Re: DOCBOOK-APPS: Sorting and non-en_US indexes

[Togan Muftuoglu <toganm@users.sourceforge.net>]

* David Cramer; <dcramer@broadjump.com> on 24 Sep, 2002 wrote:
>I think you can rule out the version of Java as an issue. I'm getting it
>to work with java 1.3.1.

I wish I could :-(

>
>I can get it to use Compare_ja in two ways:
>
>java -Duser.language=ja ...

no 
>
>or 
>
>adding lang="ja" (or an attribute value template that gets the same
>thing) to _all_ the xsl:sorts in autoidx.xsl. If you have lang set on
>the xsl:sort, it appears to use that instead of the user.language
>setting passed in from the command line.

no it does not work for me

I have placed a modified (added Turkish) test document
http://toganm.tripod.com/misc/indexterm.002.xml and the following
http://toganm.tripod.com/misc/Compare_tr.java
http://toganm.tripod.com/misc/autoidx_gen.xsl
http://toganm.tripod.com/misc/fop-common.xsl

So if anyone wants to help me before I loose all my hair you're welcomed 

I am using Saxon 6.5.2 along with
SPFACTORY="org.apache.xerces.jaxp.SAXParserFactoryImpl"
DBFACTORY="org.apache.xerces.jaxp.DocumentBuilderFactoryImpl"
org.apache.xml.resolver.tools.ResolvingXMLReader as the catalog resolver


java version "1.3.1_02"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.1_02-b02)
Java HotSpot(TM) Client VM (build 1.3.1_02-b02, mixed mode)

toganm@earth:~/projects/kitap> locale
LANG=tr_TR.UTF-8
LC_CTYPE="tr_TR.UTF-8"
LC_NUMERIC="tr_TR.UTF-8"
LC_TIME=POSIX
LC_COLLATE=tr_TR.UTF-8
LC_MONETARY="tr_TR.UTF-8"
LC_MESSAGES=en_US
LC_PAPER="tr_TR.UTF-8"
LC_NAME="tr_TR.UTF-8"
LC_ADDRESS="tr_TR.UTF-8"
LC_TELEPHONE="tr_TR.UTF-8"
LC_MEASUREMENT="tr_TR.UTF-8"
LC_IDENTIFICATION="tr_TR.UTF-8"


--
Togan Muftuoglu