[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Gathering package upstream meta-data in the UDD. (was: Re: more formally indicating the registration URL)

On Tue, Jan 12, 2010 at 10:38:26AM +0900, Charles Plessy wrote:
> no problem for exporting in JSON, YAML (same as JSON without all those quotes
> and braces), or even XML or SQL pseudo-dumps.

YAML would be fine as well.
> If there is important documentation to read, please let me know. Otherwise, I
> will dig the sources of the UDD.

IMHO the best thing to read is the code at


For something really simple I would advise to read the importer from
data of screenshots.debian.net.  It consists of a script which fetches
the json data:


and a script which imports these data into UDD


This is the typical scheme for all UDD importers but this is a quite
simple one which might properly serve as an example for
upstream-metadata.  In the file sql/screenshots.sql you can see the
table definition for the screenshots - this is actually the point I
tried to discuss here: What will be a proper table layout for the
upstream metadata.

BTW, on blends.debian.net I'm running a "test and play around" copy
of UDD (but it needs some updates because I just realised that the
release type was removed - probably I have to recreate the database
so please drop me a note id you want to do serious work there). I
have also a private script which strips down UDD to a size which fits
on my laptop while featuring enough stuff to play around.
> I will move the source to git.debian.org after adding BOLA boilerplates.


Kind regards



Reply to: