Bug 135239 - Bad iso8859-2 encoding
Summary: Bad iso8859-2 encoding
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Fedora
Classification: Fedora
Component: gnuplot
Version: 6
Hardware: i386
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Ivana Varekova
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2004-10-11 09:35 UTC by Dariusz Knociński
Modified: 2007-11-30 22:10 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2007-05-21 13:17:55 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
Polish chars test file (71 bytes, application/octet-stream)
2005-10-20 20:58 UTC, Marcin Garski
no flags Details

Description Dariusz Knociński 2004-10-11 09:35:08 UTC
Description of problem:
Gnuplot generate bad format for PostScrip document wiht
iso8859-2 encoding and no generate Polish letters on X11.

Version-Release number of selected component (if applicable):
gnuplot-4.0.0-3

How reproducible:
Always

Steps to Reproduce:
1.Create sample file from gnuplot samples
2.Insert section "set encoding iso_8859-2" into top of document
3.Realize graph to X11 window or PostScript file
  
Actual results:


Expected results:


Additional info:

Comment 1 Phil Knirsch 2004-10-18 12:22:44 UTC
I've tried what you described and got perfectly working postscript
files afterwards.

What exactly did you try and what was the error message?

Read ya, Phil

Comment 2 Marcin Garski 2005-10-20 20:57:04 UTC
I probably know why Dariusz filed this bug.

Save attached file and run 'gnuplot polish-letters-sample.gp'. You won't see
polish characters in title. The file have ISO-8859-2 encoding.

When I recode this file to UTF-8 and remove set encoding line I still can't get
polish chars on screen.

Comment 3 Marcin Garski 2005-10-20 20:58:33 UTC
Created attachment 120207 [details]
Polish chars test file

Tested on gnuplot-4.0.0-7 (FC4)

Comment 4 John Thacker 2007-01-13 19:06:44 UTC
The short answer to this problem is that gnuplot 4.0 is badly broken with
respect to UTF-8.  It's basically always choosing an ISO-8859-1 font.

The first type of bad display is what happens when gnuplot attempts to display
ISO-8859-2 characters by choosing the correct codepoints, but from an ISO-8859-1
font.  The second type of bad display (after converting to UTF-8) is that
gnuplot is again using an ISO-8859-1 font and interpreting the codepoints as
codepoints in ISO-8859-1, but this time some of the characters are multibyte,
and in any case the bytes are different.

gnuplot 4.1/4.2 (in CVS) fixes a lot of this:
http://statist.wald.intevation.org/utf8.html
http://gnuplot.cvs.sourceforge.net/gnuplot/gnuplot/demo/utf8.dem?revision=1.3&view=markup

4.2 is supposed to be out very shortly, which should really make solving the
issue a lot easier (though not perfect), by adding UTF-8 support.  In fact, the
last minute comments before final release on the development mailing list for
gnuplot address this very issue of switching between different font encodings.

I'd say that this cannot be fixed until 4.2 is released.

Comment 5 Ivana Varekova 2007-05-21 13:17:55 UTC
Fixed with gnuplot-4.2.0-1.fc8.


Note You need to log in before you can comment on or make changes to this bug.