Menu
▾
▴
wxhaskell-devel
|
From: Eric K. <eri...@gm...> - 2011-11-12 16:21:53
|
I suggest a trusting-by-default policy for gaining push access to main repo on cho: - anybody who has contributed accepted patches to wxhaskell - and who requests it (or says yes to a "would you like commit access") The basic rule for using this access is - use good judgement: only push if you have some confidence in what's going in Using a DVCS means we can afford to relax a little. The worst case is a reset and a message to wxhaskell mailing lists asking folks to reset their own branches Darcsden/Patch-tag forks are great, but it may help for synchronisation to have a single focal point that moves forward quickly. Review is great, but it be more effective to make it a social norm than a technical stumbling block. Comments? -- Eric Kow <http://erickow.com> |
|
From: Eric K. <eri...@gm...> - 2011-11-21 09:00:06
|
Any comments? On 12 Nov 2011, at 16:21, Eric Kow wrote: > I suggest a trusting-by-default policy for gaining push access to main repo on cho: > > - anybody who has contributed accepted patches to wxhaskell > - and who requests it (or says yes to a "would you like commit access") -- Eric Kow <http://erickow.com> |
|
From: Dave T. <duk...@gm...> - 2011-11-21 18:31:41
|
Not surprisingly, I am in favour of this :) Given that there aren't going to be any more 2.8.x releases of wxWidgets, I'm happy to say: If you want a stable(ish) wxHaskell, then use the current hackage release along with the last stable wxWidgets release (2.8.12). If you want bleeding edge wxHaskell, then pull from code.haskell.org along with the latest dev wxWidgets release (currently 2.9.2). I should note one more time that I'm quite happy to stop supporting pre 2.9.x support now, I don't know if anyone has any objection to this? Dave, On 21 November 2011 08:59, Eric Kow <eri...@gm...> wrote: > Any comments? > > On 12 Nov 2011, at 16:21, Eric Kow wrote: > >> I suggest a trusting-by-default policy for gaining push access to main repo on cho: >> >> - anybody who has contributed accepted patches to wxhaskell >> - and who requests it (or says yes to a "would you like commit access") > > -- > Eric Kow <http://erickow.com> > > > ------------------------------------------------------------------------------ > 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 > wxh...@li... > https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel > |
|
From: Jeremy O'D. <jer...@gm...> - 2011-11-28 11:37:10
|
On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: > Not surprisingly, I am in favour of this :) > I have spent a while thinking about this, as it has considerable ramifications. I don't think we have ever seen a case of an irresponsible committer (could such a thing even exist in the Haskell community?), so I'm in favour. > Given that there aren't going to be any more 2.8.x releases of > wxWidgets, I'm happy to say: > If you want a stable(ish) wxHaskell, then use the current hackage > release along with the last stable wxWidgets release (2.8.12). > If you want bleeding edge wxHaskell, then pull from code.haskell.org > along with the latest dev wxWidgets release (currently 2.9.2). > > I should note one more time that I'm quite happy to stop supporting > pre 2.9.x support now, I don't know if anyone has any objection to > this? > The caveat is that I would like to do one more release on Hackage supporting 2.8.x, as we have a number of valuable bugfixes in the devel branches which would benefit users of 2.8.x. I will try to do this over then next two weeks, so my proposal is... Patches committed until the end of 2011 should be verified on a wxWidgets 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the version number from 0.13.x to 0.14.x. How does this sound? Jeremy |
|
From: Dave T. <duk...@gm...> - 2011-11-28 17:57:14
|
On 28 November 2011 11:37, Jeremy O'Donoghue <jer...@gm...> wrote: > On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: >> >> Not surprisingly, I am in favour of this :) > > I have spent a while thinking about this, as it has considerable > ramifications. > > I don't think we have ever seen a case of an irresponsible committer (could > such a thing even exist in the Haskell community?), so I'm in favour. > >> >> Given that there aren't going to be any more 2.8.x releases of >> wxWidgets, I'm happy to say: >> If you want a stable(ish) wxHaskell, then use the current hackage >> release along with the last stable wxWidgets release (2.8.12). >> If you want bleeding edge wxHaskell, then pull from code.haskell.org >> along with the latest dev wxWidgets release (currently 2.9.2). >> >> I should note one more time that I'm quite happy to stop supporting >> pre 2.9.x support now, I don't know if anyone has any objection to >> this? > > The caveat is that I would like to do one more release on Hackage > supporting 2.8.x, as we have a number of valuable bugfixes in the devel > branches which would benefit users of 2.8.x. I will try to do this over then > next two weeks, so my proposal is... > > Patches committed until the end of 2011 should be verified on a wxWidgets > 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the > version number from 0.13.x to 0.14.x. > > How does this sound? Well it's the most sensible new year's resolution I've heard thus far :) I shall continue pushing to my >= wx-2.9 repo on darcs den, in to which I'm aiming to get all the patches which are sent out on the mailing list as well. Dave, > > Jeremy > |
|
From: Maciek M. <ma...@mm...> - 2011-11-28 18:09:43
|
I don't have a strong opinion on which version of wx should be supported by wxHaskell, as long as there is at least one that works on Windows without the need to compile wxWidgets from source. Until there is a wxPack available for 2.9 I fear that the majority of Windows users will be stuck with wxHaskell versions that work with 2.8. That aside, focusing on supporting a single version of wxWidgets sounds like a reasonable thing to do. Maciek On Mon, Nov 28, 2011 at 5:56 PM, Dave Tapley <duk...@gm...> wrote: > On 28 November 2011 11:37, Jeremy O'Donoghue <jer...@gm...> wrote: >> On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: >>> >>> Not surprisingly, I am in favour of this :) >> >> I have spent a while thinking about this, as it has considerable >> ramifications. >> >> I don't think we have ever seen a case of an irresponsible committer (could >> such a thing even exist in the Haskell community?), so I'm in favour. >> >>> >>> Given that there aren't going to be any more 2.8.x releases of >>> wxWidgets, I'm happy to say: >>> If you want a stable(ish) wxHaskell, then use the current hackage >>> release along with the last stable wxWidgets release (2.8.12). >>> If you want bleeding edge wxHaskell, then pull from code.haskell.org >>> along with the latest dev wxWidgets release (currently 2.9.2). >>> >>> I should note one more time that I'm quite happy to stop supporting >>> pre 2.9.x support now, I don't know if anyone has any objection to >>> this? >> >> The caveat is that I would like to do one more release on Hackage >> supporting 2.8.x, as we have a number of valuable bugfixes in the devel >> branches which would benefit users of 2.8.x. I will try to do this over then >> next two weeks, so my proposal is... >> >> Patches committed until the end of 2011 should be verified on a wxWidgets >> 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the >> version number from 0.13.x to 0.14.x. >> >> How does this sound? > > Well it's the most sensible new year's resolution I've heard thus far :) > > I shall continue pushing to my >= wx-2.9 repo on darcs den, in to > which I'm aiming to get all the patches which are sent out on the > mailing list as well. > > Dave, > >> >> Jeremy >> > > ------------------------------------------------------------------------------ > 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 > wxh...@li... > https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel > |
|
From: Eric K. <eri...@gm...> - 2011-11-28 18:32:06
|
On 28 Nov 2011, at 18:09, Maciek Makowski wrote: > I don't have a strong opinion on which version of wx should be > supported by wxHaskell, as long as there is at least one that works on > Windows without the need to compile wxWidgets from source. Until there > is a wxPack available for 2.9 I fear that the majority of Windows > users will be stuck with wxHaskell versions that work with 2.8. Probably not an issue as the proposal would only affect devel versions of wxHaskell (as I understand it) As a general principle, I guess hackage versions should always support the stable versions of the underlying wxWidgets. This means making sure we've got 2.8 support until 3.0 comes out, and also making sure we can cope with the 3.0 release -- Eric Kow <http://erickow.com> |
|
From: Alessandro V. <a.v...@st...> - 2011-11-28 18:24:50
|
> I fear that the majority of Windows > users will be stuck with wxHaskell versions that work with 2.8. Well, you can always bundle the wx libraries with your application in that case and create your own wxPack. :-) - Alessandro On 28 nov. 2011, at 19:09, Maciek Makowski wrote: > I don't have a strong opinion on which version of wx should be > supported by wxHaskell, as long as there is at least one that works on > Windows without the need to compile wxWidgets from source. Until there > is a wxPack available for 2.9 I fear that the majority of Windows > users will be stuck with wxHaskell versions that work with 2.8. > > That aside, focusing on supporting a single version of wxWidgets > sounds like a reasonable thing to do. > > Maciek > > On Mon, Nov 28, 2011 at 5:56 PM, Dave Tapley <duk...@gm...> wrote: >> On 28 November 2011 11:37, Jeremy O'Donoghue <jer...@gm...> wrote: >>> On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: >>>> >>>> Not surprisingly, I am in favour of this :) >>> >>> I have spent a while thinking about this, as it has considerable >>> ramifications. >>> >>> I don't think we have ever seen a case of an irresponsible committer (could >>> such a thing even exist in the Haskell community?), so I'm in favour. >>> >>>> >>>> Given that there aren't going to be any more 2.8.x releases of >>>> wxWidgets, I'm happy to say: >>>> If you want a stable(ish) wxHaskell, then use the current hackage >>>> release along with the last stable wxWidgets release (2.8.12). >>>> If you want bleeding edge wxHaskell, then pull from code.haskell.org >>>> along with the latest dev wxWidgets release (currently 2.9.2). >>>> >>>> I should note one more time that I'm quite happy to stop supporting >>>> pre 2.9.x support now, I don't know if anyone has any objection to >>>> this? >>> >>> The caveat is that I would like to do one more release on Hackage >>> supporting 2.8.x, as we have a number of valuable bugfixes in the devel >>> branches which would benefit users of 2.8.x. I will try to do this over then >>> next two weeks, so my proposal is... >>> >>> Patches committed until the end of 2011 should be verified on a wxWidgets >>> 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the >>> version number from 0.13.x to 0.14.x. >>> >>> How does this sound? >> >> Well it's the most sensible new year's resolution I've heard thus far :) >> >> I shall continue pushing to my >= wx-2.9 repo on darcs den, in to >> which I'm aiming to get all the patches which are sent out on the >> mailing list as well. >> >> Dave, >> >>> >>> Jeremy >>> >> >> ------------------------------------------------------------------------------ >> 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 >> wxh...@li... >> 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 > wxh...@li... > https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel |
|
From: <mac...@gm...> - 2011-11-28 21:41:29
|
Last time I tried I wasn't able to get wx 2.8 to build on Windows. After spending a day trying to sort out gcc running out of memory and issues with Unicode support I resorted to wxPack. If not for wxPack I would have probably given up on using wxHaskell altogether. I'm sure it was possible to build it somehow, but I was not prepared to invest massive amount of time just to be able to use a GUI library. Maciek On Mon, Nov 28, 2011 at 6:24 PM, Alessandro Vermeulen <a.v...@st...> wrote: >> I fear that the majority of Windows >> users will be stuck with wxHaskell versions that work with 2.8. > Well, you can always bundle the wx libraries with your application in that > case and create your own wxPack. :-) > > - Alessandro > On 28 nov. 2011, at 19:09, Maciek Makowski wrote: > >> I don't have a strong opinion on which version of wx should be >> supported by wxHaskell, as long as there is at least one that works on >> Windows without the need to compile wxWidgets from source. Until there >> is a wxPack available for 2.9 I fear that the majority of Windows >> users will be stuck with wxHaskell versions that work with 2.8. >> >> That aside, focusing on supporting a single version of wxWidgets >> sounds like a reasonable thing to do. >> >> Maciek >> >> On Mon, Nov 28, 2011 at 5:56 PM, Dave Tapley <duk...@gm...> wrote: >>> On 28 November 2011 11:37, Jeremy O'Donoghue <jer...@gm...> wrote: >>>> On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: >>>>> >>>>> Not surprisingly, I am in favour of this :) >>>> >>>> I have spent a while thinking about this, as it has considerable >>>> ramifications. >>>> >>>> I don't think we have ever seen a case of an irresponsible committer (could >>>> such a thing even exist in the Haskell community?), so I'm in favour. >>>> >>>>> >>>>> Given that there aren't going to be any more 2.8.x releases of >>>>> wxWidgets, I'm happy to say: >>>>> If you want a stable(ish) wxHaskell, then use the current hackage >>>>> release along with the last stable wxWidgets release (2.8.12). >>>>> If you want bleeding edge wxHaskell, then pull from code.haskell.org >>>>> along with the latest dev wxWidgets release (currently 2.9.2). >>>>> >>>>> I should note one more time that I'm quite happy to stop supporting >>>>> pre 2.9.x support now, I don't know if anyone has any objection to >>>>> this? >>>> >>>> The caveat is that I would like to do one more release on Hackage >>>> supporting 2.8.x, as we have a number of valuable bugfixes in the devel >>>> branches which would benefit users of 2.8.x. I will try to do this over then >>>> next two weeks, so my proposal is... >>>> >>>> Patches committed until the end of 2011 should be verified on a wxWidgets >>>> 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the >>>> version number from 0.13.x to 0.14.x. >>>> >>>> How does this sound? >>> >>> Well it's the most sensible new year's resolution I've heard thus far :) >>> >>> I shall continue pushing to my >= wx-2.9 repo on darcs den, in to >>> which I'm aiming to get all the patches which are sent out on the >>> mailing list as well. >>> >>> Dave, >>> >>>> >>>> Jeremy >>>> >>> >>> ------------------------------------------------------------------------------ >>> 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 >>> wxh...@li... >>> 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 >> wxh...@li... >> https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel > > |
|
From: Jeremy O'D. <jer...@gm...> - 2011-11-28 22:25:43
|
Maciek, Alessandro, and anyone worried about the future. We will not stop supporting wxWidgets 2.8 until 2.9/3.0 builds are standard or readily available on all of the supported platforms. The only difference is that primary development will be on 2.9.x instead of 2.8.x as is the case right now, so wxHaskell *developers* will need to have wxWidgets 2.9.x builds (wxWidgets isn't *that* bad to build, although you need to follow the instructions very carefully) Basically (using non-DVCS terminology - sorry Eric) the old development branch will become a maintenance branch and the current feature development branch becomes the mainline. What this means is that 2.8.x will see only bugfixes from Jan 2012, and these will be backported from the 2.9.x branch as needed. This is tedious, but not hard (we manage more than 30 branches of the same code where I work in some cases) - and is the only realistic option given the limited development and test effort available to us. The bad news is that I volunteer for job of merge monkey(*), which means that it will get done eventually (sadly for fairly lengthy values of eventually, unless anyone can invent a 36 hour day :-) That said, if anyone has a high boredom threshold, they are welcome to take it off of my hands. (*) actually, we used to have a rubber chicken at work, given to the person currently merging - had much the same function as a critical section - so maybe that should be 'merge chicken'. Best regards Jeremy On 28 November 2011 21:41, <mac...@gm...> wrote: > Last time I tried I wasn't able to get wx 2.8 to build on Windows. > After spending a day trying to sort out gcc running out of memory and > issues with Unicode support I resorted to wxPack. If not for wxPack I > would have probably given up on using wxHaskell altogether. I'm sure > it was possible to build it somehow, but I was not prepared to invest > massive amount of time just to be able to use a GUI library. > > Maciek > > On Mon, Nov 28, 2011 at 6:24 PM, Alessandro Vermeulen > <a.v...@st...> wrote: > >> I fear that the majority of Windows > >> users will be stuck with wxHaskell versions that work with 2.8. > > Well, you can always bundle the wx libraries with your application in > that > > case and create your own wxPack. :-) > > > > - Alessandro > > On 28 nov. 2011, at 19:09, Maciek Makowski wrote: > > > >> I don't have a strong opinion on which version of wx should be > >> supported by wxHaskell, as long as there is at least one that works on > >> Windows without the need to compile wxWidgets from source. Until there > >> is a wxPack available for 2.9 I fear that the majority of Windows > >> users will be stuck with wxHaskell versions that work with 2.8. > >> > >> That aside, focusing on supporting a single version of wxWidgets > >> sounds like a reasonable thing to do. > >> > >> Maciek > >> > >> On Mon, Nov 28, 2011 at 5:56 PM, Dave Tapley <duk...@gm...> > wrote: > >>> On 28 November 2011 11:37, Jeremy O'Donoghue < > jer...@gm...> wrote: > >>>> On 21 November 2011 18:31, Dave Tapley <duk...@gm...> wrote: > >>>>> > >>>>> Not surprisingly, I am in favour of this :) > >>>> > >>>> I have spent a while thinking about this, as it has considerable > >>>> ramifications. > >>>> > >>>> I don't think we have ever seen a case of an irresponsible committer > (could > >>>> such a thing even exist in the Haskell community?), so I'm in favour. > >>>> > >>>>> > >>>>> Given that there aren't going to be any more 2.8.x releases of > >>>>> wxWidgets, I'm happy to say: > >>>>> If you want a stable(ish) wxHaskell, then use the current hackage > >>>>> release along with the last stable wxWidgets release (2.8.12). > >>>>> If you want bleeding edge wxHaskell, then pull from code.haskell.org > >>>>> along with the latest dev wxWidgets release (currently 2.9.2). > >>>>> > >>>>> I should note one more time that I'm quite happy to stop supporting > >>>>> pre 2.9.x support now, I don't know if anyone has any objection to > >>>>> this? > >>>> > >>>> The caveat is that I would like to do one more release on Hackage > >>>> supporting 2.8.x, as we have a number of valuable bugfixes in the > devel > >>>> branches which would benefit users of 2.8.x. I will try to do this > over then > >>>> next two weeks, so my proposal is... > >>>> > >>>> Patches committed until the end of 2011 should be verified on a > wxWidgets > >>>> 2.8.x release. From 1st Jan 2012, 2.8.x is dropped, and we'll bump the > >>>> version number from 0.13.x to 0.14.x. > >>>> > >>>> How does this sound? > >>> > >>> Well it's the most sensible new year's resolution I've heard thus far > :) > >>> > >>> I shall continue pushing to my >= wx-2.9 repo on darcs den, in to > >>> which I'm aiming to get all the patches which are sent out on the > >>> mailing list as well. > >>> > >>> Dave, > >>> > >>>> > >>>> Jeremy > >>>> > >>> > >>> > ------------------------------------------------------------------------------ > >>> 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 > >>> wxh...@li... > >>> 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 > >> wxh...@li... > >> 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 > wxh...@li... > https://lists.sourceforge.net/lists/listinfo/wxhaskell-devel > |
Thanks for helping keep SourceForge clean.
X