Page MenuHomePhabricator

Error recognizing pages in uploaded PDF, creating PDF thumbnails and passing it to Wikisource for creating Index pages
Closed, ResolvedPublicBUG REPORT

Description

Steps to replicate the issue:

  • Upload a PDF to Commons to create an Index page in Wikisource.

What happens?:
An error occurs that PDF pages are not scanned in Wikisource. The red error "Ошибка: неправильный интервал" ("Error: incorrect interval") appears in place of the page list on Index page.

What should have happened instead?:
The file must be scanned and a cache created.

An example:

The error started a year ago, a first bug report we had on June 2, 2024.
The file is not always scanned on Commons, unlike in this example. I don't know how files are cached, but it seems to me that it is because of the delay on Commons that sending the cache to Wikisource is delayed. And even after scanning on Commons, as in this example, due to the long period, Wikisource does not request the cache, which is why the error hangs in Wikisource.
I advised users to reset the cache on Commons. But zero editing does not help much, you have to fiddle with it for several minutes: open the page in the editor several times, click preview, even after saving the page the page does not update, and you have to repeat several times. And do the same in this file on Wikisource, and then also on the Index page.

But users, especially newbies and non-techies, do not succeed in all this. Topics about this:

Event Timeline

Vladis13 renamed this task from Error caching PDF and passing cache to Wikisource to Error recognizing pages in uploaded PDF, creating PDF thumbnails and passing it to Wikisource for creating Index pages.Apr 11 2025, 7:08 PM
TheDJ claimed this task.
TheDJ subscribed.

Sometimes this happens when the file servers or the thumbnailing servers are too slow or otherwise unavailable to do the task of creating thumbnails. I purged the page and it was fixed (that is what distinguishes this from another issue with 0x0 pdfs, where purging doesn't fix it)

Thanks, but what you answered, I also answered in user requests. answered once, another, again, again, and so on... But I cannot single-handedly make zero edits so that one of the main functions of Wikisource works - creating Index pages. This is why PDFs are uploaded for Wikisource.
Users, especially newbies and non-techies, don't succeed in this, as I written above.
What do you suggest doing? Personally, I'm tired of doing this. And I suppose that asking you personally to purging every time would be impolite and annoying.

  • Is it possible to categorize PDFs with such an error that persists for more than 1 day or more than a week?
  • And is it possible to configure the bot so that it tries to process such PDF again? And also purge the Index page in Wikisource that uses this file, since an error remains there after purge the file page on Commons.