Return-Path: X-Original-To: apmail-incubator-flex-dev-archive@minotaur.apache.org Delivered-To: apmail-incubator-flex-dev-archive@minotaur.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 37FDDCFCA for ; Tue, 22 May 2012 12:41:40 +0000 (UTC) Received: (qmail 15895 invoked by uid 500); 22 May 2012 12:41:39 -0000 Delivered-To: apmail-incubator-flex-dev-archive@incubator.apache.org Received: (qmail 15831 invoked by uid 500); 22 May 2012 12:41:39 -0000 Mailing-List: contact flex-dev-help@incubator.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: flex-dev@incubator.apache.org Delivered-To: mailing list flex-dev@incubator.apache.org Received: (qmail 15822 invoked by uid 99); 22 May 2012 12:41:39 -0000 Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 22 May 2012 12:41:39 +0000 X-ASF-Spam-Status: No, hits=1.7 required=5.0 tests=FRT_ADOBE2,RCVD_IN_DNSWL_NONE,SPF_HELO_PASS,SPF_NEUTRAL X-Spam-Check-By: apache.org Received-SPF: neutral (nike.apache.org: local policy) Received: from [80.67.31.93] (HELO smtprelay05.ispgateway.de) (80.67.31.93) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 22 May 2012 12:41:32 +0000 Received: from [10.128.0.19] (helo=exchange.df.eu) by smtprelay05.ispgateway.de with esmtps (TLSv1:RC4-MD5:128) (Exim 4.68) (envelope-from ) id 1SWoOl-0000wj-RD for flex-dev@incubator.apache.org; Tue, 22 May 2012 14:41:11 +0200 Received: from ECCR06PUBLIC.exchange.local ([10.128.2.56]) by efe06.exchange.local ([10.128.0.19]) with mapi; Tue, 22 May 2012 14:41:11 +0200 From: "christofer.dutz@c-ware.de" To: "'flex-dev@incubator.apache.org'" Date: Tue, 22 May 2012 14:41:10 +0200 Subject: AW: AW: AW: Library Versions used in Flex SDK Thread-Topic: AW: AW: Library Versions used in Flex SDK Thread-Index: Ac00+SOpIuLkE5WnRsmregOJEexCBgAAMbdQAAV2oMcALKsswAAHEUcBABs0siA= Message-ID: <66E38C42347D6446BF7FCB22C3D3878072C670A974@ECCR06PUBLIC.exchange.local> References: <66E38C42347D6446BF7FCB22C3D3878072C5549B45@ECCR06PUBLIC.exchange.local> In-Reply-To: Accept-Language: de-DE, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: de-DE, en-US Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 I allready feared that there could be problems with legal issues in distrib= uting the runtimes. I guess the only solution would be do publish the tool = to enable people to generate their own SDK versions (including the flash pl= ayers and playerglobals). Flexmojos is not only a link between Flex and Maven, but integrates several= other tools, that make it extremely valuable: - Apparat Integration to optimize the SWFs and SWCs - GraniteDS to automatically generate the ActionScript model classes from t= heir Java counterparts. So simply integrating the basic compiler feature would strip a lot of it's = functionality. It would be great however if the compiler would could be cal= led directly instead of having to go the way of simulating the commandline = call.=20 I know that Flexmojos still has quite some issues, but most of the Issues I= had to resolve were mainly related to problems in the Flex compiler code (= Just to mention one: Try using ASDoc with code comments containing the "@" = character ;-) ). I toally agree that the general perspective should be to i= ntegrate the functionality of flexmojos into Flex, or at least have them ma= intained by the same group, but I think the current SDK has far too many is= sues that should be adressed first. Trying to create a new Maven integratio= n layer would just add more to the todo list. So for now I think it would b= e great to work together. I would be glad to contribute to make Flex work b= etter. As soon as the main issues are resolved I would also be glad to help= you create something similar or to integrate flexmojos to flex. I'm also not requesting to change flex in any way to make mavenizing the sd= k easier. Currently I was only suggesting to clean up the structure of the = SDK to more reflect the parts natures and to generate some helpfull other s= tuff that makes life of people using it a lot easier: - Generate a special pom containing only dependencyManagement section. This= can be imported into other maven projects to automatically set the version= s of the sdk artifacts. - Generate a special jar containing metadata that can be used by flexmojos = to perform validation (minimum SFW-version, minimum playerglobal version, e= tc.). Just to name a few. Chris -----Urspr=FCngliche Nachricht----- Von: Alex Harui [mailto:aharui@adobe.com]=20 Gesendet: Samstag, 19. Mai 2012 18:45 An: flex-dev@incubator.apache.org Betreff: Re: AW: AW: Library Versions used in Flex SDK On 5/19/12 6:39 AM, "christofer.dutz@c-ware.de" wrote: > Currently Flex doesn't work at all with Maven. Even if you distribute=20 > the SDKs content as Maven artifacts, this is by far not enough as=20 > Maven simply doesn't know what to do with a Flex/Air project. >=20 > Flexmojos is a set of maven plugins that actually tells Maven what to=20 > do with a Maven project of packaging "swf" or "swc" and how a "war"=20 > has to be built to contain a Flex application. >=20 > Currently those plugins act as a wrapper to the Flex compilers,=20 > optimizers and runntimes so you don't have to wory about how your=20 > Maven dependencies get passed to the Flex compiler. > So I don't think that Flexmojos is a dead End, it's more the missing=20 > link to bringing Maven and Flex together. I understand (I think). This list is about Apache Flex and to me, lots of = packaging changes are now possible, if not required. For example, Apache F= lex cannot ship the same folder structure as Adobe does because it doesn't = have the right to distribute the Flash/AIR pieces. Apache Flex also now ha= s control over the compiler. In my mind, FlexMojos is a mapping between Ma= ven and the Adobe Flex SDKs and I've heard it has some issues. Given all t= hat, I am just raising the question of whether it is better to continue on = with FlexMojos which is currently an external dependency, or whether it is = better to adjust packaging and compilers and whatever to make a simpler map= ping possible and just start over. And then all the pieces are within Apac= he. >=20 > Up till now Velo (The creator of Flexmojos) had to vonvert every Flex=20 > SDK in a mavenized form, and deployed that on the Sonatype Maven repo.=20 > Now I have taken over to continue flexmojos development and as one of=20 > my first tasks I wanted to publish all of those patched SDKs Adobe=20 > released a few months ago. While I am at it, I'm refactoring the=20 > structure of the SDKs to a structure that relates better to the=20 > structure oft he Products (Flashplayer and Air runtime (together with pla= yerglobal and airglobel) are not part of the SDK. I think that's great that you are doing that. At least in February, a coup= le of large companies were in dire need of getting those patch SDKs deploye= d and needed Maven artifacts. That effort is outside of the Apache Flex sc= ope so you can do whatever you want there (assuming you have legal rights t= o redistribute the Adobe pieces). I'm just looking for your input on how t= hings could be better in Apache Flex as now is an important time as we are = deciding on our release packaging. >=20 > I guess as Maven is an Apache project and now Flex is an Apache=20 > project, it would make sense if Apache provided the SDKs as a=20 > donwloadable sdk the same way Adobe did, but would also provide the=20 > official releases in the Maven Central repository. > The structure of these SDKs however would be the "API" Flexmojos would=20 > have to rely on. That's why I'm asking you guys here about your=20 > Mavenizing plans so I don't have to generate the SDKs and adjust=20 > Flexmojos to that new structure and then do all of that again as soon=20 > as the Flex project starts distributing your SDKs. This would result=20 > in 3 structural different sets of Flex SDKs deployed in Nexuses/Artifacto= ries/etc. all over the planet. I would like to avoid this. >=20 > As a suggestion, I could provide you with the SDKs I generated and we=20 > could optimize them together. Then you could simply use the tool I=20 > created to officially distribute the new SDKs from then on. That would be a useful data point, but I really want to make sure some aspe= ct of Apache Flex shouldn't be changing to make Maven-izing easier to do an= d maintain. >=20 -- Alex Harui Flex SDK Team Adobe Systems, Inc. http://blogs.adobe.com/aharui