[protege-discussion] Protege 3.4.7 :: UTF-8 HTML Export from Frames based ontology less than optimal

[Richard MAHONEY]

Readers,

I have spent a little while trying to get Protege 3.4.7 to export a
multilingual frames based ontology to UTF-8 HTML. This exercise, although
it should be simple, has been problematic [understatement] and even with my
workaround remains less than optimal.

In order to get the Protege Client to actually export the UTF-8 characters
from a collaborative server based ontology I had to set the following
variable in the `run_protege.sh' script (FreeBSD 8.2-RELEASE with
LANG=en_US.UTF-8):

for:

# ------------------- JVM Options ------------------- 
MAXIMUM_MEMORY=-Xmx500M
OPTIONS=$MAXIMUM_MEMORY

read:

# ------------------- JVM Options ------------------- 
FILE_ENCODING=-Dfile.encoding=UTF8
MAXIMUM_MEMORY=-Xmx500M
OPTIONS="$FILE_ENCODING $MAXIMUM_MEMORY" 

This addition resulted in the CJK and Arabic characters making it into the
HTML output. What it didn't do, unfortunately, was to update the character
set encoding of the exported HTML files:

They currently read:

<meta http-equiv="content-type" content="text/html; charset=iso-8859-1"/>

when they should read:

<meta http-equiv="content-type" content="text/html; charset=utf-8" />


So to my question ... How is it possible to configure Protege 3.4.7 so that
the HTML headers are appropriate for UTF-8 output?


Kind regards,

 Richard 




-- 
Richard MAHONEY - Indica et Buddhica

Littledene, Bay Road, OXFORD 7430, NZ
Tel.: +64 3 312 1699
Handheld: +64 210 640 216
r.mahoney at indica-et-buddhica.org

http://indica-et-buddhica.org
http://camera-antipodea.indica-et-buddhica.com