Cabal dependency extrema testing

[Roman Cheplyaka]

First of all, I don't think libraries@ is the right place for this discussion.
You should probably move it to -cafe until you have a concrete proposal
for libraries changes.

To answer your question: you're interpreting the results wrong.
If cabal found a way to build the package with these constraints and the
build itself failed, that would be an argument for restricting bounds.

In this case cabal simply can't find a suitable install plan, which
means that *your* constraints are incompatible. That *probably* means
(although it's easy to come up with a contrived counterexample) that
dependencies can be made a bit tighter, but it's not a maintainer's task
to analyse compatibility of various packages — that's what cabal is for.

Maintainer only specifies constraint within which the package is
supposed to build. The only way such constraints can be proven wrong is
showing a build within the constraints which fails.

Roman

* Bram Neijt <bneijt at gmail.com> [2013-06-16 00:18:24+0200]
> Ok, so I had my first go at this, but it seems all the things I try
> will fail. My approach was simple: take a single dependency and change
> ">=" into "==" for that single dependency, then try to build. Do that
> for every dependency in the cabal file.
> 
> Turns out, none of the things I tries would work. Here is a concrete example:
> mkdir test
> cd test
> snap init
> #Edit the test.cabal file and change bytestring >= 0.9.1 to bytestring == 0.9.1
> cabal fetch .
> 
> This fails, telling me:
> $ cabal fetch .
> Resolving dependencies...
> cabal: Could not resolve dependencies:
> trying: test-0.1 (user goal)
> next goal: xmlhtml (dependency of test-0.1)
> rejecting: xmlhtml-0.2.0.3/installed-4c6... (conflict: xmlhtml =>
> bytestring==0.10.0.1/installed-7e2..., test => bytestring==0.9.1 && <0.11)
> trying: xmlhtml-0.2.1
> trying: unordered-containers-0.2.3.0/installed-7fe... (dependency of
> xmlhtml-0.2.1)
> next goal: hashable (dependency of
> unordered-containers-0.2.3.0/installed-7fe...)
> rejecting: hashable-1.2.0.2/installed-bbd... (conflict: unordered-containers
> => hashable==1.1.2.5/installed-d93...)
> rejecting: hashable-1.1.2.5/installed-d93... (conflict: hashable =>
> bytestring==0.10.0.1/installed-7e2..., test => bytestring==0.9.1 && <0.11)
> rejecting: hashable-1.2.0.10, 1.2.0.9, 1.2.0.8, 1.2.0.7, 1.2.0.6, 1.2.0.5,
> 1.2.0.4, 1.2.0.3, 1.2.0.2, 1.2.0.1, 1.2.0.0, 1.1.2.5, 1.1.2.4, 1.1.2.3,
> 1.1.2.2, 1.1.2.1, 1.1.2.0, 1.1.1.0, 1.1.0.0, 1.0.1.1, 1.0.1.0, 1.0.0
> (conflict: unordered-containers => hashable==1.1.2.5/installed-d93...)
> 
> 
> Am I donig something wrong, or does this just mean that the bytstring
> dependency should be fixed to something above 0.9.1?
> 
> Greetings,
> 
> Bram
> 
> On Wed, Jun 12, 2013 at 9:29 AM, Erik Hesselink <hesselink at gmail.com> wrote:
> > On Wed, Jun 12, 2013 at 12:16 AM, Roman Cheplyaka <roma at ro-che.info> wrote:
> >> * Roman Cheplyaka <roma at ro-che.info> [2013-06-12 01:10:56+0300]
> >>> * Roman Cheplyaka <roma at ro-che.info> [2013-06-12 01:05:29+0300]
> >>> > * Bram Neijt <bneijt at gmail.com> [2013-06-11 23:15:39+0200]
> >>> > > Hi all,
> >>> > >
> >>> > > I think it would be interesting to test build a package with the
> >>> > > "latest" and "oldest" allowed dependencies, to see if the dependency
> >>> > > boundaries hold. For example, build the text[1] package with version
> >>> > > 0.2 of test-framework-hunit (oldest allowed according to .cabal file)
> >>> > > to see if that is still works.
> >>> > >
> >>> > > Does anybody know of previous work on that?
> >>> > > Does anybody have pointers on how to build with a specific version set?
> >>> > > What is the easiest way to parse the dependencies in the cabal file,
> >>> > > API pointers are also very much appreciated :)
> >>> >
> >>> > I believe Bram Schuur (CC'd) worked on this during OdHac.
> >>>
> >>> Trying his another email...
> >>
> >> Still no luck. This is his github account, maybe try to reach him there:
> >> https://github.com/craffit
> >
> > I believe this address still works.
> >
> > Regards,
> >
> > Erik
> >
> > _______________________________________________
> > Libraries mailing list
> > Libraries at haskell.org
> > http://www.haskell.org/mailman/listinfo/libraries
> 
> _______________________________________________
> Libraries mailing list
> Libraries at haskell.org
> http://www.haskell.org/mailman/listinfo/libraries