Cannot copy text from specific pdf in evince

[drpjkurian]

Binary package hint: evince

1) lsb_release -rd
Description: Ubuntu Vivid Vervet (development branch)
Release: 15.04

2) apt-cache policy evince
evince:
  Installed: 3.14.1-0ubuntu1
  Candidate: 3.14.1-0ubuntu1
  Version table:
 *** 3.14.1-0ubuntu1 0
        500 http://us.archive.ubuntu.com/ubuntu/ vivid/main amd64 Packages
        100 /var/lib/dpkg/status

3) What is expected to happen via https://bugs.launchpad.net/ubuntu/+source/evince/+bug/545176/+attachment/1239564/+files/emrscheme.pdf is that one may select any of the text, just like Adobe Reader.

4) What happens instead is only a handful of letters are selectable as per https://launchpadlibrarian.net/41742091/Screenshot-2.png .

ProblemType: Bug
Architecture: i386
Date: Tue Mar 23 21:09:22 2010
DistroRelease: Ubuntu 9.10
ExecutablePath: /usr/bin/evince
Package: evince 2.28.1-0ubuntu1.2
ProcEnviron:
 LANGUAGE=en_IN.UTF-8
 PATH=(custom, user)
 LANG=en_IN.UTF-8
 SHELL=/bin/bash
ProcVersionSignature: Ubuntu 2.6.31-20.58-generic
SourcePackage: evince
Uname: Linux 2.6.31-20-generic i686

[madbiologist (me-again)]

Can you attach the problem document to this bug report?

[drpjkurian (drpjkurian-gmail)]

Hi
I am using Karmic Kaola with the latest kernel
When I try to copy the text from a paragraph in a pdf file, The programme selects few random alphabets (usually the first letter of the word)from the paragraph. I was not able to select the whole paragraph to copy.

I hope this is what you meant(problem document). Sorry If I'am wrong

[madbiologist (me-again)]

The image you attached is useful. It shows us the problem you are experiencing, more clearly than a written description. Can you also attach the PDF file so that we can do some testing?

[drpjkurian (drpjkurian-gmail)]

Hi
Please find herewith the attached copy of the .pdf file

With regards
Dr Kurian

[madbiologist (me-again)]

Confirmed on Ubuntu 10.04 "Lucid Lynx" alpha 3 with updates.

Package versions: evince 2.29.92-0ubuntu1
                              poppler 0.12.4-0ubuntu2
                              libcairo2 1.8.10-2ubuntu1

[Gianfranco Costamagna (costamagnagianfranco)]

Still confirmed with evince 2.30 on lucid beta 2

[drpjkurian (drpjkurian-gmail)]

The problem still persist in Lucid stable version

[madbiologist (me-again)]

Still occurring on Maverick.

uname: Linux 2.6.35-22-generic i686
poppler 0.14.3-0ubuntu1
evince 2.32.0-0ubuntu1

[Sitaram Shelke (sitaramshelke)]

Is the PDF format correct?
I have tried to open the PDF with Ubuntu-evince ,Ubuntu-okular and Mac-default PDF Reader and I have attached respective screenshots if you want to take a look at it. All ther reader gives different selection with none of them perfect.

[penalvch (penalvch)]

Created attachment 118948
emrscheme-1.pdf

Upstreaming as advised in:
https://bugzilla.gnome.org/show_bug.cgi?id=741623

Downstream report:
https://bugs.launchpad.net/ubuntu/+source/evince/+bug/545176

lsb_release -rd
Description: Ubuntu Wily Werewolf (development branch)
Release: 15.10

apt-cache policy poppler-utils
poppler-utils:
  Installed: 0.33.0-0ubuntu3
  Candidate: 0.33.0-0ubuntu3
  Version table:
 *** 0.33.0-0ubuntu3 0
        500 http://us.archive.ubuntu.com/ubuntu/ wily/main amd64 Packages
        100 /var/lib/dpkg/status

What is expected to happen with Evince is that when one opens the attached PDF file, one may select any of the text, just like in Adobe Reader.

What happens instead is only a handful of letters are selectable as per https://launchpadlibrarian.net/41742091/Screenshot-2.png . First reported against Ubuntu 9.10 evince 2.28.1-0ubuntu1.2 / poppler-utils 0.12.4-0ubuntu4.

[Jason Crain (jcrain)]

I don't see why it would be useful to select text from this document. You aren't going to be able to copy and paste any readable text from it. The document doesn't use a standard encoding or include a way to map from charcode to text, so at best you're going to get a bunch of random letters and control characters.

[penalvch (penalvch)]

Jason Crain, thanks for taking a look. Given your point about the documents encoding is confirmed with Windows 10's built-in PDF reader, this is considered closed.

[penalvch (penalvch)]

drpjkurian, as per upstream:
"The document doesn't use a standard encoding or include a way to map from charcode to text"

Hence, the root cause here resides with how the original author created the PDF, versus a software bug in a PDF viewer.