Le Thu, Feb 23, 2012 at 10:13:52AM +0100, Andreas Tille a écrit : > On Thu, Feb 23, 2012 at 05:51:59PM +0900, Charles Plessy wrote: > > the two approaches have symmetrical flaws: > > > > - VCS URLs may be wrong, causing some packages to be missed. > > Currently we have a difference of about 30% missing upstream files and > we have not yet developed means to verify the completeness.
Hi again, I reported that I loaded the data for the the pacakges corresponding to the med-bio task, nothing else. Sorry but I did not have much free time this week to do more. I may be uploading and committing frequently during week days, but this is synergistic with my work, while this development on debian/upstream is not, and therefore it only progresses on week-ends. I want first to concentrate on moving the bibliographic data out of the tasks files first. I think that this is the highest priority, and the implementation I have is sufficient for this. Tasks like med-bio are not manageable anymore in my opinion. We are dumping a linear list of packages, but with the bibliographic information, it is very difficult to review and organise their contents. Let's use the current gatherer to solve that problem first. > I would add the following specification for the gatherer: > > 1. Fetches *all* available debian/upstream files > 2. Does this automatically (= without manual triggering) It is not as simple as a two-point list. But you can prove me wrong. Cheers, -- Charles -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/[email protected]
