From magnus at therning.org Fri May 1 06:42:36 2015 From: magnus at therning.org (Magnus Therning) Date: Fri, 1 May 2015 08:42:36 +0200 Subject: [arch-haskell] Fwd: RFQ: Proposal for new versioning of packages In-Reply-To: References: <20150429201117.GC11928@tatooine> <20150429231024.6c07a583@archlinux.ca> Message-ID: <20150501064236.GA1041@tatooine> On Thu, Apr 30, 2015 at 07:07:48PM +0900, Nicola Squartini wrote: > Release numbers would still be reset on adding: > > xrev == 0 --> > haskell-zlib-0.5.4.2-76-x86_64.pkg.tar.xz > add xrev == n && n > 0 --> haskell-zlib-0.5.4.2-1.n-x86_64.pkg.tar.xz > add , xrev == 0 -->haskell-zlib--1-x86_64.pkg.tar.xz > > The only concern that I have with your versioning is that having 0.x in the > release number might suggest that the package is still in a testing stage. > And I make xrev == 0 first class (not adding the ".n") because I believe > Hackage revisions are ugly and should not exist. > Anyway, whatever scheme you choose it's good for me :) It's addition after bump of x-revision that I worry about: xrev == 0 --> haskell-zlib-0.5.4.2-76-x86_64.pkg.tar.xz add xrev == n && n > 0 --> haskell-zlib-0.5.4.2-1.n-x86_64.pkg.tar.xz rebuild --> haskell-zlib-0.5.4.2-2.n-x86_64.pkg.tar.xz add --> haskell-zlib-0.5.4.2-1.-x86_64.pkg.tar.xz add , xrev == 0 --> haskell-zlib--1-x86_64.pkg.tar.xz As you see, if the release is put before the x-rev and the release number is reset on add, then we end up going backwards in version numbers. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus The results point out the fragility of programmer expertise: advanced programmers have strong expectations about what programs should look like, and when those expectations are violated--in seemingly innocuous ways--their performance drops drastically. -- Elliot Soloway and Kate Ehrlich -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 163 bytes Desc: not available URL: From magnus at therning.org Fri May 1 06:55:57 2015 From: magnus at therning.org (Magnus Therning) Date: Fri, 1 May 2015 08:55:57 +0200 Subject: [arch-haskell] For the brave: alternative versioning in testing repo Message-ID: <20150501065557.GB1041@tatooine> The repo [haskell-testing] now contains all packages from [haskell-core] modified according to one of the alternatives for a new versioning scheme: -_- All the existing packages have been taken apart, had their meta data modified, and then been re-assembled again. If you feel brave, please turn on the testing repo and see what issues pop up (so far it's worked fine for me). /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Java is, in many ways, C++--. -- M Feldman -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 163 bytes Desc: not available URL: From tensor5 at gmail.com Fri May 1 07:41:22 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Fri, 1 May 2015 16:41:22 +0900 Subject: [arch-haskell] Fwd: RFQ: Proposal for new versioning of packages In-Reply-To: <20150501064236.GA1041@tatooine> References: <20150429201117.GC11928@tatooine> <20150429231024.6c07a583@archlinux.ca> <20150501064236.GA1041@tatooine> Message-ID: Got it. Thanks for the clarification. On Fri, May 1, 2015 at 3:42 PM, Magnus Therning wrote: > On Thu, Apr 30, 2015 at 07:07:48PM +0900, Nicola Squartini wrote: > > Release numbers would still be reset on adding: > > > > xrev == 0 --> > > haskell-zlib-0.5.4.2-76-x86_64.pkg.tar.xz > > add xrev == n && n > 0 --> haskell-zlib-0.5.4.2-1.n-x86_64.pkg.tar.xz > > add , xrev == 0 -->haskell-zlib--1-x86_64.pkg.tar.xz > > > > The only concern that I have with your versioning is that having 0.x in > the > > release number might suggest that the package is still in a testing > stage. > > And I make xrev == 0 first class (not adding the ".n") because I believe > > Hackage revisions are ugly and should not exist. > > Anyway, whatever scheme you choose it's good for me :) > > It's addition after bump of x-revision that I worry about: > > xrev == 0 --> haskell-zlib-0.5.4.2-76-x86_64.pkg.tar.xz > add xrev == n && n > 0 --> haskell-zlib-0.5.4.2-1.n-x86_64.pkg.tar.xz > rebuild --> haskell-zlib-0.5.4.2-2.n-x86_64.pkg.tar.xz > add --> > haskell-zlib-0.5.4.2-1.-x86_64.pkg.tar.xz > add , xrev == 0 --> haskell-zlib--1-x86_64.pkg.tar.xz > > As you see, if the release is put before the x-rev and the release > number is reset on add, then we end up going backwards in version > numbers. > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > The results point out the fragility of programmer expertise: advanced > programmers have strong expectations about what programs should look like, > and when those expectations are violated--in seemingly innocuous > ways--their performance drops drastically. > -- Elliot Soloway and Kate Ehrlich > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From xyne at archlinux.ca Sat May 2 15:31:53 2015 From: xyne at archlinux.ca (Xyne) Date: Sat, 2 May 2015 15:31:53 +0000 Subject: [arch-haskell] RFQ: Proposal for new versioning of packages In-Reply-To: <20150430062517.GA3678@arch-rddm.charmaie.local> References: <20150429201117.GC11928@tatooine> <20150429231024.6c07a583@archlinux.ca> <20150430055510.GA978@arch-rddm.charmaie.local> <20150430062517.GA3678@arch-rddm.charmaie.local> Message-ID: <20150502153153.05771bad@archlinux.ca> On 2015-04-30 08:25 +0200 Fabien Dubosson wrote: >> The issue I'm worried about is that of trailing zeroes. Is 2.1_1 < 2.1.1_0? > >warning: downgrading package dummy (2.1.1_0-1 => 2.1_1-1) In case you were not aware of it, pacman provides an executable named "vercmp" to compare versions. It also provides a vercmp manpage to explain how ALPM's version comparison function works. $ vercmp 2.1_1 2.1.1_0 -1 The underscore system seems to work as expected. I was afraid that it would trigger alphanumeric comparisons (e.g. 2.3_0 > 2.24_0) but it does not: $ vercmp 2.r3_0 2.r24_0 -1 Regards, Xyne From pardar at bk.ru Sat May 2 17:43:36 2015 From: pardar at bk.ru (=?UTF-8?B?0JTQsNC90LjQu9CwINCg0Y/Qt9Cw0L3QvtCy0YHQutC40Lk=?=) Date: Sat, 02 May 2015 20:43:36 +0300 Subject: [arch-haskell] =?utf-8?q?ghc-mod_in_repo?= Message-ID: <1430588616.205551944@f391.i.mail.ru> Hello, Sorry for bothering you, I didn't find any feedback email for haskell-core repo except this one. As I noticed, haskell-ghc-mod package is missing from repo for some time by now. Could you please re-add it back? Are there any problems? Thank you, Daniel -------------- next part -------------- An HTML attachment was scrubbed... URL: From pradermecker at yahoo.ca Sat May 2 18:42:47 2015 From: pradermecker at yahoo.ca (Pierre Radermecker) Date: Sat, 2 May 2015 18:42:47 +0000 (UTC) Subject: [arch-haskell] ghc-mod in repo In-Reply-To: <1430588616.205551944@f391.i.mail.ru> References: <1430588616.205551944@f391.i.mail.ru> Message-ID: <1301104557.344434.1430592167903.JavaMail.yahoo@mail.yahoo.com> I assume it is not there because the latest `ghc-mod` hackage release does not compile with ghc-7.10 yet. You can follow the issue here : https://github.com/kazu-yamamoto/ghc-mod/issues/437 Note that the master branch does compile fine so the best option is to `cabal install` at this stage.Be aware that there are still known bugs/issues with master on ghc-7.10 (my understanding is that these issues hold back a new hackage release) Cheers, On Saturday, May 2, 2015 7:43 PM, ?????? ??????????? wrote: Hello, Sorry for bothering you, I didn't find any feedback email for haskell-core repo except this one. As I noticed, haskell-ghc-mod package is missing from repo for some time by now. Could you please re-add it back? Are there any problems? Thank you, Daniel _______________________________________________ arch-haskell mailing list arch-haskell at haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell -------------- next part -------------- An HTML attachment was scrubbed... URL: From bo.bjornsen at gmail.com Sun May 3 06:35:50 2015 From: bo.bjornsen at gmail.com (=?ISO-8859-1?Q?Bj=F8rn_=D8ivind_Bj=F8rnsen?=) Date: Sun, 03 May 2015 08:35:50 +0200 Subject: [arch-haskell] ghc-mod in repo In-Reply-To: <1430588616.205551944@f391.i.mail.ru> References: <1430588616.205551944@f391.i.mail.ru> Message-ID: <2417524.AZQGVm67Gh@archdukebox> On Saturday 02 May 2015 20:43:36 ?????? ??????????? wrote: > Hello, > Sorry for bothering you, I didn't find any feedback email for haskell-core > repo except this one. As I noticed, haskell-ghc-mod package is missing from > repo for some time by now. Could you please re-add it back? Are there any > problems? Thank you, > Daniel Hello, Since the GHC 7.10 update, the version of ghc- mod in hackage does not compile. Here's a bug report (one of many) about it: https://github.com/kazu-yamamoto/ghc-mod/issues/437 master of that repository now builds fine. Don't know if it is missing any features, I haven't noticed any problems yet. So either clone that git repo and build/install into sandbox yourself, or wait for a new release on hackage :) Bj?rn ?ivind -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: This is a digitally signed message part. URL: From fabien.dubosson at gmail.com Sun May 3 11:35:15 2015 From: fabien.dubosson at gmail.com (Fabien Dubosson) Date: Sun, 3 May 2015 13:35:15 +0200 Subject: [arch-haskell] RFQ: Proposal for new versioning of packages In-Reply-To: <20150502153153.05771bad@archlinux.ca> References: <20150429201117.GC11928@tatooine> <20150429231024.6c07a583@archlinux.ca> <20150430055510.GA978@arch-rddm.charmaie.local> <20150430062517.GA3678@arch-rddm.charmaie.local> <20150502153153.05771bad@archlinux.ca> Message-ID: <20150503113515.GA1621@asus-home> > In case you were not aware of it, pacman provides an executable named "vercmp" > to compare versions. It also provides a vercmp manpage to explain how ALPM's > version comparison function works. > > $ vercmp 2.1_1 2.1.1_0 > -1 I wasn't aware, thanks for the tip! > The underscore system seems to work as expected. I was afraid that it would > trigger alphanumeric comparisons (e.g. 2.3_0 > 2.24_0) but it does not: > > $ vercmp 2.r3_0 2.r24_0 > -1 Good idea to think about and test this point. ++ Fab -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 801 bytes Desc: not available URL: From magnus at therning.org Sun May 3 12:17:43 2015 From: magnus at therning.org (Magnus Therning) Date: Sun, 3 May 2015 14:17:43 +0200 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: <20150501065557.GB1041@tatooine> References: <20150501065557.GB1041@tatooine> Message-ID: The testing repo now contains the full set of packages, an updated cblrepo, and I've made some updates of packages. I'm satisfied with how it works so I plan on moving it all over to core shortly. /M On 1 May 2015 8:56 am, "Magnus Therning" wrote: > The repo [haskell-testing] now contains all packages from > [haskell-core] modified according to one of the alternatives for a new > versioning scheme: > > -_- > > All the existing packages have been taken apart, had their meta data > modified, and then been re-assembled again. > > If you feel brave, please turn on the testing repo and see what issues > pop up (so far it's worked fine for me). > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > Java is, in many ways, C++--. > -- M Feldman > -------------- next part -------------- An HTML attachment was scrubbed... URL: From neitsab at esrevart.net Mon May 4 10:46:53 2015 From: neitsab at esrevart.net (Bastien Traverse) Date: Mon, 04 May 2015 12:46:53 +0200 Subject: [arch-haskell] gitit status update: feed fixed and uploaded, possibility of reincluding gitit? Message-ID: <55474E1D.4040203@esrevart.net> Hi all, feed has just been updated to work with utf8-string ? 1.0 and time 1.5.*. [1][2][3] gitit was updated on Apr 22 to fix building with GHC 7.10.1 [4], and although no new release has been made incorporating it [5] it seems from the Hackage dependencies list [6] it should build ok (but I may be wrong, I'm not that great at parsing Haskell deps :) Therefore it seems gitit should be able to build now with packages present in haskell-core and -happstack: can somebody try it and if ok reupload it to the repo? It's been a while since I couldn't upgrade my stack :) Thanks! [1] https://hackage.haskell.org/package/feed [2] https://github.com/bergmark/feed/commit/ae16db4f426f28a14d23da00e82553823a574736 [3] https://github.com/haskell-infra/hackage-trustees/issues/12#issuecomment-98665970 [4] https://github.com/jgm/gitit/pull/488 [5] https://github.com/jgm/gitit/releases [6] https://hackage.haskell.org/package/gitit From tensor5 at gmail.com Wed May 6 01:07:01 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 6 May 2015 10:07:01 +0900 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: References: <20150501065557.GB1041@tatooine> Message-ID: A problem raised when I tried to updates my repository: $ cblrepo add -d Diff,0.3.2_0,1 option -d: Cannot parse 'Diff,0.3.2_0,1' as PKG,VER,REL Seems like cblrepo doesn't parse the underscore character in the version field. On Sun, May 3, 2015 at 9:17 PM, Magnus Therning wrote: > The testing repo now contains the full set of packages, an updated > cblrepo, and I've made some updates of packages. I'm satisfied with how it > works so I plan on moving it all over to core shortly. > > /M > On 1 May 2015 8:56 am, "Magnus Therning" wrote: > >> The repo [haskell-testing] now contains all packages from >> [haskell-core] modified according to one of the alternatives for a new >> versioning scheme: >> >> -_- >> >> All the existing packages have been taken apart, had their meta data >> modified, and then been re-assembled again. >> >> If you feel brave, please turn on the testing repo and see what issues >> pop up (so far it's worked fine for me). >> >> /M >> >> -- >> Magnus Therning OpenPGP: 0xAB4DFBA4 >> email: magnus at therning.org jabber: magnus at therning.org >> twitter: magthe http://therning.org/magnus >> >> Java is, in many ways, C++--. >> -- M Feldman >> > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Wed May 6 05:28:30 2015 From: magnus at therning.org (Magnus Therning) Date: Wed, 6 May 2015 07:28:30 +0200 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: References: <20150501065557.GB1041@tatooine> Message-ID: On 6 May 2015 at 03:07, Nicola Squartini wrote: > A problem raised when I tried to updates my repository: > > $ cblrepo add -d Diff,0.3.2_0,1 > option -d: Cannot parse 'Diff,0.3.2_0,1' as PKG,VER,REL > > Seems like cblrepo doesn't parse the underscore character in the version > field. Unfortunately old x-revisions aren't available in the Hackage index file, so I can't really see the value of adding with an explicit x-revision. That's probably just my limited imagination though ;) Is it just that you want to be able to pass the output of `cblrepo list` back into `cblrepo add`, or did you have some other use case in mind? /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus From magnus at therning.org Wed May 6 05:48:04 2015 From: magnus at therning.org (Magnus Therning) Date: Wed, 6 May 2015 07:48:04 +0200 Subject: [arch-haskell] Kiwilight is going away, thank you Kaiting Message-ID: This isn't really affecting anyone, but some of you have been around long enough to remember that the repo used to be hosted on auriga.kiwilight.com. Yesterday I received an announcement from Kaiting Chen that he's taking down the Kiwilight server. It's been badly out-of-date for a long time, and somewhat neglected, but it has soldiered on and I'm still using it regularly for the ArchHaskell work. Thank you Kaiting for providing this excellent service for so long, it will be missed, but I see no issues with managing without it :) # A bit of history Early on in the life of ArchHaskell we were in need of a place to host the repo. Kaiting graciously gave us access to one of his Kiwilight servers (it wasn't Auriga at first, but I've forgotten its name). Soon after that he also gave me root access and the Kiwilight server wasn't only our repo, but also the machine where all packages were built. At some point, due to Kaiting being extremely busy, it started falling behind on updates (it was rather heavily configured, using Kerberos and other stuff I know nothing of, so I didn't dare trying to keep it up-to-date myself). Then the HTTP server failed, but we already had the Xsounds mirror by then. I kept on building the packages on Kiwilight though, until finally that didn't work either and I had to go back to building on my laptop. Still, every ArchHaskell package you've downloaded over the year has passed via the Kiwilight server, nowadays it's used only for signing. I'll miss the Kiwilight server, but the ArchHaskell work won't really be affected at all by its passing :) /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus From tensor5 at gmail.com Wed May 6 07:59:09 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 6 May 2015 16:59:09 +0900 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: References: <20150501065557.GB1041@tatooine> Message-ID: It's for the [haskell-happstack] repository, I need to update the dependencies on [haskell-core] packages, Diff is just one example. As I understand the new format introduced after the versioning change does not support x-revisions for distro pkgs yet. Or am I not using it correctly? On Wed, May 6, 2015 at 2:28 PM, Magnus Therning wrote: > On 6 May 2015 at 03:07, Nicola Squartini wrote: > > A problem raised when I tried to updates my repository: > > > > $ cblrepo add -d Diff,0.3.2_0,1 > > option -d: Cannot parse 'Diff,0.3.2_0,1' as PKG,VER,REL > > > > Seems like cblrepo doesn't parse the underscore character in the version > > field. > > Unfortunately old x-revisions aren't available in the Hackage index > file, so I can't really see the value of adding with an explicit > x-revision. That's probably just my limited imagination though ;) Is > it just that you want to be able to pass the output of `cblrepo list` > back into `cblrepo add`, or did you have some other use case in mind? > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Wed May 6 08:11:00 2015 From: magnus at therning.org (Magnus Therning) Date: Wed, 6 May 2015 10:11:00 +0200 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: References: <20150501065557.GB1041@tatooine> Message-ID: On 6 May 2015 at 09:59, Nicola Squartini wrote: > It's for the [haskell-happstack] repository, I need to update the > dependencies on [haskell-core] packages, Diff is just one example. As I > understand the new format introduced after the versioning change does not > support x-revisions for distro pkgs yet. Or am I not using it correctly? Ah, no, you are completely correct! You see! My limited imagination in action :) After breaking with the `[extra]` repo a while back I don't myself use the `Distro` package type. I didn't even consider that it needed a modification to deal with the addition of explicit x-revisions. Please create an issue on github (for `cblrepo`). I'll try to fix this ASAP. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus From tensor5 at gmail.com Wed May 6 08:44:46 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 6 May 2015 17:44:46 +0900 Subject: [arch-haskell] For the brave: alternative versioning in testing repo In-Reply-To: References: <20150501065557.GB1041@tatooine> Message-ID: Issue created! On Wed, May 6, 2015 at 5:11 PM, Magnus Therning wrote: > On 6 May 2015 at 09:59, Nicola Squartini wrote: > > It's for the [haskell-happstack] repository, I need to update the > > dependencies on [haskell-core] packages, Diff is just one example. As I > > understand the new format introduced after the versioning change does not > > support x-revisions for distro pkgs yet. Or am I not using it correctly? > > Ah, no, you are completely correct! You see! My limited imagination > in action :) > After breaking with the `[extra]` repo a while back I don't myself use > the `Distro` package type. I didn't even consider that it needed a > modification to deal with the addition of explicit x-revisions. > > Please create an issue on github (for `cblrepo`). I'll try to fix this > ASAP. > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tensor5 at gmail.com Wed May 6 08:50:32 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 6 May 2015 17:50:32 +0900 Subject: [arch-haskell] gitit status update: feed fixed and uploaded, possibility of reincluding gitit? In-Reply-To: References: <55474E1D.4040203@esrevart.net> Message-ID: Unfortunately there will be a delay in the update of [haskell-happstack] due to the recent changes to cblrepo ( https://github.com/magthe/cblrepo/issues/42). I'll update as soon as the issue is resolved. Nicola On Tue, May 5, 2015 at 10:19 PM, Nicola Squartini wrote: > It will be reintroduced soon, probably tomorrow. > > Nicola > > On Mon, May 4, 2015 at 7:46 PM, Bastien Traverse > wrote: > >> Hi all, >> >> feed has just been updated to work with utf8-string ? 1.0 and time >> 1.5.*. [1][2][3] >> >> gitit was updated on Apr 22 to fix building with GHC 7.10.1 [4], and >> although no new release has been made incorporating it [5] it seems from >> the Hackage dependencies list [6] it should build ok (but I may be >> wrong, I'm not that great at parsing Haskell deps :) >> >> Therefore it seems gitit should be able to build now with packages >> present in haskell-core and -happstack: can somebody try it and if ok >> reupload it to the repo? It's been a while since I couldn't upgrade my >> stack :) >> >> Thanks! >> >> [1] https://hackage.haskell.org/package/feed >> [2] >> >> https://github.com/bergmark/feed/commit/ae16db4f426f28a14d23da00e82553823a574736 >> [3] >> >> https://github.com/haskell-infra/hackage-trustees/issues/12#issuecomment-98665970 >> [4] https://github.com/jgm/gitit/pull/488 >> [5] https://github.com/jgm/gitit/releases >> [6] https://hackage.haskell.org/package/gitit >> _______________________________________________ >> arch-haskell mailing list >> arch-haskell at haskell.org >> http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tensor5 at gmail.com Thu May 7 12:17:10 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Thu, 7 May 2015 21:17:10 +0900 Subject: [arch-haskell] gitit status update: feed fixed and uploaded, possibility of reincluding gitit? In-Reply-To: References: <55474E1D.4040203@esrevart.net> Message-ID: I tried building gitit but there is a problem with the new `time` package. The issue was already reported here . Hopefully it will be fixed soon. Nicola On Wed, May 6, 2015 at 5:50 PM, Nicola Squartini wrote: > Unfortunately there will be a delay in the update of [haskell-happstack] > due to the recent changes to cblrepo ( > https://github.com/magthe/cblrepo/issues/42). I'll update as soon as the > issue is resolved. > > Nicola > > On Tue, May 5, 2015 at 10:19 PM, Nicola Squartini > wrote: > >> It will be reintroduced soon, probably tomorrow. >> >> Nicola >> >> On Mon, May 4, 2015 at 7:46 PM, Bastien Traverse >> wrote: >> >>> Hi all, >>> >>> feed has just been updated to work with utf8-string ? 1.0 and time >>> 1.5.*. [1][2][3] >>> >>> gitit was updated on Apr 22 to fix building with GHC 7.10.1 [4], and >>> although no new release has been made incorporating it [5] it seems from >>> the Hackage dependencies list [6] it should build ok (but I may be >>> wrong, I'm not that great at parsing Haskell deps :) >>> >>> Therefore it seems gitit should be able to build now with packages >>> present in haskell-core and -happstack: can somebody try it and if ok >>> reupload it to the repo? It's been a while since I couldn't upgrade my >>> stack :) >>> >>> Thanks! >>> >>> [1] https://hackage.haskell.org/package/feed >>> [2] >>> >>> https://github.com/bergmark/feed/commit/ae16db4f426f28a14d23da00e82553823a574736 >>> [3] >>> >>> https://github.com/haskell-infra/hackage-trustees/issues/12#issuecomment-98665970 >>> [4] https://github.com/jgm/gitit/pull/488 >>> [5] https://github.com/jgm/gitit/releases >>> [6] https://hackage.haskell.org/package/gitit >>> _______________________________________________ >>> arch-haskell mailing list >>> arch-haskell at haskell.org >>> http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell >>> >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tensor5 at gmail.com Fri May 8 01:35:41 2015 From: tensor5 at gmail.com (Nicola Squartini) Date: Fri, 8 May 2015 10:35:41 +0900 Subject: [arch-haskell] gitit status update: feed fixed and uploaded, possibility of reincluding gitit? In-Reply-To: References: <55474E1D.4040203@esrevart.net> Message-ID: And finally gitit is back! On Thu, May 7, 2015 at 9:17 PM, Nicola Squartini wrote: > I tried building gitit but there is a problem with the new `time` package. > The issue was already reported here > . Hopefully it will be fixed > soon. > > Nicola > > On Wed, May 6, 2015 at 5:50 PM, Nicola Squartini > wrote: > >> Unfortunately there will be a delay in the update of [haskell-happstack] >> due to the recent changes to cblrepo ( >> https://github.com/magthe/cblrepo/issues/42). I'll update as soon as the >> issue is resolved. >> >> Nicola >> >> On Tue, May 5, 2015 at 10:19 PM, Nicola Squartini >> wrote: >> >>> It will be reintroduced soon, probably tomorrow. >>> >>> Nicola >>> >>> On Mon, May 4, 2015 at 7:46 PM, Bastien Traverse >>> wrote: >>> >>>> Hi all, >>>> >>>> feed has just been updated to work with utf8-string ? 1.0 and time >>>> 1.5.*. [1][2][3] >>>> >>>> gitit was updated on Apr 22 to fix building with GHC 7.10.1 [4], and >>>> although no new release has been made incorporating it [5] it seems from >>>> the Hackage dependencies list [6] it should build ok (but I may be >>>> wrong, I'm not that great at parsing Haskell deps :) >>>> >>>> Therefore it seems gitit should be able to build now with packages >>>> present in haskell-core and -happstack: can somebody try it and if ok >>>> reupload it to the repo? It's been a while since I couldn't upgrade my >>>> stack :) >>>> >>>> Thanks! >>>> >>>> [1] https://hackage.haskell.org/package/feed >>>> [2] >>>> >>>> https://github.com/bergmark/feed/commit/ae16db4f426f28a14d23da00e82553823a574736 >>>> [3] >>>> >>>> https://github.com/haskell-infra/hackage-trustees/issues/12#issuecomment-98665970 >>>> [4] https://github.com/jgm/gitit/pull/488 >>>> [5] https://github.com/jgm/gitit/releases >>>> [6] https://hackage.haskell.org/package/gitit >>>> _______________________________________________ >>>> arch-haskell mailing list >>>> arch-haskell at haskell.org >>>> http://mail.haskell.org/cgi-bin/mailman/listinfo/arch-haskell >>>> >>> >>> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ramana at member.fsf.org Fri May 8 03:01:15 2015 From: ramana at member.fsf.org (Ramana Kumar) Date: Fri, 8 May 2015 04:01:15 +0100 Subject: [arch-haskell] pacman output for archhaskell packages Message-ID: When (re)installing ArchHaskell packages, pacman produces some unusual and typically uninformative output compared to for non-haskell packages. I mean, things like: Reading package info from stdin ... done. and, worse, there is often an "error" when upgrading a package "command failed to execute correctly", which appears to be harmless but is confronting nevertheless. Would it be possible to engineer the ArchHaskell PKGBUILDs so as to suppress these messages? -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Fri May 8 06:11:30 2015 From: magnus at therning.org (Magnus Therning) Date: Fri, 8 May 2015 08:11:30 +0200 Subject: [arch-haskell] pacman output for archhaskell packages In-Reply-To: References: Message-ID: On 8 May 2015 at 05:01, Ramana Kumar wrote: > When (re)installing ArchHaskell packages, pacman produces some unusual and > typically uninformative output compared to for non-haskell packages. > > I mean, things like: > Reading package info from stdin ... done. > > and, worse, there is often an "error" when upgrading a package "command > failed to execute correctly", which appears to be harmless but is > confronting nevertheless. > > Would it be possible to engineer the ArchHaskell PKGBUILDs so as to suppress > these messages? A fairly recent change to `cblrepo`[1] should take care of the "Reading package..." bit. I have opted to not spend the hours necessary to rebuild all packages, instead all added and updated packages will be more silent. When the next version of ghc comes out all packages will be quieter. I'd really like to get a reliable way to reproduce the second issue, so I can see what changes to the install script that's necessary to deal with it. /M [1]: http://is.gd/s26Bd9 -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus From ramana at member.fsf.org Fri May 8 06:48:49 2015 From: ramana at member.fsf.org (Ramana Kumar) Date: Fri, 8 May 2015 07:48:49 +0100 Subject: [arch-haskell] pacman output for archhaskell packages In-Reply-To: References: Message-ID: On 8 May 2015 at 07:11, Magnus Therning wrote: > A fairly recent change to `cblrepo`[1] should take care of the > "Reading package..." bit. I have opted to not spend the hours > necessary to rebuild all packages, instead all added and updated > packages will be more silent. When the next version of ghc comes out > all packages will be quieter. > Awesome! Thanks. > I'd really like to get a reliable way to reproduce the second issue, > so I can see what changes to the install script that's necessary to > deal with it. > When I next come across it I'll try to isolate it. > > /M > > [1]: http://is.gd/s26Bd9 > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Fri May 8 07:54:44 2015 From: magnus at therning.org (Magnus Therning) Date: Fri, 8 May 2015 09:54:44 +0200 Subject: [arch-haskell] pacman output for archhaskell packages In-Reply-To: References: Message-ID: On 8 May 2015 at 08:48, Ramana Kumar wrote: > On 8 May 2015 at 07:11, Magnus Therning wrote: >> A fairly recent change to `cblrepo`[1] should take care of the >> "Reading package..." bit. I have opted to not spend the hours >> necessary to rebuild all packages, instead all added and updated >> packages will be more silent. When the next version of ghc comes out >> all packages will be quieter. > > > Awesome! Thanks. Yeah, it's been irritating me for a long time too, so I decided to actually do something about it :) /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus From neitsab at esrevart.net Tue May 12 11:55:14 2015 From: neitsab at esrevart.net (Bastien Traverse) Date: Tue, 12 May 2015 13:55:14 +0200 Subject: [arch-haskell] Best way to workaround haskell-nats blocking upgrade? Message-ID: <5551EA22.2030807@esrevart.net> Hi, When I try to upgrade I get this blocking error: > :: Starting full system upgrade... > resolving dependencies... > looking for conflicting packages... > error: failed to prepare transaction (could not satisfy dependencies) > :: haskell-nats: requires ghc=7.8.4-1 > :: haskell-nats: requires haskell-hashable=1.2.3.1-2 Then I saw in https://github.com/archhaskell/habs/issues/183 this was a known issue, but I don't really get what the best workaround for it is: - I tried to `--ignore haskell-nats`, to no avail; - I tried to remove it but `haskell-semigroups` depends on it; - I tried to reinstall `haskell-semigroups` but it gave a long list of errors with many packages saying they depend on ghc=7.8.4-1 (my databases are up to date though); - I tried to reinstall haskell-gitit, which pulls haskell-pandoc but I also get a "failed to prepare transaction (could not satisfy dependencies)" error. So, is the only way to workaround this to purge the whole Haskell platform and then reinstall it, as suggested in the previously mentioned issue? Thanks From neitsab at esrevart.net Tue May 12 11:57:12 2015 From: neitsab at esrevart.net (Bastien Traverse) Date: Tue, 12 May 2015 13:57:12 +0200 Subject: [arch-haskell] gitit status update: feed fixed and uploaded, possibility of reincluding gitit? In-Reply-To: References: <55474E1D.4040203@esrevart.net> Message-ID: <5551EA98.4060302@esrevart.net> On Thu, May 7, 2015 at 9:17 PM, Nicola Squartini wrote: > I tried building gitit but there is a problem with the new `time` package. > The issue was already reported here > . Hopefully it will be fixed soon. On Fri, May 8, 2015 at 01:35 AM, Nicola Squartini wrote: > And finally gitit is back! Thanks a lot Nicola you rock!