On Thu, Sep 11, 2003 at 10:56:28AM +0200, Gerfried Fuchs wrote: > Hi! > > Something did go wrong with the recent changs to the l10n pages. They > produce tons of output in the tidy/validate analyses[1]. From what I can > see it is because unescaped translator names that are taken 1:1 into the > corresponding pages. > > Can someone with enough l10n/i18n knowledge fix that? Would be really > nice, thanks. > > So long, > Alfie > [1] http://www-master.debian.org/~barbier/tidy/ and s/tidy/validate/
Yup, we're on it. Denis did detect this quite quickly, and Tomohiro KUBOTA did provide us the Perl code sniplet which will fix the issue. It was discussed off list because one of us did switch to private email by error, sorry for not informing you about that. It only has to be done. Denis did say that he may fix the issue, so I leave him a bit more time. He cannot mail at day and I cannot mail at night, so coordinating our efforts is a bit difficult and takes some days. But if he don't want to do this, I'll patch the data extraction scripts happily. The point is that those scripts run outside of my intervention range: I'm not DD, and I cannot remove the DB if it become corrupted because of my broken commits against the scripts. That makes me kind of reluctant to change these points without Denis's approval and review. Thanks for reporting anyway, Mt. -- Let's call it an accidental feature. -- Larry Wall
