Closed Popolechien closed 1 month ago
The limit is a recommendation of the spec. See log.
We've decided to enforce our own recommendation at scraperlib's level.
What's left is:
Update zimfarm to enforce limits at ZF edition level
This would probably be the most efficient
As explained in another ticket ; we'd still need to discuss and adapt the scrapers that uses dynamic metadata: YT, TED for instance
@rgaudin
As explained in another ticket ; we'd still need to discuss and adapt the scrapers that uses dynamic metadata: YT, TED for instance
AFAIK this part has been achieved meanwhile. @benoit74 you conform, otherwise can you please list the blocking issues?
Now TED and Youtube are not creating anymore ZIMs with titles longer than 30 characters, this is true, and this is correct that it should be reflected in the Zimfarm configuration.
We probably have other scrapers in the same situation, to be checked when this issue will be implemented.
I propose to move this to prio1
(it is already prio1, sorry ^^)
Removed from zimit2 project unfortunately
I've moved the part around displaying current number of characters used in https://github.com/openzim/zimfarm/issues/1025
Creating a zim file whose title is longer than 30 characters will cause the job to crash. This should either be explicitely indicated in the
Title
field (with a small counter to indicate where one is) or if there is no particular reason for this limit then extend to something along the lines of 60 characters.