Re: [swfmill] swfmill needs your help.
I am not attempting to get off topic here, but I just thought I would mention to those interested in an IDE based on swfmill and mtasc the existence of flashdevelop. I believe that it is open source, and written in c# (unfortunately not cross platform). It has integration for both the mtasc/swfmill combo as well as using the new Adobe command line tools for flex2 development. Dan Shryock On 1/9/07, Marc Reichelt [EMAIL PROTECTED] wrote: Hi Dan, I'm using swfmill since around a year now, and I'm using it for two projects. One that I do for university is nearly finished, and it's nice to see how great the tools MTASC and swfmill work together. The biggest job is done by MTASC here (I'm not using any complex functions of swfmill right now, but big ActionScript 2.0 files), but as I need some complex movie clips from other *.swf Files (which do only work since 0.2.11.15 or so) I'm fixed to use the latest prereleases. The project does work with 0.2.11.22 as it should, no problems here. Finally, because I have a valid Flash MX license (yes, I really spent money on it), I could create some *.swf files containing complex movie clips and test if they can be imported by swfmill. [...] So. I learned a lesson: Dont do open-source for the money. It took me a long time to create a senseful connection between open source and money. The question I constantly asked myself was: How can projects like Apache, Linux, Mozilla etc. can be so much better than proprietary products? The companies spent much more money on proprietary products, so how can this be? The final answer I found a few months ago was that my question was wrong: The big open source projects earn much more money than proprietary ones - the only difference here is that _many_ companies (and private persons) spend a small amount of money and working time (which is money!) on open source projects. Overall there is more money spent on big open source projects. The conclusion is: Open Source projects work well if there is a big community to support them. That means that swfmill has to be used more often in order to earn more money with it. I don't think that there will be many donations more, but at least if swfmill is used more often companies will realise that they need additional functions, and they need a person that will implement those functions for them. And who is better on swfmill: Is it a programmer that has to spend much time to learn how swfmill works until he can modify it, or is it you, Dan? ;) That's how open source and money work together. * test the latest prerelease (0.2.11.22), and *report if it works*. As the useOutlines issue showed, my fear of introducing regressions is not completely psychological. I want to see at least about 5 reports of i use swfmill in this-or-that-context, and 0.2.11.22 works flawless before finally going 0.2.12. If you want to help more, start organizing the construction of an extended set of simple test cases for automatic regression tests. As I wrote above, 0.2.11.22 works fine here. * swfmill needs documentation. there is a lot floating around, but it needs a loving hand of collecting and organizing it into something like a manual. Please, someone, step forward and take on this issue. It could be one weekend of work, but would make an incredible difference and you could be certain of the thanks of many a newbie. I'm interested in it, but right now I have to work hard in university. What would be the best method for the documentation? A HTML page? I would appreciate that one. :) when and if swfmill gets a release and some better docs, it is back on the road to being a stable tool. is that in your interest? Yes, of course. Or what did you expect? ;) I won't wait for the documentation to be done before the release of 0.2.12 - just release it, the documentation will be done afterwards. I'm really interested in the idea of Brian who noticed that an IDE would be useful. As you wrote, swfmill is really stable now - why not creating an IDE (e.g. written in Java, which will be open-sourced in a few weeks) that uses MTASC for ActionScript code and swfmill to glue all together? I'm thinking of an IDE that looks like the official IDE by Adobe. That would have many advantages: - There is an IDE for users who don't like commandline tools - The community is getting bigger (bigger community = more money) - The development of swfmill and MTASC continues, as they are a requirement for the IDE I'm not the one to start this IDE because I have some other projects which are waiting to be developed for a long time, but IMHO it's a great idea. Regards Marc Reichelt || http://www.marcreichelt.de/ PS: Sorry for the long mail and wasting your time. ;) ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org
Re: [swfmill] Generating a SWF from a XML String
If you look at the command line options, there is an option that lets you send the data on the stdin stream for the process, which you should be able to use from most programming languages. Dan On 12/1/06, David Hassoun [EMAIL PROTECTED] wrote: Is there a command to pass a xml string to swfMill instead of a pointer to a XML flat file? I want to make an app that I can automatically feed swfmill the xml to generate a swf from without having to write the xml to a file first. Thanks. David. Access over 1 million songs - Yahoo! Music Unlimited. ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org
Re: [swfmill] [help needed] release script for OSX universal binaries
If I recall correctly, a universal binary isn't actually a single file, it is multiple executable files stored in side of an app bundle under Contents/MacOS/ and then referenced in the Contents/Info.plist. I think the first place to look is at the format of Info.plist for app bundles, that should tell you some info on what meta is needed to consider something a universal binary. Oh, and if that is the case, it would almost be pointless becase an app bundle is specifically for when an application has a gui, which would mean there is no point for swfmill. Hope this helps. DanOn 8/25/06, Mark Winterhalder [EMAIL PROTECTED] wrote: On 8/25/06, Steve Webster [EMAIL PROTECTED] wrote: My new employer have blessed me with a new Intel MacBook Pro (dodgy, explosive battery included free of charge) so I can take a look at trying to get an intel build out.Our designer just had his MBP recalled from the vendor, for a newmotherboard due to unspecified reasons. His battery isn't one of theaffected ones, though. As for how universal binaries are produced I have no idea; it can't possibly be as simple as passing an extra --arch switch to gcc, as suggested by the top few results in google.I tried your new employer's search engine (for once, normally I go with the market leader because I already know their search operators)and a link brought me here: http://developer.apple.com/documentation/Porting/Conceptual/PortingUnix/compiling/chapter_4_section_3.htmlCongrats for the new job!Mark___swfmill mailing list swfmill@osflash.orghttp://osflash.org/mailman/listinfo/swfmill_osflash.org ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org
Re: [swfmill] [help needed] release script for OSX universal binaries
Actually it looks like I am wrong, according to this page: http://developer.apple.com/documentation/MacOSX/Conceptual/universal_binary/universal_binary_compiling/chapter_2_section_7.html#//apple_ref/doc/uid/TP40002217-CH206-240191 you can specify two different arch flags to gcc 4.0 like the following: -arch ppc -arch i386 Sorry if I caused any confusion. Dan On 8/25/06, Dan Shryock [EMAIL PROTECTED] wrote: If I recall correctly, a universal binary isn't actually a single file, it is multiple executable files stored in side of an app bundle under Contents/MacOS/ and then referenced in the Contents/Info.plist. I think the first place to look is at the format of Info.plist for app bundles, that should tell you some info on what meta is needed to consider something a universal binary. Oh, and if that is the case, it would almost be pointless becase an app bundle is specifically for when an application has a gui, which would mean there is no point for swfmill. Hope this helps. Dan On 8/25/06, Mark Winterhalder [EMAIL PROTECTED] wrote: On 8/25/06, Steve Webster [EMAIL PROTECTED] wrote: My new employer have blessed me with a new Intel MacBook Pro (dodgy, explosive battery included free of charge) so I can take a look at trying to get an intel build out. Our designer just had his MBP recalled from the vendor, for a new motherboard due to unspecified reasons. His battery isn't one of the affected ones, though. As for how universal binaries are produced I have no idea; it can't possibly be as simple as passing an extra --arch switch to gcc, as suggested by the top few results in google. I tried your new employer's search engine (for once, normally I go with the market leader because I already know their search operators) and a link brought me here: http://developer.apple.com/documentation/Porting/Conceptual/PortingUnix/compiling/chapter_4_section_3.html Congrats for the new job! Mark ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org
[swfmill] Possible on svg import and generated movieclip properties.
I am using swfmill to import an svg into my swf, and everything is working perfectly, except some of the properties of the generated movie clips. I am importing a map of the United States, and each state is broken up into a separate group or path. When I check the properties of each state inside of the instance1 object, their width and height match the actual drawn width and height. But when I check the x and y properties, they always report at 0. This wouldn't be a huge issue, except that I am creating mouseovers which zoom each state outwards by adjusting the _xscale and _yscale, and when the zooming occurs, it appears to be including the offset from left and top in the zoom, meaning that each state moves down and to the right (and it gets excessively worse as I move to the eastern states). If the width reflected the offset, or the x and y were the offset, I could account for the difference and move it where it belongs. I have tried drawing some simple shapes in MMC, and their properties appear to work as I would expect. This leads me to believe that the issue occurs when swfmill imports the svg, and I was wondering if there would be any way to get this fixed, or suggestions to work around the issue. Any help is greatly appreciated. DanPS. swfmill is awesome, it made it possible to automate the generation of the map and actions for each state saving me many hours of work, thanks for all of the hard work! ___ swfmill mailing list swfmill@osflash.org http://osflash.org/mailman/listinfo/swfmill_osflash.org
