Thanks for the responses guys, just to let you know I'm currently going through the backlog here and merging relevant patches in to my wx-2.9-only-wxhaskell-devel branch[1].
I'll send further periodic updates to the list when I reach milestones / get stuck. It's worth noting that the wxWidgets people "hope to make 3.0 in the spring of 2012"[2], and that 2.9 is the development release for 3.0. I'd like to cautiously suggest that we could aim to do a major version bump and release to Hackage in tandem with them. Dave, [1] http://darcsden.com/DukeDave/wxhaskell-dev [2] http://trac.wxwidgets.org/wiki/Roadmap On 12 November 2011 17:38, Alessandro Vermeulen <a.vermeu...@students.uu.nl> wrote: > Hi Dave, > > I would like to fall in line with Eric here. > > It is WIP in the sense that it works on my (our) machines but is far from > finished yet. > > For example, with the patches and the wp-config library you are now able to > build wxHaskell against 2.9 on Lion but not on Snow Leopard. Other platforms > are > completely untested. > > Another problem seems to be with painting on canvases. The Asteroids program > [1] > from Utrecht University, kind of works. But the game elements aren't drawn. > > This might have something to do with mistakes/errors/changes in the API to > wxWidgets. Maybe the types of some functions didn't change but the meaning of > the parameters did. I'm not familiar enough with the both versions to check > what > is the case. > > So probably the best way to continue is to build programs that used to work > and > that you have a working wxHaskell 2.8 off to compare. If you want them to send > in or to send in patches, please! > > Regards, > Alessandro > > [1] http://dl.dropbox.com/u/3031364/asteroids.zip > On 12 nov. 2011, at 17:42, Eric Kow wrote: > >> Thanks, Dave, >> >> Unfortunately, I'm going to have to switch back to hunker-down mode for >> wxHaskell. But here's what I can add: >> >>> I just have a few requests/questions for you Eric: >>> 1. Could you indicate exactly which patches are required (I've never >>> applied a darcs patch from an email, so I'm curious to try), or: >> >> Not really, but I can guess from the patch titles >> >> Minimal Dependencies >> -------------------- >> Some semantic dependencies may be missing: >> >> * wxEventType is an extern type in wxWidgets 2.9.2 >> * Rudimentary replacement for wx-config. >> * Don't forget parent dir in wx-config prefix search. >> * Fix oversight: include the wxWidgets libs in wx-config output. >> * use Haskell WxConfig library instead of wx-config on Windows >> >> Minimal wxWidgets 2.9 >> --------------------- >> * Only install the wx-config executable on Windows >> * WIP: do a better job grabbing extra lib arguments >> * Disable webkit and power events that were causing compile errors. >> * wxDrag+varia, >> * wxcore: remove wxPostScriptDC::{Set,Get}Resolution >> * PolygonFillModeFix >> * wxcore: Convert wxString to wchar_t* not char* >> * WIP: Remove references to wxPendingEvents >> * WIP: Remove more missing stuff >> >> Minimal GHC 7.2 >> --------------- >> * Fix build on GHC 7.2. >> * wx: FlexibleInstances for GHC 7.2 >> >> Minimal No More Eiffel >> ---------------------- >> * wxcore: No more Eiffel (can probably stand alone) >> * wxcore: Remove elj prefix from cpp files. >> * wxcore: No more Eiffel followup >> >> Probably not needed >> ------------------- >> * Fix inconsistent newline terminator >> * wxdirect: Strip away ability to deal with Eiffel files. >> * wxdirect: Modernise exception handling. >> * wxdirect: Bump to 0.14 >> * Update Graphics.UI.WXCore.WxcDefs haddock to reflect manual maintenance. >> * Trailing newlines on EOF. >> * Modernise examples not to require haskell98 package >> >>> 2. Are the patches available on a darcs repo somewhere? >>> Either way I'd like to get the de-eiffel-ification in to my darcsden >>> wxhaskell-devel branch[1] >> >> I'd like to push them to a fork but I need to sort out some technical issues >> with Alex first. >> >> Meanwhile, know that you can cherry pick from emailed patches with darcs >> apply -i >> >>> 3. I see that you've removed all the "wxACCEL_ALT : INTEGER is 1" >>> style constant exports; with them removed how to do the constants get >>> exposed in Haskell? >> >> Before the patch, wxdirect parses the Eiffel files and spits out equivalent >> Haskell code. The patch simply adds the results of this process to the repo >> and drops the bit of Setup.hs that does this parsing. >> >> Careful: it may actually be worthwhile to retain the *functionality* in >> wxdirect that does this parsing. Not sure, but maybe this would be helpful >> in cases where people for some reason want to install an old wxcore, without >> us having to ask them to downgrade wxdirect. >> >>> 4. How "WIP" do you mean by "WIP"? :) >> >> I think it means two things >> >> (A) I don't know if this is a good idea, but it's otherwise complete >> (B) This works for me but I'm pretty sure it won't work for you >> >> In the case of the "do a better job" patch, updating this may actually be >> quite easy. You just need to ignore the basename stuff on Windows, and >> append a "-2.8" (or whatever as the case may be in case we're not using >> wxWidgets 2.8) >> >> The other patches probably belong in the B category. De-WIPping them >> probably just consists in thinking about whether or not this would actually >> be a good idea in practice and darcs amend --edit'ing the name. >> Unfortunately, having the two patches in a repo at a same time would be >> conflict, so one would have to uvpull the wippy versions first before >> pulling the rename. >> >> Regards, >> >> -- >> Eric Kow <http://erickow.com> >> >> >> ------------------------------------------------------------------------------ >> RSA(R) Conference 2012 >> Save $700 by Nov 18 >> Register now >> http://p.sf.net/sfu/rsa-sfdev2dev1 >> _______________________________________________ >> wxhaskell-devel mailing list >> wxhaskell-devel@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel > > ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-novd2d _______________________________________________ wxhaskell-devel mailing list wxhaskell-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel
