couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dave Cottlehuber <>
Subject Re: State of Debian Packaging
Date Thu, 20 Jun 2013 21:02:43 GMT
On 20 June 2013 19:53, Randall Leeds <> wrote:
> I saw that my packaging efforts came up in the IRC meeting yesterday
> and I wasn't there to report.
> Here's what's done:
>  - Imported the most up-to-date .dsc from unstable using git-buildpackage tools
>  - Imported the rc.2 release tarball that passed vote
>  - Updated to use the virtual libcurl-dev, so it should work with the
> gnutls or openssl version
>  - Split the packaging into couchdb and couchdb-bin as Ubuntu did downstream
>  - Dropped patches now included upstream
>  - Published to github at
> You can try it by installing git-buildpackage, cloning the pkg-couchdb
> repository, using git-dch to create a new debian/changelog entry for
> 1.3.1, and running git-buildpackage.
> Here's what remaining to be done, based on my notes from previous
> conversations with Laszlo and others:
>  - Use system libyajl-dev
>  - Use system libspeedy-dev
>  - Use system libjs-json
>  - Use system libjs-coffeescript
>  - Publish to
> The first 4 are Debian policy issues. As far as I'm concerned, it's
> not critical to fix this before we publish a .deb on the Apache
> mirrors. Doing this work ourselves would mean that Laszlo doesn't have
> to and we'd have Debian packaging in a repo that's ready for the
> official repo distribution.
> For the last, please advise what the best practice is. Is there an
> official place within our project that we're currently storing the
> windows and mac binary packaging or are you all doing that outside the
> ASF and just uploading the result?

I'd like to merge my windows build scripts into couchdb proper (I will
open a JIRA for that), and during the release process, we currently
stash binaries / packages into svn

> I opened these as issues on the github repo I set up.
> Laszlo: If you already have a repo for this work anywhere that you
> would prefer I use I would be glad to collaborate there.

View raw message