Bug 334692 - Missing characters (French é è à) when document printed ; appear correctly on the screen ; printed correctly in evince
Summary: Missing characters (French é è à) when document printed ; appear correctly on...
Status: RESOLVED UPSTREAM
Alias: None
Product: okular
Classification: Applications
Component: printing (show other bugs)
Version: 0.19.0
Platform: Kubuntu Linux
: NOR normal
Target Milestone: ---
Assignee: Okular developers
URL: https://www.formulaires.modernisation...
Keywords:
Depends on:
Blocks:
 
Reported: 2014-05-13 08:29 UTC by Mahendra Tallur
Modified: 2014-05-24 07:09 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In:


Attachments
Page 1 of test case printed then scanned back (164.96 KB, image/jpeg)
2014-05-14 18:14 UTC, Mahendra Tallur
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Mahendra Tallur 2014-05-13 08:29:56 UTC
Issue triggered with the official French administration document that can be found here : https://www.formulaires.modernisation.gouv.fr/gf/cerfa_14952.do

Most characters with accents do appear correctly on the screen but are NOT printed, when using Okular (replaced by a blank character). They are printed correctly with evince on the very same machine (Kubuntu 14.04 LTS 32 bits).

Reproducible: Always

Steps to Reproduce:
1. Open specificed PDF file in  Okular
2. Print
3. Compare screen with printed documents
Actual Results:  
Accents are missing on the printed version.

Expected Results:  
Printed and onscreen version should be identical.

I didn't figure out the factors that trigger the issue. It happens all the time with this document, though. As it's an official administration document, I think it's pretty important it works (many similar ones can be downloaded for different purposes).

It works with Evince.

Using : Kubuntu 14.04 LTS 32 bits, KDE 4.13.0. No PPA used, clean install.
Comment 1 Albert Astals Cid 2014-05-13 20:36:31 UTC
You mean things that are part of the document like "Prénom" or contents you fill in in the Forms of the document?
Comment 2 Mahendra Tallur 2014-05-13 20:54:50 UTC
Sorry, I didn't specify. Parts of the documents ; I didn't take the form into account at all.
Comment 3 Albert Astals Cid 2014-05-13 21:00:58 UTC
Do you see them in print preview? Can you specify which parts of the document specially? Can you take a photo of the printed page and attach it here?
Comment 4 Mahendra Tallur 2014-05-14 04:44:55 UTC
Characters do appear properly in print preview. (or onscreen after PDF/PS export)

I'll take a picture ASAP ; on the printed document, all "é" "à" "è" etc. disappear (except, for some reason, on the capital letters on the second page).

Thanks a lot :) I'll get back to you with a scanned version.
Comment 5 Mahendra Tallur 2014-05-14 18:14:52 UTC
Created attachment 86631 [details]
Page 1 of test case printed then scanned back
Comment 6 Albert Astals Cid 2014-05-20 21:09:06 UTC
Hi, this is correct, i have confirmed that in my printer the é from Prénom is not printerd either.

I am closing this bug as resolved upstream because Okular is not the one at fault here, what we are doing while printing is basically
pdftops myfile.pdf
lpr myfile.ps
that is, convert the file from pdf to ps and then print it.

Since pdftops is creating a ps file that gs shows with the é i am unsure if the bug is on pdftops or in lpr/the printer software.

What i know it is that it's not in Okular.

I suggest opening a bug in the poppler project at https://bugs.freedesktop.org saying something along the lines "ps created by pdftops in this pdf file does not get printed correctly" and see if they think the bug is on their side or in lpr's side.

Thanks for caring about Okular :)
Comment 7 Mahendra Tallur 2014-05-21 09:45:40 UTC
Hi ! Thanks for your reply ! :)

I have one last question : what would evince do differently (pdftops / lpr ?) ? As this bug doesn't occur on the very same machine / system / printer / driver, with evince.

Also our printer are probably no the same at all (Samsung ML 1650 here) ?

Thanks again, cheers, 
Mahen
Comment 8 Albert Astals Cid 2014-05-23 18:18:55 UTC
Because postscript is a very complex language and you can construct the same thing in a billion ways, none of them better or worse than the rest.
Comment 9 Mahendra Tallur 2014-05-24 07:09:36 UTC
Bug reported upstream as suggested : https://bugs.freedesktop.org/show_bug.cgi?id=79168

Thanks, I'll keep you informed !