On 2009-01-26 19:35, Felipe Augusto van de Wiel (faw) wrote: > Actually the statistics is generated based on the number > of files, the assumption is that a translator wouldn't commit a > file "half-translated". In this scenario, if en_US has 8 files > and pt_BR has 6 files with proper pseudo-headers, pt_BR is 75% > translated.
OK, this makes sense if the number of files is relatively high (~150 for d-i) and more or less equally sized. For the current structure of the release notes one would have at least to take the number of lines of each file into account. I added a quick Python hack (transcount) to have at least a rough number: ca: ~0%. (no magic headers yet!) cs: ~0%. (no magic headers yet!) pt_BR: ~11%. -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
