Search does not work on PDFs with russian font

[Benjamin Eltzner]

This is a copy of a bug reported in Debian:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=750648

As of now it is unclear, wheter the problem is generic. I have requested additional information from the reporter in Debian's BTS and will post any additional information here.

[Adam Reichold (adamreichold)]

Sorry for being too eager to triage this, I confused it with the other Debian bug report.

[Adam Reichold (adamreichold)]

Hello again,

thanks for forwarding. Until further information is given is suspect problems with handling the encoding of the text in the document, maybe related to [1].

Best regards, Adam.

[1] https://bugs.freedesktop.org/show_bug.cgi?id=36111

[Adam Reichold (adamreichold)]

Hello again,

after having a look at the files "100144927890_20140531.pdf" and "ig-jupiter2.djvu", it is clear why one can not search them: Using "djvutext", one finds that the DjVu files does not contain any text layers and hence its text can neither be extracted nor searched. The PDF document does contain a text layer, however the encoding used seems to be problematic. Using "pdftotext" to extract the text yields garbage output without any intelligble russian characters. As the fonts are all embedded, this might be due to an non-standard encoding which Poppler currently can't identify or process.

From this, I'd say that the DjVu issue is related to the documents itself and the PDF issue should go upstream to the Poppler project?

Best regards, Adam.

P.S.: I also tried MuPDF (directly and via our Fitz plug-in) and both did yield the same garbled output.