A new version of Bengali Train data released just 4 month before. Could any one update for Bengali Wikisource?
https://github.com/tesseract-ocr/tessdata/blob/master/ben.traineddata
Last time Uer:Phe update/improve bengali OCR, https://bn.wikisource.org/wiki/user_talk:Jayantanth#OCR, but now OCR Button not appeared at edit mode.
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | None | T117711 OCR scripts need updating at tools labs by updating the "tesseract-ben" package | |||
| Invalid | None | T118643 Backport newer tesseract-ocr-* packages from sid |
Not an bug with the Proofread Extension, but a bug regarding an gadget called OCR.js and the OCR tool at labs, which is being used on most wikisource wikis.
Regarding the problem of not having an OCR button, the source of that issue is this edit - as the link to said OCR.js gadget was commented out, and no longer works because of that.
Jayantanth should remove the two slashes at the start of line 27 in Mediawiki:Common.js and the OCR button should show up again.
I can not comment on the other issue mentioned in this thread - which is the lack of updates to the bengali OCR in the labs OCR tool. That issue can only be solved by Phe or Tpt, as they are the maintainers.
The javascript file has been fixed at bnWS. The update of the tools labs library is an update needed at tools, and I am not sure whether that is for @Phe himself, or something that needs more generally updating.
Tool Labs has these files installed via apt-get: http://packages.ubuntu.com/trusty/tesseract-ocr-ben , but the last release was in 2012 (!). More recent debian releases do have newer versions (e.g. https://packages.debian.org/source/sid/tesseract-ben).