Re: haskell-http

2009-07-24 Thread Joachim Breitner
Hi, Am Freitag, den 24.07.2009, 16:49 -0700 schrieb David Fox: > This new version of cdbs fails its unit tests when built on older > distributions - is there a safe way to build haskell-http without it? > Or maybe only a small patch to the current cdbs is required? you can build with an old cdbs

Re: haskell-http

2009-07-24 Thread David Fox
This new version of cdbs fails its unit tests when built on older distributions - is there a safe way to build haskell-http without it? Or maybe only a small patch to the current cdbs is required? On Tue, Jul 21, 2009 at 12:44 PM, Joachim Breitner wrote: > Hi, > > Am Dienstag, den 21.07.2009, 12

hsql-odbc and hsql-sqlite3

2009-07-24 Thread Joachim Breitner
Hi, I have put haskell-hsql-{,-mysql,-postgresql} back into shape. They had new upstream versions that made them compile with ghc-6.10.4. But hsql-odbc and hsql-sqlite3 do not, and also don’t build on hackage. Does anybody know if these are still useful, or maintained? If no one steps up to do up

Re: Bug#538098: [Pkg-haskell-maintainers] Bug#538098: Please upgrade to ≥0.5.1.0.

2009-07-24 Thread Joachim Breitner
[taking the bugnumber out of the loop, zlib has been updated] Am Freitag, den 24.07.2009, 23:32 +0100 schrieb Ganesh Sittampalam: > On Fri, 24 Jul 2009, Joachim Breitner wrote: > > don?t worry, I believe that darcs? dependency is there for a good > > reason. I just wanted to bring up this issue of

Re: Bug#538098: [Pkg-haskell-maintainers] Bug#538098: Please upgrade to ≥0.5.1.0.

2009-07-24 Thread Ganesh Sittampalam
On Fri, 24 Jul 2009, Joachim Breitner wrote: don?t worry, I believe that darcs? dependency is there for a good reason. I just wanted to bring up this issue of reasonable dependency vs. haskell platform in general. Would shipping multiple versions in the same .deb be at all feasible? Ganesh

Re: Bug#537705: ITP: agda-executable -- Commandline interface to Agda -- a dependently typed functional programming language and proof assistant

2009-07-24 Thread Joachim Breitner
Hi, Am Freitag, den 24.07.2009, 14:30 +0100 schrieb Iain Lane: > On 24 Jul 2009, at 14:10, Joachim Breitner wrote: > > > there is also the relatively common variant "agda-bin". > > How is this any better than -executable though? Seems to me like the > same issues apply to both. it is more co

Re: haddock depwaits

2009-07-24 Thread Philipp Kern
On Fri, Jul 24, 2009 at 01:50:38PM +0200, Joachim Breitner wrote: > the first upload of haddock causes FTBFS, and not all packages have > Depwaits on haddock that should have. > > The attached file first sets dep-waits for all packages that have a > depwait on ghc6 (automatically generated), and t

Re: Bug#537705: ITP: agda-executable -- Commandline interface to Agda -- a dependently typed functional programming language and proof assistant

2009-07-24 Thread Iain Lane
Hi, On 24 Jul 2009, at 14:10, Joachim Breitner wrote: there is also the relatively common variant "agda-bin". How is this any better than -executable though? Seems to me like the same issues apply to both. Regards, Iain This message has been checked for viruses but the contents of an at

Re: Bug#537705: ITP: agda-executable -- Commandline interface to Agda -- a dependently typed functional programming language and proof assistant

2009-07-24 Thread Joachim Breitner
Hi, Am Freitag, den 24.07.2009, 13:03 +0100 schrieb Iain Lane: > On 23 Jul 2009, at 02:50, Trent W. Buck wrote: > > Why can't it just be "agda"? Even if there's a separate library > > package, that would have the source name "haskell-agda" and binary > > names > > "libghc6-agda-{dev,prof,doc}",

Re: [Pkg-haskell-maintainers] Bug#538098: Bug#538098: Please upgrade to ≥0.5.1.0.

2009-07-24 Thread Iain Lane
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, On 23 Jul 2009, at 09:32, Joachim Breitner wrote: now it starts: The platform specifies zlib-0.5.0.0... but using something deprecated should be avoided. I’m still not fully convinced that strictly to the platform is a good idea. This is int

Re: Bug#537705: ITP: agda-executable -- Commandline interface to Agda -- a dependently typed functional programming language and proof assistant

2009-07-24 Thread Iain Lane
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, On 23 Jul 2009, at 02:50, Trent W. Buck wrote: Why can't it just be "agda"? Even if there's a separate library package, that would have the source name "haskell-agda" and binary names "libghc6-agda-{dev,prof,doc}", so there'd be no conflict

haddock depwaits

2009-07-24 Thread Joachim Breitner
Hi, the first upload of haddock causes FTBFS, and not all packages have Depwaits on haddock that should have. The attached file first sets dep-waits for all packages that have a depwait on ghc6 (automatically generated), and then depwaits for those package that have already tried to build a packa

libghc6-haskell-doc fixed

2009-07-24 Thread Joachim Breitner
Hi, libghc6-haskell-doc hat strict dependency on ghc6-doc, but (being arch:all), it would not be updated by a binNMU. I did a sourceful upload to fix this: dw xmonad haskell-hgl haskell-x11-xft xmonad-contrib . ALL . -m 'libghc6-haskell-doc (>= 1.4.5-6)' Danke, Joachim -- Joachim "nomeata" B

Packages with Build-Depends/Build-Depends-Indep split

2009-07-24 Thread Joachim Breitner
Hi, you probably have noticed that a new ghc6 version is uploaded. One of the changes is to move .haddock (due to it’s interface changing regularly with the ghc6 releases) to the -dev package, where it can be re-build by a binNMU. This of course requires haddock to be available at install time. K

Re: Bug#538098: [Pkg-haskell-maintainers] Bug#538098: Please upgrade to ≥0.5.1.0.

2009-07-24 Thread Joachim Breitner
Hi, Am Freitag, den 24.07.2009, 12:14 +1000 schrieb Trent W. Buck: > On Thu, Jul 23, 2009 at 10:32:09AM +0200, Joachim Breitner wrote: > >> Package: libghc6-zlib-dev > >> Version: 0.5.0.0-2+b1 > >> Severity: wishlist > >> > >> Please upgrade haskell-zlib in Debian. The upcoming release of > >> Da