fireos be part of cordova's official release? > I see

releasing 3.2.0 on jira

2013-11-27 Thread Shazron
Going to do it later today - push 3.2.0 outstanding issues to 3.3.0. Prepare your inboxes.

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Steven Gill
I don't think it makes a lot of sense to back port to 3.0 and 3.1. Users would need to do an npm install to get the updated version. Instead of doing that, we recommend people update to the latest. If we don't do a 3.3 release in Dec, I could still see Ubuntu and FireOS landing with the next tools

Re: Intro email

2013-11-27 Thread Steven Gill
Welcome Kyle! The best place to add them would be to docs.cordova.io ( https://git-wip-us.apache.org/repos/asf?p=cordova-docs.git;a=summary). Let me know if you have any questions! On Wed, Nov 27, 2013 at 9:55 AM, Brian LeRoux wrote: > Welcome aboard Kyle! > > > On Wed, Nov 27, 2013 at 9:37

Re: [FxOS] Special needs for contact plugin

2013-11-27 Thread Steven Gill
The CSP is quite restrictive eh. This will definitely make it more difficult to port apps that require plugins that make the app "privileged". No way around this it seems. I would add a short summary + link to the CSP in the quirks section for all of the plugins that will turn the application into

Re: android target level

2013-11-27 Thread Michal Mocny
Would be nice to change scripts to check for >= target-level instead of exactly ==. I currently just re-installed my android SDK locally with only 4.4 (API 19), and now fail to create a project with CLI since it isnt exactly the version we look for. Since there won't be another android release fo

Re: android target level

2013-11-27 Thread Joe Bowser
The problem is that we should never be relying on the CLI for this info, and that the platforms should be doing it. That's why I edited the bug in the first place. This is also a problem with iOS, although less of a problem due to >=. On Wed, Nov 27, 2013 at 11:43 AM, Michal Mocny wrote: > Woul

Re: android target level

2013-11-27 Thread Michal Mocny
Ah, indeed. CLI has hard coded req on 17 in module.exports.check_requirements in android_parser.js. That needs to change especially quickly now since the platform SDK dependency was bumped to 18 in our 3.2 release (so CLI and platform workflows have atm different requirements). On Wed, Nov 27,

Re: android target level

2013-11-27 Thread Braden Shepherdson
The correct fix here is to scrap CLI's half-assed custom check_requirements implementation and have it call the real Android platform script instead. Then that script should do the right thing re: API versions. Braden On Wed, Nov 27, 2013 at 11:59 AM, Michal Mocny wrote: > Ah, indeed. CLI has

Re: android target level

2013-11-27 Thread Joe Bowser
I'm bumping it to 19 now. You just change project.properties to do this on Android. :) On Wed, Nov 27, 2013 at 11:59 AM, Michal Mocny wrote: > Ah, indeed. CLI has hard coded req on 17 in > module.exports.check_requirements in android_parser.js. That needs to > change especially quickly now sin

Test changes for windows cordova-cli

2013-11-27 Thread Jesse
The recent e2e tests have serious issues on windows, too much for me to consider jumping into right now, so instead of fixing them, I just added a new test fixture. If you want to run tests on windows, you can use : $ npm run win-test I have also added a runtime platform check to the hooker.spec.j

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Brian LeRoux
Oh ya, man this decoupled world is so much better. That works well if that support is landed in cli/plugman then. As far as userland is concerned that is landing support. On Wed, Nov 27, 2013 at 11:20 AM, Steven Gill wrote: > I don't think it makes a lot of sense to back port to 3.0 and 3.1

CLI testing

2013-11-27 Thread David Kemp
a recent set of commits landed about 1/2 hour ago has broken the CLI tests 343 tests, 572 assertions, 59 failures looks like most are : Error: ENOENT, no such file or directory

Re: CLI testing

2013-11-27 Thread Jesse
okay, yeah, just switched over to OSX, and there are numerous (59) failures here too as well. I have verified that these are all a result of the e2e tests folder, if you run : jasmine-node --color spec You will not see any of these failures. blame @mmocny ;) @purplecabbage risingj.com On Wed

Re: CLI testing

2013-11-27 Thread Steven Gill
I think we should revert these out and put them on a branch until ready. Thoughts? On Wed, Nov 27, 2013 at 12:59 PM, Jesse wrote: > okay, yeah, just switched over to OSX, and there are numerous (59) failures > here too as well. > > I have verified that these are all a result of the e2e tests f

Re: CLI testing

2013-11-27 Thread Lorin Beer
+1 on reverting these out. Pushing large commits that causes widespread failing tests on one or more platforms should be reverted out. They should not be pushed up to master in the first place. On Wed, Nov 27, 2013 at 1:16 PM, Steven Gill wrote: > I think we should revert these out and put them

Cordova CLI minor update

2013-11-27 Thread Lorin Beer
Michael and I released another version of the Cordova CLI this morning. This is a minor version bump, and includes a commit to fix a blocking issue for projects that consume the Cordova CLI as a library. This was tested on OSX and Windows before the release, and should not affect the Cordova CLI

Re: CLI testing

2013-11-27 Thread Jesse
they all live in their own folder, so it is easy to run tests without running e2e tests. ( ie we do at least have some good separation ) Just modify package.json 's scripts{test:} to not include e2e/ until they are fixed. @purplecabbage risingj.com On Wed, Nov 27, 2013 at 1:20 PM, Lorin Beer wro

any plan to have a Qt5 backend?

2013-11-27 Thread haithem rahmani
Hi, the Qt backend has been removed since a while from official cordova releases. Qt5 has been released since almost last year, and it is aimed to run really everywhere, including android, ios. So I'm wondering whether there are plans to a cordova Qt5 backend or not? thanks haithem -- *Never s

Re: CLI testing

2013-11-27 Thread Steven Gill
I see it is easy to fix by changing npm test to not run e2e tests. The issue here is that this code should never have made it to master while broken. I am going to revert the e2e tests out and they can be resubmitted on a branch and merged in when ready. We all need to be diligent in making sure br

Re: CLI testing

2013-11-27 Thread Braden Shepherdson
Revert them out if you like, that's the right flow for broken commits. There's no need for a lecture on not checking in broken code - it wasn't broken on our machines! Braden On Wed, Nov 27, 2013 at 1:36 PM, Steven Gill wrote: > I see it is easy to fix by changing npm test to not run e2e test

Re: android target level

2013-11-27 Thread Michal Mocny
But the file warns that its auto-generated and shouldn't be edited ;) I've got a patch for this. Seems of the other platforms were already doing it right. Reviewboard incoming since I don't much with CLI often. On Wed, Nov 27, 2013 at 3:03 PM, Joe Bowser wrote: > I'm bumping it to 19 now. Y

Review Request 15889: [android] call out to platform check_req script

2013-11-27 Thread Michal Mocny
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15889/ --- Review request for cordova. Repository: cordova-cli Description --- .

Re: android target level

2013-11-27 Thread Michal Mocny
https://reviews.apache.org/r/15889/ On Wed, Nov 27, 2013 at 4:38 PM, Michal Mocny wrote: > But the file warns that its auto-generated and shouldn't be edited ;) > > I've got a patch for this. Seems of the other platforms were already > doing it right. Reviewboard incoming since I don't much w

Re: Review Request 15889: [android] call out to platform check_req script

2013-11-27 Thread Joe Bowser
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15889/#review29514 --- I think that this should be shipped. However, someone who is more f

Re: android target level

2013-11-27 Thread Michal Mocny
It seems that all the platform scripts do something subtly different. Some of them do support custom platform paths, others don't, etc. It seems like something we could easily consolidate. On Wed, Nov 27, 2013 at 4:51 PM, Michal Mocny wrote: > https://reviews.apache.org/r/15889/ > > > On Wed,

Re: Review Request 15889: [android] call out to platform check_req script

2013-11-27 Thread Braden Shepherdson
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15889/#review29517 --- src/metadata/android_parser.js

Review Request 15891: Don't require JAVA_HOME to be defined

2013-11-27 Thread Mark Koudritsky
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15891/ --- Review request for cordova. Bugs: CB-5422 https://issues.apache.org/jira/br

Re: Review Request 15889: [android] call out to platform check_req script

2013-11-27 Thread Michal Mocny
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15889/#review29518 --- src/metadata/android_parser.js

Re: Review Request 15891: Don't require JAVA_HOME to be defined

2013-11-27 Thread Joe Bowser
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15891/#review29519 --- Ship it! Ship It! - Joe Bowser On Nov. 27, 2013, 10:08 p.m., Mar

Re: Review Request 15891: Don't require JAVA_HOME to be defined

2013-11-27 Thread Steven Gill
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15891/#review29521 --- Ship it! Ship It! - Steven Gill On Nov. 27, 2013, 10:08 p.m., Ma

Re: Review Request 15891: Don't require JAVA_HOME to be defined

2013-11-27 Thread Mark Koudritsky
> On Nov. 27, 2013, 10:25 p.m., Steven Gill wrote: > > Ship It! Would be grateful if someone with commit right could merge it. Thanks for review. - Mark --- This is an automatically generated e-mail. To reply, visit: https://reviews.apa

Re: any plan to have a Qt5 backend?

2013-11-27 Thread Brian LeRoux
I beleive our friends at Canonical have the freshest copy of Qt stuff as it relates to Cordova. No idea how much to genericize that would be. Hmm. Is genericize a word? On Wed, Nov 27, 2013 at 1:28 PM, haithem rahmani wrote: > Hi, > > the Qt backend has been removed since a while from official

[Android] Adding Robotium to Platform Tests

2013-11-27 Thread Joe Bowser
Hey I'm adding some more messages to Cordova for the plugins to handle, namely onScrollChanged, and some of the Context Menu plugins, and since I'm currently writing tests, I'm wondering how we feel about adding Robotium as a dependency for testing. I think that it's appropriate for our needs, an

Re: Cordova for Ubuntu

2013-11-27 Thread Steven Gill
Cordova-plugman + cordova-cli both have ubuntu branches. When I run npm test on cordova cli, I get one failing test. I am still in the process of manually testing both of these (setting up vagrant with ubuntu). You will notice I made a commit to the CLI to use apache's repo to grab cordova-ubuntu (

RE: Test changes for windows cordova-cli

2013-11-27 Thread Maxime LUCE
I confirm that hooker.spec.js is failing on windows because of file system methods. We can also try to spy on file system or shelljs methods and force result instead of using real FS which is OS dependant. In near term it's better to use your method but in long term it could be great to study

Re: Review Request 15889: [android] call out to platform check_req script

2013-11-27 Thread Carlos Santana
--- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/15889/#review29528 --- Did you performed a test running on Windows? Remember that CLI is n

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Carlos Santana
Steven if you add support in 3.2.0-0.2.0, I think the platforms should be release with version 3.2.0 to keep consistency and avoid user more confuse about our already complex versions scheme. On Wed, Nov 27, 2013 at 3:40 PM, Brian LeRoux wrote: > Oh ya, man this decoupled world is so much

Re: Updating FileTransfer

2013-11-27 Thread Andrew Grieve
AFAIK, on Android and iOS, it's not possible to create a native Blob / File object that points to a file on the filesystem. The exception is on Android *before* kitkat, we could maybe use the to do this. Without this ability, we'll need to keep FileTransfer around. Neat! URLs enable non-upload/do

Re: amazon-fireos core plugin pull requests..Please review

2013-11-27 Thread Steven Gill
I already have flexible 3.2.0 tags on both Cordova-Ubuntu and Cordova-Amazon-fireos. I am using it for testing the cli currently. Flexible because I am willing to delete and retag while we are dealing with integration and testing. Instead of 3.2.0-0.2.0 I will probably tag it 3.2.0-1.0.0. Maybe o

Re: cordova-plugins repo release

2013-11-27 Thread Andrew Grieve
Back to the original question - we should specify how cordova-plugins should be included in plugin releases. Really, the thing we need is to be able to know if there are any unreleased commits and to update releasenotes.md for each plugin. I think it'd be enough to run "git log -- plugin-dir" and

Re: cordova-plugins repo release

2013-11-27 Thread Michal Mocny
I think the conclusion was that cordova-plugins are just *not* included in plugin releases. On Wed, Nov 27, 2013 at 11:32 PM, Andrew Grieve wrote: > Back to the original question - we should specify how cordova-plugins > should be included in plugin releases. > > Really, the thing we need is to