So, the compile-with-file-name issue turned out to be a little more complicated than my first assessment. I've added a GwtCompilerTask for Ant that accepts a module file path, but it also needs to know the source root. I think I could get around that by starting with the module file name and prepending elements from the absolute path until I find something that exists on the classpath. That'll wait for another day, since what I have works for now.
The build.demos target and GwtCompilerTask are available in r1130. On Sat, Oct 11, 2008 at 12:17 PM, Emily Crutcher <[EMAIL PROTECTED]> wrote: > That would be awesome! The zip isn't even necessary, as long as the output > represents a predictable static file structure. > > My guess is that 2970 will probably not be implemented right away. You're > right about ant, so if you like python or want to create a specialized ant > task, that would work. > > In fact, the script I'm using to upload static content to the Google app > engine is already in python, so I'm going to have to deal with python/ant > integration anyway. > > Also, in an ideal world we'd some day also generate a master index page as > well, so having the base demo-generator be in an expressive language > definitely has some advantages. > > Thanks a million! > > Emily > > > > > On Sat, Oct 11, 2008 at 11:50 AM, Isaac Truett <[EMAIL PROTECTED]> wrote: >> >> > Some sort of ant/python/shell/ script that crawls through the new >> > src-demo and automatically >> > builds all the demos there. We could then use that to easily publish >> > on google app engine >> > up-to-date demos for all incubator widgets before each public drop. >> >> I'll volunteer for that. This is pretty easy to do in an Ant script. >> Building the demos will be trivial if issue #2970 is accepted and >> implemented. I'm not familiar with deploying to GAE. Would a zip file of all >> the compiled demos, plus a simple index.html, suffice? >> >> >> On Fri, Oct 10, 2008 at 6:02 PM, Emily Crutcher <[EMAIL PROTECTED]> wrote: >>> >>> I actually tend to agree with Authur here, that in gwt-incubator we have >>> widgets at different stages in the pipeline and while some we don't want >>> users to touch, others are much more stable, and would benefit from wider >>> user distribution and are at least as good, or better, then almost any other >>> available gwt library. >>> >>> As widgets are moved into gen2, they are being all given default style >>> sheets though improvements are always appreciated. >>> >>> Internally, we've started to toy with the idea of labeling >>> widgets/libraries: Here is the initial proposal >>> >>> @ReleaseCandidate: This widget has been slotted for a GWT release. Its >>> bug reports are taken as seriously as any normal gwt widgets and its API is >>> less likely to change. >>> @Beta: This is a widget intended for wide distribution and is being >>> actively supported by at least one developer. The code should be reasonably >>> bug free as well. >>> @DoNotUse or no tag: Don't use, is either in the process of being dropped >>> from incubator or is still in alpha. >>> >>> >>> Also, here is my wish list of things I'd love to get volunteers for: >>> >>> A maven repository to pick up the most recent gwt-incubator + gwt-trunk >>> milestone build. If possible, would also create a nightly build as well. >>> Some sort of ant/python/shell/ script that crawls through the new >>> src-demo and automatically builds all the demos there. We could then use >>> that to easily publish on google app engine up-to-date demos for all >>> incubator widgets before each public drop. >>> Someone to go through the documentation, fix what can be easily fixed, >>> and then e-mail out those widgets/libraries whose documentation need serious >>> attention. >>> Also, someone who would be willing to go through the current demos and >>> improve them/flag developers if they are missing and/or just suck. >>> Finally, if people could nominate widgets they would like to see leave >>> incubator, that would be very helpful as well! >>> >>> >>> >>> On Fri, Oct 10, 2008 at 4:57 PM, Arthur Kalmenson <[EMAIL PROTECTED]> >>> wrote: >>>> >>>> OK that makes sense. Maybe we need a project that's in between >>>> incubator and GWT, something that has regular releases and uses >>>> polished content from the incubator but content that's not so polished >>>> that it would go into GWT. While what's in the incubator is a work in >>>> progress, it is still a) much faster then what the other libraries >>>> offer, b) mostly cleaner and better written then the other libraries, >>>> c) gives people an idea of where GWT is headed. >>>> >>>> While making custom widgets is easy in GWT, there is a lot of overlap >>>> in what people need and there is usually a rich suite of widgets that >>>> people would like out of the box. It's not very easy to pick up GWT >>>> and dive in making great and interactive apps. There's a lot of group >>>> work that needs to be done to build rich widgets. It seems a waste >>>> that this group work is done on every GWT project by all the various >>>> organizations that use GWT. I know that the GWT team said GWT was >>>> meant to be pretty low level and they wanted the community to build on >>>> top of that, but it seems that the community is not delivering. I know >>>> Bruce mentioned that better widgets are on the agenda, so I don't >>>> know.... >>>> >>>> Regards, >>>> Arthur Kalmenson >>>> >>>> On Oct 10, 4:36 pm, "Isaac Truett" <[EMAIL PROTECTED]> wrote: >>>> > We don't want people to be afraid of the Incubator, but we do want >>>> > them to >>>> > be cautious. It isn't a library per se, but a workshop for ideas. >>>> > Things in >>>> > the Incubator may be half finished or in the middle of refurbishing at >>>> > any >>>> > given time. It's an environment where you may have to get your hands >>>> > dirty. >>>> > It's not something that all GWT users will have the patience or risk >>>> > tolerance for. >>>> > Improving the contents of the Incubator is, of course, important. >>>> > That's >>>> > what it's there for. But I would expect that as things are "cleaned >>>> > up" >>>> > they'll be promoted to the main GWT project or spun off into separate >>>> > libraries. What's left in the Incubator will always be >>>> > works-in-progress. >>>> > >>>> > On Fri, Oct 10, 2008 at 4:15 PM, Arthur Kalmenson >>>> > <[EMAIL PROTECTED]>wrote: >>>> > >>>> > >>>> > >>>> > > Hello everyone, >>>> > >>>> > > There was some discussion on the IRC channel about this, and I >>>> > > figured >>>> > > I'd put it up for everyone here. >>>> > >>>> > > I think that the incubator has a lot of useful things (FooBundle, >>>> > > CssResource, DatePicker), a lot of great ideas (Declarative UI) and >>>> > > some not so great things (PagingScrollTable). The incubator has a >>>> > > lot >>>> > > of potential, it just needs some polish and promotion. I think it >>>> > > needs the following: >>>> > >>>> > > 1. Better and nicer examples of widgets and use of some nice CSS >>>> > > (just >>>> > > use one of the GWT themes). >>>> > > 2. Clean up the existing widgets and make them easier to use. >>>> > > 3. A showcase to show off all these widgets and make them accessible >>>> > > like the main GWT showcase. >>>> > > 4. Promoting incubator as a great resource on the regular GWT group >>>> > > and on the main GWT site. >>>> > >>>> > > As it stands right now, few people know about the incubator and >>>> > > those >>>> > > that do are usually scared away. Most people end up going to widget >>>> > > libraries like ExtGWT and others. Just about every widget library >>>> > > I've >>>> > > seen is poorly done and ends up giving GWT a really bad name (ExtGWT >>>> > > especially). I think the incubator can offer a much better and >>>> > > cleaner >>>> > > widget library and components, but it needs to be cleaned up and >>>> > > promoted so more people contribute and give feedback. >>>> > >>>> > > What do you think? >>>> > >>>> > > Regards, >>>> > > Arthur Kalmenson >>>> >>> >>> >>> >>> -- >>> "There are only 10 types of people in the world: Those who understand >>> binary, and those who don't" >>> >>> >> >> >> > > > > -- > "There are only 10 types of people in the world: Those who understand > binary, and those who don't" > > > > --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/Google-Web-Toolkit-Contributors -~----------~----~----~----~------~----~------~--~---
