Re: [LAZY] Asking Infra to make all git repos read-write.

Here is David@Infra's response. This is the bug he is referring to:[1] David, > Folks outside the US are experiencing slowness in acccessing the git > repos. Is this something that can/will be resolved? > Also, what happens to the existing repos? Will they stay around as > read-only forever? We a

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:38 PM, Alex Harui wrote: > > > > On 3/17/13 9:22 PM, "Om" wrote: > > > On Sun, Mar 17, 2013 at 9:09 PM, Alex Harui wrote: > > > >> What does "fully tested" really mean? If a unix diff doesn't find > anything > >> missing, isn't that sufficient? > >> > > > > That and t

Re: [LAZY] Asking Infra to make all git repos read-write.

On 3/17/13 9:22 PM, "Om" wrote: > On Sun, Mar 17, 2013 at 9:09 PM, Alex Harui wrote: > >> What does "fully tested" really mean? If a unix diff doesn't find anything >> missing, isn't that sufficient? >> > > That and the history, I guess. > We've spot checked history in all of the repos,

Re: [LAZY] Asking Infra to make all git repos read-write.

OK, I looked at utilities. The Git migration did not take any files from SVN that were deleted. All of the missing history is associated with files that are not in SVN at the head. Interestingly, I cannot find a log entry for moving those files. This makes me think that Infra did it correctly b

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:24 PM, Frédéric THOMAS wrote: > The problem is that git remote add -f origin > https://git-wip-us.apache.org/**repos/asf/flex-sdk.gitdownload > all the repo and then, once downloaded, it filters the directories. > > A

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:19 PM, Alex Harui wrote: > Fred and Justin said your version cloned the entire repo first. This post > claims to not require that. > > Not true. You can try it out yourself. It took a few minutes for me. After following those steps, I have only the .git and bigosmall

Re: [LAZY] Asking Infra to make all git repos read-write.

The problem is that git remote add -f origin https://git-wip-us.apache.org/repos/asf/flex-sdk.git download all the repo and then, once downloaded, it filters the directories. Anyway, I just give another try downloading the flex-whiteboard and it is quicker than this afternoon, actually, it dow

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:09 PM, Alex Harui wrote: > What does "fully tested" really mean? If a unix diff doesn't find anything > missing, isn't that sufficient? > That and the history, I guess. > > IMO, Falcon is ready to be read/write. Same for flex-sdk and flex-asjs and > flex-tlf. > > Bu

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:05 PM, Frédéric THOMAS wrote: > It's not fully tested yet, flex-utilities (bad history lenght), I will ping Infra about this issue. > flex-whiteboard (not agree), Should we wait till we agree to make it read-write? Not sure what Infra can do about this. > falcon

Re: [LAZY] Asking Infra to make all git repos read-write.

On Sun, Mar 17, 2013 at 9:01 PM, Alex Harui wrote: > Did the "external" folder get moved over to git? (Not TLF as an > svn:external, there is a folder called "external" at our SVN top-level. > > Not yet. It was outstanding last night. I will ping Infra. > Didn't Fred find an issue with "utili

Re: [LAZY] Asking Infra to make all git repos read-write.

Fred and Justin said your version cloned the entire repo first. This post claims to not require that. On 3/17/13 9:13 PM, "Om" wrote: >> >> >> BTW, with whiteboard, I won't have time to try this but can one of you? >> >> http://stackoverflow.com/questions/4114887/is-it-possible-to-do-a-spar

Re: [LAZY] Asking Infra to make all git repos read-write.

> > > BTW, with whiteboard, I won't have time to try this but can one of you? > > http://stackoverflow.com/questions/4114887/is-it-possible-to-do-a-sparse-che > ckout-without-checking-out-the-whole-repository?lq=1 > > I sent out the commands to do a sparse checkout earlier today. It works for me.

Re: [Falcon] building

Unfortunately, I haven't been able to do a hard link to a directory using ln on cygwin, remains mklink which works on windows. -Fred -Message d'origine- From: Frédéric THOMAS Sent: Monday, March 18, 2013 4:24 AM To: dev@flex.apache.org Subject: Re: [Falcon] building I didn't about it

Re: [LAZY] Asking Infra to make all git repos read-write.

What does "fully tested" really mean? If a unix diff doesn't find anything missing, isn't that sufficient? IMO, Falcon is ready to be read/write. Same for flex-sdk and flex-asjs and flex-tlf. But utilities needs another look. BTW, with whiteboard, I won't have time to try this but can one of y

Re: [LAZY] Asking Infra to make all git repos read-write.

It's not fully tested yet, flex-utilities (bad history lenght), flex-whiteboard (not agree), falcon (not fully tested) and I don't know the status of the others. I guess should really take the time to test everything we can before definitly open those repos. -Fred -Message d'origine

Re: [LAZY] Asking Infra to make all git repos read-write.

Did the "external" folder get moved over to git? (Not TLF as an svn:external, there is a folder called "external" at our SVN top-level. Didn't Fred find an issue with "utilities" history? On 3/17/13 8:53 PM, "Om" wrote: > Anyone has an objection to this? If no one objects, I will ask Infra to

Re: Committers - preparing for Git

On 3/17/13 8:49 PM, "Om" wrote: > On Sun, Mar 17, 2013 at 8:37 PM, Alex Harui wrote: > >> So does this mean we picked a SCM technology that does not scale? N-years >> down the road won't all of our other repos have tons of branches and enough >> objects to add up to 240MB? What happens whe

[LAZY] Asking Infra to make all git repos read-write.

Anyone has an objection to this? If no one objects, I will ask Infra to go ahead and make all the git repos read-write. Thanks, Om On Sun, Mar 17, 2013 at 9:54 AM, Erik de Bruin wrote: > +1, INFRA seems to have picked up their game, all should be well (enough) > in git. > > Let's get this over

Re: Committers - preparing for Git

inline answers. -Fred -Message d'origine- From: Alex Harui Sent: Monday, March 18, 2013 4:37 AM To: dev@flex.apache.org Subject: Re: Committers - preparing for Git So does this mean we picked a SCM technology that does not scale? N-years down the road won't all of our other repos hav

Re: Committers - preparing for Git

On Sun, Mar 17, 2013 at 8:37 PM, Alex Harui wrote: > So does this mean we picked a SCM technology that does not scale? N-years > down the road won't all of our other repos have tons of branches and enough > objects to add up to 240MB? What happens when new folks want to get the > code? > If yo

Re: [Falcon] building

If you do whatever git command that pulls from the remote or pushes to the remote, what will happen to files in the linked in directories? I excluded 3.0.33 from being included in git operations. TLF is particular but for others, there's still the possibily to use submodules, I don't like it t

Re: Committers - preparing for Git

So does this mean we picked a SCM technology that does not scale? N-years down the road won't all of our other repos have tons of branches and enough objects to add up to 240MB? What happens when new folks want to get the code? No Git user in the world has a project that large? We must be missi

[jira] [Commented] (FLEX-33418) Error #2044: Unhandled error:. text=PDFError

[ https://issues.apache.org/jira/browse/FLEX-33418?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13604848#comment-13604848 ] Jed Hatamosa commented on FLEX-33418: - By the way im running the pdf inside desktop ai

Re: [Falcon] building

A question about file links: If you do whatever git command that pulls from the remote or pushes to the remote, what will happen to files in the linked in directories? When we finally get falcon production ready, are we going to link it into the SDK? I hope not. Falcon is also being used for Fl

Re: [Falcon] building

I didn't about it before and I still have to check it works but because we've got cygwin, creating a hard link shouldn't be a big deal using ln. -Fred -Message d'origine- From: Alex Harui Sent: Monday, March 18, 2013 4:08 AM To: dev@flex.apache.org Subject: Re: [Falcon] building Some

Re: Git and revision numbers

And you loose all of your local commit history then. I'm not sure if that is ideal or not - guess it depends on how big the change is but that local history log may be useful. Well, you'll still have the details in the logs, well, just found a good reading about the different option [1] have a

Re: Committers - preparing for Git

On Mar 17, 2013 6:44 PM, "Frédéric THOMAS" wrote: >> >> That's problem is solved by applying this policy naming > > > Because in the withboard projects we don't use jira (at least the moment) "bug/jira-#2342" doesn't work, "whiteboard/fthomas" is the same than "fthomas". > It goes beyond that, the

[jira] [Commented] (FLEX-33418) Error #2044: Unhandled error:. text=PDFError

[ https://issues.apache.org/jira/browse/FLEX-33418?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13604845#comment-13604845 ] Jed Hatamosa commented on FLEX-33418: - Hi Patrick Barry, I do have same issue. Right

Re: FXG 2.0 donation progress concerns and Adobe design tool support

Falcon is made up of sub-compilers. There is one for FXG. Just because it spits out AS of a certain format right now doesn't the JS output will be a straight up translation of it. I can be anything. I think Om and I are waiting to see if the HTML5 button with skinning really does have rendering

Re: [Falcon] building

Sometime soon, unless someone else wants to do it, I will rework the build scripts for SDK to pull stuff from TLF without using file links or submodules. I think it is the best way. On 3/17/13 9:17 AM, "Frédéric THOMAS" wrote: Jose, We could decide to add TLF as submodule but we ‘ll have to

Re: Git and revision numbers

Hi, > which I squash a merge time using git merge --squash -m 'new ' newFeature. And you loose all of your local commit history then. I'm not sure if that is ideal or not - guess it depends on how big the change is but that local history log may be useful. Still no idea how we indicate what cha

Re: Committers - preparing for Git

Hi, > Yes, we need to take the one time git to make things a bit more organized. With my current bandwidth that's make it difficult. I'd vote for keeping the whiteboards in SVN. I want to cherry pick what I checkout of that like I could do with SVN and not have to check out everyones work. Just

Re: Git and revision numbers

What I use to do is I commit generaly by JIRA Ticket, it happens I can do more commits but amend them to still have one commit for a specific JIRA Ticket at push time, if my ticket take time and a lot of commits, I create a feature branch which I squash a merge time using git merge --squash -m '

Github mirrors status

Hi, looks like: https://github.com/apache/flex-sdk/tree/develop Isn't tracking change we make to git - any idea what needs to be done to fix it? I assume the other github mirrors are in the same state. Justin

Re: Committers - preparing for Git

On Mar 17, 2013 6:21 PM, "Justin Mclean" wrote: > > Hi, > > > Why is that a problem? > Because you have to initially checkout/download all branches. > > Justin > Yes, we need to take the one time git to make things a bit more organized.

Re: Git and revision numbers

Hi, >> Also how do we identify which GIT check in fixed a JIRA bug? Used to be >> that we just add the SVN revision checkin but now a push can result in >> several checkins. > > I would say indicating the JIRA Ticket id in the commit message In every commit message? You may make a dozen or mor

[jira] [Commented] (FLEX-33427) Missing translation for postcode components

[ https://issues.apache.org/jira/browse/FLEX-33427?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13604838#comment-13604838 ] Justin Mclean commented on FLEX-33427: -- Checked in de_DE, pt_PT, pt_BR and ja_JP.

Re: Git and check in messages

You are very likely to have unstaged changes if your built the SDK That's why I committed a .gitignore file. However it's not possible to run (unlike git push) Only do this on commits that haven’t been pushed an external repository. -Fred -Message d'origine- From: Justin Mclean S

Re: Git and revision numbers

As far as I can see Git has no check in revision numbers. Currently the build release process uses SVN version number I think the number is appended to the RSLs and may be used elsewhere. Anyone have any ideas on how to resolve this? Good question, really, at the moment I don't know, I'm going

Re: Git and check in messages

Hi, > git rebase -i HEAD~4 However it's not possible to run (unlike git push) if you have uncommitted local changes, that makes it less than useful IMO. You are very likely to have unstaged changes if your built the SDK. Justin

Re: Git and check in messages

What about those who are not using the command line but relying on Tower or Git Tortoise? Is that still an option? yep. Who was supposed to to that, I'm not sure, Carlos ? -Fred -Message d'origine- From: Justin Mclean Sent: Monday, March 18, 2013 3:34 AM To: dev@flex.apache.org Su

Re: Git and check in messages

Hi, > Another possibility is to use interactive rebase where you squash files/even > lines of file inside a single commit you there you can even change the commit > orders, use git rebase -i HEAD~4 to interactily change your commits. This really need to be documented on the WIki people who are

Re: Git and check in messages

use git rebase -i HEAD~4 to interactily change your 4 last commits -Message d'origine- From: Frédéric THOMAS Sent: Monday, March 18, 2013 3:30 AM To: dev@flex.apache.org Subject: Re: Git and check in messages Sorry, I meant the usual behavior* Another possibility is to use interactive

Re: Git and check in messages

Sorry, I meant the usual behavior* Another possibility is to use interactive rebase where you squash files/even lines of file inside a single commit you there you can even change the commit orders, use git rebase -i HEAD~4 to interactily change your commits. -Fred -Message d'origine-

Git and revision numbers

Hi, As far as I can see Git has no check in revision numbers. Currently the build release process uses SVN version number I think the number is appended to the RSLs and may be used elsewhere. Anyone have any ideas on how to resolve this? Also how do we identify which GIT check in fixed a JIRA

Re: Git and check in messages

That's usage behavior but there is the possibility to amend your last commit overriding the last message too using git commit --amend -Fred -Message d'origine- From: Justin Mclean Sent: Monday, March 18, 2013 3:19 AM To: dev@flex.apache.org Subject: Git and check in messages Hi, Jus

Re: Git check in

Yes, I checked in the .gitignore file. -Fred -Message d'origine- From: Justin Mclean Sent: Monday, March 18, 2013 3:15 AM To: dev@flex.apache.org Subject: Git check in Hi, I seem to be having trouble checking into git into the Flex SDK repo. Has anyone else been able to push to the

Git and check in messages

Hi, Just noticed that the checkin messages are set up to do one message per (local) commit. Is this the behaviour we want? If you do a large number of small commits and then push you get a lot of emails sent to the list. Thanks, Justin

Re: Git check in

Hi, Never mind - sorted now. Thanks, Justin

Git check in

Hi, I seem to be having trouble checking into git into the Flex SDK repo. Has anyone else been able to push to the repo? Justin

Re: Committers - preparing for Git

That's problem is solved by applying this policy naming Because in the withboard projects we don't use jira (at least the moment) "bug/jira-#2342" doesn't work, "whiteboard/fthomas" is the same than "fthomas". It goes beyond that, the pattern should be , but it's still to risky to go by conve

[jira] [Issue Comment Deleted] (FLEX-33427) Missing translation for postcode components

[ https://issues.apache.org/jira/browse/FLEX-33427?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Justin Mclean updated FLEX-33427: - Comment: was deleted (was: @Marcus; agreed ;-)) > Missing translation for postcode component

Re: Committers - preparing for Git

That's problem is solved by applying this policy naming: master develop whiteboard/fthomas whiteboard/cdutz whiteboard/mclean feature/add-maven-descriptor feature/add-installer-fp-download feature/add-fp-download bug/jira-#2342 ... using nominal branches, which are structured to manage virtual fo

[jira] [Created] (FLEX-33430) Adobe Reader not loading on Flex

Francis Menguito created FLEX-33430: --- Summary: Adobe Reader not loading on Flex Key: FLEX-33430 URL: https://issues.apache.org/jira/browse/FLEX-33430 Project: Apache Flex Issue Type: Bug

Re: Committers - preparing for Git

can really get quickly messy, I wouldn't take this risk. Again, if folks have several projects and those projects several branches as it common using GIT, without strict maming convention, the list of the branches which can grow a lot will mess the people to retreive even their own branch, im

Re: Committers - preparing for Git

Hi, > Why is that a problem? Because you have to initially checkout/download all branches. Justin

Re: Committers - preparing for Git

I think the correct way would be, that each branch is made par on each project wherever experience in addition to the branches of tickets or integration necessary. Thus whiteboard space is confined in each repository. -- Jose Barragan Chief Software Architect Codeoscopic Madrid C/. Infanta Merced

Re: Committers - preparing for Git

On Mar 17, 2013 5:56 PM, "Frédéric THOMAS" wrote: > > Hi, > > I agree with you Justin, each persons branch is a bad pratice, the repo can really get quickly messy, I wouldn't take this risk. Describe 'messy', please. I am not sure what the concern is. > Maybe one repo by person is not feasible,

Re: Committers - preparing for Git

Hi, I agree with you Justin, each persons branch is a bad pratice, the repo can really get quickly messy, I wouldn't take this risk. Maybe one repo by person is not feasible, how do we know without asking ? If it's not feasible, I would stay in SVN too and if I really want to work with GIT, I

Re: Committers - preparing for Git

On Mar 17, 2013 5:12 PM, "Justin Mclean" wrote: > > Hi, > > > I vote for creating a branch for each committer under whiteboard. Anyone > > else want to chime in? > By branch I assume you mean repo not sure if everyone having their own branch make sense as each persons branch would contain differe

Re: Committers - preparing for Git

Hi, > I vote for creating a branch for each committer under whiteboard. Anyone > else want to chime in? By branch I assume you mean repo not sure if everyone having their own branch make sense as each persons branch would contain different files etc etc. But currently using git for the white bo

[jira] [Commented] (FLEX-33427) Missing translation for postcode components

[ https://issues.apache.org/jira/browse/FLEX-33427?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13604809#comment-13604809 ] Fréderic Cox commented on FLEX-33427: - nl_NL (Dutch): invalidCharPostcodeError=De pos

Re: [DISCUSS] Apache Flex SDK Installer 2.5 RC3 (was: [DISCUSS] Apache Flex SDK Installer 2.5 RC2)

Hi, As a reminder, there's an installer which is waiting for your votes. -Fred

Re: [VOTE] Release InstallApacheFlex 2.5.2 - RC3

+1 (Binding) Tested on Windows 7/x64 too. -Fred -Message d'origine- From: Nicholas Kwiatkowski Sent: Sunday, March 17, 2013 5:43 PM To: dev@flex.apache.org Subject: Re: [VOTE] Release InstallApacheFlex 2.5.2 - RC3 +1 Tested on Windows 7/x64. No certificate issues. Package download

Re: Frédéric Thomas joins Apache Flex PMC

Congrats! -Mark On Sat, Mar 16, 2013 at 11:15 AM, Erik de Bruin wrote: > Hi all, > > In recognition for his continued and substantial contributions as a > committer on this project, please join me in welcoming Frédéric Thomas > as the newest member of the Apache Flex PMC [1]. > > EdB > > 1: ht

[jira] [Commented] (FLEX-33427) Missing translation for postcode components

[ https://issues.apache.org/jira/browse/FLEX-33427?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13604781#comment-13604781 ] Justin Mclean commented on FLEX-33427: -- Russian from Elena Geller invalidCharPostcod

Re: Committers - preparing for Git

On Mar 17, 2013 12:06 PM, "Frédéric THOMAS" wrote: > > The 67 Gb I mentioned earlier was wrong, I misread as I said in the next post. > You downloaded 205.68 Mb but, because git re-use the same objects, once the file structure is done, the space taken on the disk at the end is 2,61 Gb. > Well, any

Re: Committers - preparing for Git

The 67 Gb I mentioned earlier was wrong, I misread as I said in the next post. You downloaded 205.68 Mb but, because git re-use the same objects, once the file structure is done, the space taken on the disk at the end is 2,61 Gb. Well, anyway, the 205.68 Mb are normaly downloaded quickly at me, b

Re: Committers - preparing for Git

But it does not download the 67 Gb worth of data as you mentioned earlier. The git repo objects is around 240MB.I suggest you do it as a one time thing and after that use only your directory from the whiteboard. Longer term, maybe each of us could create our own branches under the whiteboard

Re: Committers - preparing for Git

The git remote fetch operation you do will get the entire repo, you then apply a filter using the .git/info/sparse-checkout and you'll see only your directory but first you fetched the entire repo, with your bandwith that's ok, not with mine, while you are at 624 KiB/s I'm at 20 KiB/s -Fred

Re: Committers - preparing for Git

In git bash: omuppira@OMUPPIRA /c/p/flex_os/workspace/flexroot/git/test *$ git init* Initialized empty Git repository in c:/p/flex_os/workspace/flexroot/git/test/.git/ omuppira@OMUPPIRA /c/p/flex_os/workspace/flexroot/git/test (master) *$ git remote add -f origin https://git-wip-us.apache.org/rep

Re: Committers - preparing for Git

It does seems to work for me, can you please give me the commands you used. -Fred -Message d'origine- From: Om Sent: Sunday, March 17, 2013 6:04 PM To: dev@flex.apache.org Subject: Re: Committers - preparing for Git On Sun, Mar 17, 2013 at 9:33 AM, Frédéric THOMAS wrote: Well, fro

Re: Committers - preparing for Git

On Sun, Mar 17, 2013 at 9:33 AM, Frédéric THOMAS wrote: > Well, from what I'm reading about spare checkouts, I need to clone the > entire repo first, and only after, I can indicate that only this/those > tree(s) interest me, but I can't do it before it is already cloned. > > Do you know something

Re: Committers - preparing for Git

flex-utilities: - History -> FAIL (back to 2012-07-12 for svn, 2012-08-15 for git) - Build/run Installer -> OK -Fred -Message d'origine- From: Frédéric THOMAS Sent: Sunday, March 17, 2013 5:33 PM To: dev@flex.apache.org Subject: Re: Committers - preparing for Git Well, from what I'm

Re: Git migration update - Falon is ready

+1, INFRA seems to have picked up their game, all should be well (enough) in git. Let's get this over with! EdB -- Ix Multimedia Software Jan Luykenstraat 27 3521 VB Utrecht T. 06-51952295 I. www.ixsoftware.nl (http://www.ixsoftware.nl) On Sunday, March 17, 2013 at 16:16, Alex Harui wro

Re: TLF Tables

I'm not sure how well attachments work for this list, but I'm going to try. I found this file in TLF which SEEMS to spell out what's left to do with the table support. I have not yet totally wrapped my head around the TLF architecture. I expect to get there though… ;-) Here's the interesting pa

Re: [Falcon] building

Ok Fréd, This reasoning seems correct, since it is less dramatic add a symbolic link at that point to change the scripts and also manage a git submodule, just for that. Well ... It was just a suggestion :) Thanks, -- Jose Barragan Chief Software Architect Codeoscopic Madrid C/. Infanta Mercede

Re: [VOTE] Release InstallApacheFlex 2.5.2 - RC3

+1 Tested on Windows 7/x64. No certificate issues. Package downloads, installs and tests fine. Good work guys! -Nick On Sat, Mar 16, 2013 at 12:09 AM, Frédéric THOMAS wrote: > *Issues addressed in this Release Candidate:* > > 1. Enable Flex SDK download stats tracking > 2. > https://issue

Re: Committers - preparing for Git

Well, from what I'm reading about spare checkouts, I need to clone the entire repo first, and only after, I can indicate that only this/those tree(s) interest me, but I can't do it before it is already cloned. Do you know something else about spares I didn't get ? -Fred -Message d'origine

Re: Committers - preparing for Git

Hi Om, Good morning so :) Well, not yet, actually, I was look for a such feature, I'm trying it right now. Thanks for that tip, -Fred -Message d'origine- From: Om Sent: Sunday, March 17, 2013 5:01 PM To: dev@flex.apache.org Subject: Re: Committers - preparing for Git On Mar 17, 20

Re: [Falcon] building

Jose, We could decide to add TLF as submodule but we ‘ll have to change the build files in order to tell them to go down of 1 level to pick up the textLayout sub directory as source, in more, at the moment, this directory is picked up from the TLF version, all in all, that would be possible but

Re: Committers - preparing for Git

On Mar 17, 2013 8:43 AM, "Frédéric THOMAS" wrote: > > @Om, > > At the end, the flex-withboard repo is 2,61 Gb on disk, it took more than 2:30 to download, I really don't think that's something acceptable, would it be possible to split the repos, one by committer ? > Can you check that with David ?

Re: Committers - preparing for Git

Otherwize, the history is good but the structure seems alterated, just right after I cloned the remote repo, I did a git status [1], it should have returned "nothing to commit, working directory clean" -Fred U:\sources\asf\flex\flex-whiteboard>git status # On branch master # Changes not staged

Re: Committers - preparing for Git

@Om, At the end, the flex-withboard repo is 2,61 Gb on disk, it took more than 2:30 to download, I really don't think that's something acceptable, would it be possible to split the repos, one by committer ? Can you check that with David ? -Fred -Message d'origine- From: Michael Schm

Re: [Falcon] building

Hi Jose, We can't, the submodule feature takes a entire repo, not a sub directory inside a repo. -Fred -Message d'origine- From: Jose Barragan Sent: Sunday, March 17, 2013 4:25 PM To: dev@flex.apache.org Subject: Re: [Falcon] building Ops... sorry for sintax :P Would not it be bet

Re: [Falcon] building

Ops... sorry for sintax :P Would not it be better, add the TLF repository as a submodule GIT, at that point the SDK? -- Jose Barragan Software Architect Chief Codeoscopic Madrid C/. Infanta Mercedes, 92. Planta 5. 505. 28020 Madrid. Tel.: +34 912 94 80 80 On Mar 17, 2013, at 4:23 PM, Jose Bar

Re: [Falcon] building

Would not it be better, add the TLF repository as a submodule GIT SDK at that point? -- Jose Barragan Software Architect Chief Codeoscopic Madrid C/. Infanta Mercedes, 92. Planta 5. 505. 28020 Madrid. Tel.: +34 912 94 80 80 On Mar 17, 2013, at 4:06 PM, Frédéric THOMAS wrote: > For TLF, the p

Re: Git migration update - Falon is ready

I did a unix diff on compiler.js and checked the history of one file. That should be good enough. I will trust that there is a develop branch in asjs and I will eventually figure out how to get it. Let's open everything up in git. Even the whiteboards. Anything missing can be hand migrated. O

Re: TLF Tables

And falcon its master, not like there is anything else. Mike Quoting Harbs : Thanks. But TLF does not have a develop branch? On Mar 17, 2013, at 5:11 PM, Frédéric THOMAS wrote: Hi Harbs, You should consider to checkout only the master branch to have the last code on this repo, for the f

Re: TLF Tables

Thanks. But TLF does not have a develop branch… On Mar 17, 2013, at 5:11 PM, Frédéric THOMAS wrote: > Hi Harbs, > > You should consider to checkout only the master branch to have the last code > on this repo, for the flex-sdk, that's the develop branch. > > -Fred > > -Message d'origine--

Re: TLF Tables

Hi Harbs, You should consider to checkout only the master branch to have the last code on this repo, for the flex-sdk, that's the develop branch. -Fred -Message d'origine- From: Harbs Sent: Sunday, March 17, 2013 4:08 PM To: dev@flex.apache.org Subject: Re: TLF Tables I'm trying, b

Re: [Falcon] building

Ok, Well I have no idea and don't feel like learning anything new at the moment. :) Things work fine right now for what I am doing. I will wait until it gets sorted out. Mike Quoting Frédéric THOMAS : For TLF, the point is that with svn, we used svnexternal to link the TLF/textLayout sub

Re: TLF Tables

I'm trying, but I'm having a heck of a time wrapping my head around git… :-( TLF has the following branches: 1, 1.1, 2, 2.1, 3.0, HEAD and Master. Where would I look for the code? (not that I've figured out the difference between Checkout, Fetch and Pull yet…) Harbs On Mar 17, 2013, at 4:57

Re: [Falcon] building

For TLF, the point is that with svn, we used svnexternal to link the TLF/textLayout sub directory to frameworks\projects\textLayout\3.0.33, this feature doesn't exist with git, so, at the moment I go by a hard link, on my windows 7, I use Link Shell Extension [1] -Fred [1] http://schinagl.pri

Re: [Falcon] building

I didn't notice, maybe there is. If there is then great. :) I think now that I have looked with word wrap below, the FlashPlayer.exe is not getting copied to the temp dir when running the falcon tests. It seems only the functional tests are failing and that is why. I can build falcon fin

Re: [Falcon] building

If you only want the jar and not run tests, I thought there was a sub-task you could run. On 3/17/13 7:59 AM, "Michael Schmalle" wrote: > Here is a stack trace when running the build of compiler.tests > > > > > [junit] 2086 bytes written to > C:\Users\Work\Documents\git-apache\flex-falc

Re: [Falcon] building

Here is a stack trace when running the build of compiler.tests [junit] 2086 bytes written to C:\Users\Work\Documents\git-apache\flex-falcon\compiler.tests\temp\MXMLPropertyStringTests1425684253960474506.swf in 0.305 seconds [junit] java.io.IOException: Cannot run program "C:\Users

Re: TLF Tables

I'd scan the source code for "Table" and see what you find. On 3/17/13 6:22 AM, "Harbs" wrote: > There were a couple of bugs in the Adobe JIRA on table support in TLF [1] [2] > These were transferred over to Apache JIRA here. [3] [4] > > There have been inferences that it was supposed to be av

Re: [Falcon] building

On this note; I have done my duty... I can confirm all tests pass int eh falcon.jx.tests except the 4 that depend on flexjs. So I'm good to go other than the text layout barf. Falcon was successfully built as well after my half ass attempt at the sdk. All the tests that don't depend on swf

Re: Namespaces in FXG

I don't think you need XSLT to filter out certain namespaces. But it is up to you. On 3/17/13 2:57 AM, "Sebastian Mohr" wrote: > Thanks Alex, > > I have had a look for XSLT with AIR. Is > that the way [1] you whould propose it? > > [1] http://blogs.adobe.com/briggs/2008/05/using_air_for_xslt

  1   2   >