Future enhancements -------------------
The final version will work from modules residing in a subfolder of the Plugins folder, probably to be called something cryptic like "FlipsIDE Modules".
Each modile will be simply a stack file with a stack script containing two handlers: one for loading a specified IDE and one for purging it.
It will ship with at least two prefab modules, one for MC and one for running with no IDE at all.
It will be easy to write the load/purge handlers, so we could see modules for any third party set of tools that comprises a complete development environment.
From the list of installed modules you'll be able to set one as a default, so that launching Rev with FlipsIDE installed will cause FlipsIDE to automatically flip into the specified IDE if desired.
So in fact, we will need a separate module for each release of each IDE since the list of components may vary between releases (this applies only to Rev at this time). I wonder whether each IDE should not have a built-in function that returns a list of its components, so FlipsIDE (and any other utils) can get it dynamically. This would keep the number of prefab modules down on the long term. Rev's suspend development environment function must use such a list.
Rev's "Suspend Development Environment" merely hides windows; Rev relies on an extensive array of backscripts and frontScripts which it leaves in place when "suspended"; FlipsIDE purges those to load MC with a clean slate.
My interest in using modules was to provide a flipping mechanism without encumbering an IDE publisher to modify their work just for the modest convenience of FlipsIDE.
While I contribute to the MC IDE, I have no control over the Rev IDE. As much as I appreciate Kevin's efforts to help resolve FlipsIDE's crashing issue I don't presume that he would want to provide support for it directly.
Yes, IDEs will change from time to time. But the sort of changes that would break FlipsIDE are likely few, and I could probably update a module faster than we could expect RunRev to get around to modifying one more internal element in their IDE. Until all elements of the Rev IDE are completely error-free I think what benefits it most is removing things from it rather than adding to it.
For the foreseeable future I imagine we'd see no more than four or five modules total: one for MC, one for Rev, one for "No IDE", possibly one for FreeGUI if Alain Farmer resumes interest in that valuable idea, and maybe one other I don't know about yet. Most toolkits (Chipp's Alt series, Scott Rossi's wonders, devolution, etc.) may be comprehensive but I don't know that any of them are designed to provide a complete IDE, from stack creation to standalone building. Speaking for myself, I doubt I'll ever have enough free time to expand devolution from a toolkit into a full IDE, and if I had the time I'd probably elect to spend it making another consumer app over a dev tool anyway (or taking another vacation in Malta -- gotta get back to the Bacchus restaurant in Mdina ASAP <g>).
Given what is likely a small and finite number of IDEs FlipsIDE would need to support, as an alternative to modules I might just leave the code inside itself as it is now. The only upside to a module is that on the odd chance that some third party wants to make an IDE they wouldn't have to go mucking through FlipsIDE's code, and anyone could write one so it needn't necessarily encumber the IDE publisher.
Rev, MC, and "None" are currently supported. I don't know of any other IDE except FreeGUI, and reading the FreeGUI list at <http://groups.yahoo.com/group/freeGUI/> it seems that Alain may have lost interest in it in favor of experimenting with making an authoring GUI from Mozilla's DOM. And since FreeGUI predates the MC IDE's going open source, in its current state it's not an IDE itself but rather an add-on to MC; it could be made a more convenient experience by forking the MC IDE into a completely separate FreeGUI, but that would require some modification and I'm unsure whether Alain's up for it.
If he resumes work on FreeGUI I'll jump at the chance to make sure it's well supported in FlipsIDE. I think the goal of providing a HyperCard-like experience for the Rev engine is valuable, and FreeGUI is most of the way to doing that.
And of course if there are other IDEs that I don't know about it would be good to be in touch with those authors so we could define what's simplest for them.
In the meantime, the most critical thing with FlipsIDE is to make sure the crashing error is resolved. Kevin has been very helpful on this, but unfortunately my time is spoken for through most of Q1 on client work, so any additional data from testers is very helpful:
If you get a crash please let me know which IDE you were in when it happened, how many times you had flipped back and forth before the crash, and if on a Mac please include the crash log.
-- Richard Gaskin Fourth World Media Corporation __________________________________________________ Rev tools and more: http://www.fourthworld.com/rev _______________________________________________ metacard mailing list metacard@lists.runrev.com http://lists.runrev.com/mailman/listinfo/metacard
