- cabal-devel - Haskell.org

[Hackage] #500: develop a tool to find the maximal consistent set of hackage packages
by Hackage 15 Feb '09

15 Feb '09

#500: develop a tool to find the maximal consistent set of hackage packages ------------------------------+--------------------------------------------- Reporter: duncan | Owner: Type: enhancement | Status: new Priority: normal | Milestone: Component: miscellaneous | Version: Severity: normal | Keywords: Difficulty: project(> week) | Ghcversion: Platform: | ------------------------------+--------------------------------------------- To a great extent the point of libraries is code re-use. It follows that is it useful to be able to install and use two packages simultaneously. The extension of this is that we would like to know the maximal set of packages that we can install consistently. Them being consistent implies that we can use any of them in another package and have consistent dependencies. If we knew a package was within the maximal subset then we can give it a positive mark on hackage. We may also be able to encourage package authors to update their dependencies to as to get their packages within this set. Calculating the maximal consistent subset is certainly an NP-complete problem. It is at least as hard as working out if any single package is installable and that is already NP-complete. However we do not necessarily have to get an exact maximal set, approximation would probably do. We would also like to influence the choice of the set by the importance of packages, especially when it comes to resolving choices about divisive packages (see #499). We would also like to influence the choices by manually supplied preferences. This can probably be implemented as an extension of the standard package dependency resolution algorithm and using the code developed for ticket #499. -- Ticket URL: <http://hackage.haskell.org/trac/hackage/ticket/500> Hackage <http://haskell.org/cabal/> Hackage: Cabal and related projects

1 5

[Hackage] #499: develop tool to analyse package dependency splits
by Hackage 14 Feb '09

14 Feb '09

#499: develop tool to analyse package dependency splits ------------------------------+--------------------------------------------- Reporter: duncan | Owner: Type: enhancement | Status: new Priority: normal | Milestone: Component: miscellaneous | Version: Severity: normal | Keywords: Difficulty: project(> week) | Ghcversion: Platform: | ------------------------------+--------------------------------------------- Looking at packages on hackage as a whole we often see cases where different packages require mutually incompatible versions of some dependency. The most notable cases at the present time are parsec 2 vs 3, HaXml 1.13 vs 1.19, HTTP 3000 vs 4000 etc. What we want to know is, which dependencies are the divisive ones. In such cases is there a clear winner? Which are the most divisive packages? A divisive package is one where looking at the packages that depend on it, the intersection of versions constraints is empty. That is we cannot configure all the other packages on hackage without them disagreeing about the devisive package. When we say all, we really only mean the latest recommended version of each package, not all versions of all packages. Looking at the packages depending on a divisive package we may see a clear consensus that can resolve the issue if we ignore the minority. Frist we must discover the 'camps', that is the sets of packages that can agree on versions of the dependency. For example it might be simply two camps, `pkg < 3` vs `pkg >= 3`. Then for each camp we score it based on the importance of the packages in that camp. The importance of packages is calculated separately (based on downloads and importance of dependents). The camp with the biggest score wins. Some packages are more divisive than others. A package is more divisive if the the loosing camps are still quite important. That corresponds to excluding many packages from the maximal consistent package set. It would be useful to make a ranking of the most divisive packages because these correspond to cross-package problems that need resolution in a more coordinated fashion and their impact is bigger. It would also be useful to track trends in the balance of the camps for divisive packages. It may be that such splits can be naturally resolved. If there is a clear trend then we can simply make that the de-facto recommendation. Imagine a summary page on hackage for platform managers to see what the most divisive packages are and the trends for each one. It would help them and maintainers to make decisions on recommendations for dependencies. -- Ticket URL: <http://hackage.haskell.org/trac/hackage/ticket/499> Hackage <http://haskell.org/cabal/> Hackage: Cabal and related projects

1 1

patch applied (cabal): Validate the docbook xml before processing.
by Duncan Coutts 13 Feb '09

13 Feb '09

Fri Feb 13 05:41:36 PST 2009 Duncan Coutts <duncan(a)haskell.org> * Validate the docbook xml before processing. Apparently xsltproc does not validate against the dtd. This should stop errors creaping back in. M ./Makefile -1 +6 View patch online: http://darcs.haskell.org/cabal/_darcs/patches/20090213134136-adfee-90fdd69f…

1 0

Re: [Haskell-cafe] Help needed with apache config for hackage
by Duncan Coutts 13 Feb '09

13 Feb '09

On Fri, 2009-02-13 at 22:39 +1100, George Pollard wrote: > On Fri, 2009-02-13 at 10:58 +0000, Duncan Coutts wrote: > > Hi folks, > > > > Does anyone have any experience with apache configuration, particularly > > mime types and handling browser quirks and would like to help us with an > > issue we have on hackage? > > > > http://hackage.haskell.org/trac/hackage/ticket/498 > > > > The problem is described in the ticket but basically IE6 gets confused > > by the Content-Type and Content-Encoding and ends up saving ".tar.gz" > > files with the wrong name ".tar.tar". > > > > We need help working out how to configure apache to use a workaround and > > with testing that the solution actually works. > > > > Thanks! > > > > Duncan > > Currently the browser receives: > > > HTTP/1.1 200 OK > > Date: Fri, 13 Feb 2009 11:15:22 GMT > > Server: Apache/2.2.3 (Debian) > > Last-Modified: Mon, 09 Feb 2009 07:55:57 GMT > > ETag: "38c010-46d-b361bd40" > > Accept-Ranges: bytes > > Content-Length: 1133 > > Content-Type: application/x-tar > > Content-Encoding: x-gzip > > You could try adding a Content-Disposition header to specify a file name: > > Content-Disposition: attachment; filename=APackage.tar.gz > > In Apache you can (apparently [1],[2]) do it like this: > > RewriteRule "^packages/archive/[^/]+/[^/]+/(.+)$" - [env=pkgname:$1] > Header set Content-Disposition "attachment; filename=\"%{pkgname}e\"" env=pkgname > > [1]: http://httpd.apache.org/docs/2.0/mod/core.html#files > [2]: http://www.experts-exchange.com/Software/Server_Software/Web_Servers/Apache… Can we do that just for one user agent? I don't think we want to use non-standard stuff in general. Apparently Content-Disposition is not in the official HTTP spec, but IE is known to follow it. Duncan

1 0

patch applied (cabal): Make documentation validate
by Duncan Coutts 13 Feb '09

13 Feb '09

Thu Feb 12 15:50:57 PST 2009 Samuel Bronson <naesten(a)gmail.com> * Make documentation validate M ./doc/Cabal.xml -17 +21 View patch online: http://darcs.haskell.org/cabal/_darcs/patches/20090212235057-90c7a-a7f25535…

1 0

patch applied (cabal): Folly the directions for docbook-xsl
by Duncan Coutts 13 Feb '09

13 Feb '09

Thu Feb 12 18:26:15 PST 2009 Samuel Bronson <naesten(a)gmail.com> * Folly the directions for docbook-xsl As it says in http://docbook.sourceforge.net/release/xsl/current/README: - Use the base canonical URI in combination with one of the pathnames below. For example, for "chunked" HTML, output: http://docbook.sourceforge.net/release/xsl/current/html/chunk.xsl M ./Makefile -1 +1 View patch online: http://darcs.haskell.org/cabal/_darcs/patches/20090213022615-90c7a-bf669086…

1 0

Help needed with apache config for hackage
by Duncan Coutts 13 Feb '09

13 Feb '09

Hi folks, Does anyone have any experience with apache configuration, particularly mime types and handling browser quirks and would like to help us with an issue we have on hackage? http://hackage.haskell.org/trac/hackage/ticket/498 The problem is described in the ticket but basically IE6 gets confused by the Content-Type and Content-Encoding and ends up saving ".tar.gz" files with the wrong name ".tar.tar". We need help working out how to configure apache to use a workaround and with testing that the solution actually works. Thanks! Duncan

1 0

[Hackage] #498: package downloads confuse IE6
by Hackage 13 Feb '09

13 Feb '09

#498: package downloads confuse IE6 --------------------------------+------------------------------------------- Reporter: duncan | Owner: Type: defect | Status: new Priority: normal | Milestone: Component: hackageDB website | Version: Severity: normal | Keywords: Difficulty: normal | Ghcversion: Platform: Windows | --------------------------------+------------------------------------------- In IE6 the `.tar.gz` packages from the Hackage website get downloaded and saved by the browser with the wrong name. For example `foo-1.0.tar.gz` gets saved as `foo-1.0.tar.tar`. Currently the apache config for the hackage website is such that it send these headers for `.tar.gz` files: {{{ Content-Type: application/x-tar Content-Encoding: x-gzip }}} This apparently confuses IE6. This has been [http://mail- archives.apache.org/mod_mbox/httpd- dev/200302.mbox/%3c5.2.0.9.2.20030204125407.02768620@pop3.rowe- clan.net%3e. reported elsewhere previously]. So we are looking for help in finding a workaround. We want to adjust our apache config so that for the IE6 user agent it reports a mime time and encoding that will make IE6 save the file properly. For example we guess that sending `Content-Type: application/x-gzip` and no `Content-Encoding` would make IE6 do the right thing. So how do we actually configure apache to do this? Can we get some help testing it? -- Ticket URL: <http://hackage.haskell.org/trac/hackage/ticket/498> Hackage <http://haskell.org/cabal/> Hackage: Cabal and related projects

1 0

darcs patch: Folly the directions for docbook-xsl
by Samuel Bronson 13 Feb '09

13 Feb '09

Thu Feb 12 21:26:15 EST 2009 Samuel Bronson <naesten(a)gmail.com> * Folly the directions for docbook-xsl As it says in http://docbook.sourceforge.net/release/xsl/current/README: - Use the base canonical URI in combination with one of the pathnames below. For example, for "chunked" HTML, output: http://docbook.sourceforge.net/release/xsl/current/html/chunk.xsl

1 0

darcs patch: Link to #configurations from #system-dependant in the ...
by Samuel Bronson 13 Feb '09

13 Feb '09

Thu Feb 12 19:03:14 EST 2009 Samuel Bronson <naesten(a)gmail.com> * Link to #configurations from #system-dependant in the Docbook documentation

1 0