[ckan-dev] Openness checking: request for review

Friedrich Lindenberg friedrich at pudo.org
Sat Nov 27 09:20:56 UTC 2010

Hey Olly, 

these changes look very good, altough I assume we'll have to find a more dynamic way to classify MIME types in the future. Maybe the general HTTP stuff could also go into a common lib, such as swiss. 

This could also be a first good application of the workers: having a link checker that just runs off the queue and once a day, submitting some info back into CKAN (HEAD status, content-length, ...). Getting fancy, this information would be stored in PackageResourceExtras, not PackageExtra, but that's probably for another time. 


On Nov 26, 2010, at 1:00 PM, Oliver Cope wrote:

> Hi all,
> I'm back on my feet again at last and have just pushed my code for issue #277 (openness checking) to my bitbucket repo for review. Here are the changesets:
>  http://bitbucket.org/ollyc/ckan/changeset/d365a3396761
>  http://bitbucket.org/ollyc/ckan/changeset/c179383dd0b8
>  http://bitbucket.org/ollyc/ckan/changeset/1b16fbe9aa65
>  http://bitbucket.org/ollyc/ckan/changeset/5ceef55e1334
> This implements all the core functions for checking, plus a cli script for manually updating all scores. I haven't implemented the worker, but this should now be very straightforward to add.
> Also mercurial forced me to manually resolve a conflict before letting me push. I'm baffled as to why, because it's in a file I have never touched:
>  http://bitbucket.org/ollyc/ckan/changeset/888ec23b7dbf
> I have no idea if I picked the right version of the affected line or not, so maybe I've reverted someone's edit. Apologies if that's the case.
> Olly.
> -- 
> Oliver Cope, Unicorn sprl
> http://www.ollycope.com/
> +44 20 8133 1190

More information about the ckan-dev mailing list