I totally agree with the reasoning behind the longer compile times for streamlined build process that take quality control into account. I just wanted to make sure I wasn't losing it. For the life of me, I was stuck on thinking that speed was the main factor in the reasons to use mxmlc. Thanks for helping me get over it, lol.
I really like the Flex 2 build/debug as you go capability. But you are right there needs to be a good process in place to handle tests and compiling issues for large applications, and using build scripts and mxmlc are perfect way to (semi-)automate it. Renaun --- In [email protected], "Johannes Nel" <[EMAIL PROTECTED]> wrote: > > >>My question then would be, is the speed of compilation a main factor > >>in the choice of compilation methods? > > well as i orignally stated the appserver approach is the fastest, so if its > speed you are after, yeah go for it. some of the larger projects i have > worked on has taken up to a 1:30 but thats rare. use of swc's and the like > can speed things like this up substantially. > > the fact of the matter (IMO at least is) that i would take much slower > compilation speed for a streamlined process that highlight bugs quicker > (unit tests for example) and the flexibility that ant gives you, since in > the long run you save time > > > > On 3/22/06, Renaun Erickson <[EMAIL PROTECTED]> wrote: > > > > The use of the multiple targeted build scripts (unittests, app, or > > all) makes sense. > > > > My question then would be, is the speed of compilation a main factor > > in the choice of compilation methods? > > > > It sounds like there are other factors that depend on the development > > teams practices that weigh in the to decision then just speed of > > compilation. > > > > In regards to just the factor of speed, on an application with roughly > > 50 .mxml and 50+ .as files whats the typical compile times? 20s-30s? > > > > Renaun > > > > --- In [email protected], "Johannes Nel" <johannes.nel@> > > wrote: > > > > > > this is both for develpers (developer chooses which target to compile - > > > unitests,app or all) and our continues integration server. we only > > deploy a > > > swf to production. > > > > > > On 3/22/06, Renaun Erickson <renaun@> wrote: > > > > > > > > Thanks Johannes for insight. > > > > > > > > Is this what you do as a developer on a dev box? Or only on the major > > > > builds on the deployment (staging/production) servers? > > > > > > > > Renaun > > > > > > > > --- In [email protected], "Johannes Nel" <johannes.nel@> > > > > wrote: > > > > > > > > > > hear hear on what david says. using ant we > > > > > update from svn > > > > > build unit tests app using mxmlc, if all tests pass > > > > > use mxmlc to build app > > > > > > > > > > so yeah it has its benefits :) > > > > > > > > > > so it is critical to the process as > > > > > On 3/22/06, Dave Wolf <gatorj24@> wrote: > > > > > > > > > > > > That depends what you mean by better. In Flex 1.5 you will > > find very > > > > > > quickly that the JIT server compiler leaks memory very badly > > and wont > > > > > > work with an application of any real size. For any kind of > > serious > > > > > > application the server compilation simply wont work for you. > > > > > > > > > > > > We use mxmlc called from autoamted ANT builds. We're pretty > > attament > > > > > > about it all. We have a very well designed build and source > > control > > > > > > process that has made real impacts on the quality and > > reliability of > > > > > > our solutions. > > > > > > > > > > > > Yeah, performance isnt stunning, but given that its a two pass > > > > > > compilation thats not shocking either. > > > > > > > > > > > > We're hoping for some marked improvement in Flex2. > > > > > > > > > > > > -- > > > > > > Dave Wolf > > > > > > Cynergy Systems, Inc. > > > > > > Macromedia Flex Alliance Partner > > > > > > http://www.cynergysystems.com > > > > > > > > > > > > Email: dave.wolf@ > > > > > > Office: 866-CYNERGY > > > > > > > > > > > > > > > > > > > > > > > > --- In [email protected], "Renaun Erickson" <renaun@> > > wrote: > > > > > > > > > > > > > > After some more reading I am under the assumption that for > > > > development > > > > > > > of Flex 1.5 apps its better to use the Java web server access to > > > > > > > compile with caching on. Is this correct? > > > > > > > > > > > > > > Renaun > > > > > > > > > > > > > > --- In [email protected], "Renaun Erickson" <renaun@> > > > > wrote: > > > > > > > > > > > > > > > > ## This post is in regard to Flex 1.5 ## > > > > > > > > > > > > > > > > I have finally got around to trying out mxmlc and fastmxmlc > > > > with a xml > > > > > > > > build script. The purpose was to see if it was faster to use > > > > mxmlc to > > > > > > > > compile on my development environment versus using the web > > server > > > > > > > > (accessing the mxml by a url). > > > > > > > > > > > > > > > > I am not seeing a big difference in compile time between > > the two > > > > > > > > methods. I tried RSL's but dont think they are setup > > correctly. > > > > > > > > > > > > > > > > My big question is how do I get my compile times down on > > my local > > > > > > > > development machine? > > > > > > > > > > > > > > > > Renaun > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > > Flexcoders Mailing List > > > > > > FAQ: > > http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt > > > > > > Search Archives: > > > > http://www.mail-archive.com/flexcoders%40yahoogroups.com > > > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > j:pn > > > > > http://www.lennel.org > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > Flexcoders Mailing List > > > > FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt > > > > Search Archives: > > http://www.mail-archive.com/flexcoders%40yahoogroups.com > > > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > j:pn > > > http://www.lennel.org > > > > > > > > > > > > > > > > > -- > > Flexcoders Mailing List > > FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt > > Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com > > Yahoo! Groups Links > > > > > > > > > > > > > > > > > -- > j:pn > http://www.lennel.org > -- Flexcoders Mailing List FAQ: http://groups.yahoo.com/group/flexcoders/files/flexcodersFAQ.txt Search Archives: http://www.mail-archive.com/flexcoders%40yahoogroups.com Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/flexcoders/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
