Problem while trying to generate Multilingual PDF

[Srinivas Adusumalli]

Hi,
 
I am facing the routine problem with generating multilingual pdf using Apache 
FOP
 
Requirement: We should be able to generate PDF in Asian/multiple 
languages(Japanese, Chinese etcc.,) using Apache FOP
 
Current Set up: Current system is using Apache FOP to generate PDF in english 
at the runtime. The .fo file (containing information about the blocks and data 
to be written to PDF) and userconfig file(containing information about the 
fonts) content is constructed at the run time based on the data available in 
the Database.
 
Solution Tried so far: After analyzing i found that we need to embedd fonts 
related to japanese or chinese etcc specifically in to the PDF, following are 
the steps I followed to do the same.
 
1) Downloaded latest fop.jar (fop-0.95-bin.zip ) from 
http://mirrors.ecvps.com/apache/xmlgraphics/fop/binaries/ 
2) Created font metric files for ARIALUNI and MS Gothic fonts as specified at 
http://www.zvon.org/HowTo/Output/cid-fonts.pdf 
3) Ensured that the userconfig.xml has the following font entries
 
    <font metrics-file="ARIALUNI.xml"  kerning="yes" embed-file="ARIALUNI.TTF">
        <font-triplet name="Arial Unicode MS" style="normal" weight="normal" />
    </font>
     
    
    <font metrics-file="msgothic.xml"  kerning="yes" embed-file="msgothic.ttc">
        <font-triplet name="MS-Gothic" style="normal" weight="normal"/>
    </font>
    
 4) Ensured that .fo file generated uses the above specified font for the 
certain blocks which contains the Japanese text.
 
Even after performing the above steps, I see the content being displayed as 
 
&#12388;&#12398;&#12467;&#12513;&#12531;&#12488;&#12434;&#12486;&#12473;&#12488;
 
or as ?????????????????????
 
instead of displaying it in Japanese text.
 
NOTE: 
 
1) The japanese text that needs to be displayed is picked from the database, 
and in the database we see the information as 
&#12388;&#12398;&#12467;&#12513;&#12531;&#12488;&#12434;&#12486;&#12473;&#12488;
 when we try to view it using TOAD. I am fetching the same information and 
displaying it in one of the text field in the page where in it is displaying 
appropriately as Japanese text (This is because i am setting the 
charset/pageEncoding to UTF-8).
 
2) There was no "MS Gothic.ttf"  in C:\WINDOWS\fonts dir so i have used "MS 
Gothic.ttc" to generate the font metric file for the same initially and later i 
downloaded MSGothic.ttf from internet and tries the same , and in the 
userconfig.xml I tried to specify both (one at a time) for embed-file property
 
i tried the following entry once 
 
 <font metrics-file="MSGothic.xml"  kerning="yes" embed-file="MSGothic.ttf">
 <font-triplet name="MS Gothic" style="normal" weight="normal"/>
 </font>
 
and then
 
 <font metrics-file="MSGothic.xml"  kerning="yes" embed-file="MSGothic.ttc">
 <font-triplet name="MS Gothic" style="normal" weight="normal"/>
 </font>
 
 
Questions I have:
 
1)  The .fo file created at runtime has the Japanese content as 
&#12388;&#12398;&#12467;&#12513;&#12531;&#12488;&#12434;&#12486;&#12473;&#12488;
 isntead of Japanese characters. Is this fine? Enclosed is the sample 
"Template_user10_.fo" file that is getting generated.
 
2) Enclosed are userconfig.xml and font metric files for MS Gothic and 
ARIALUNI. Check if these are fine.
 
3) Guide/Suggest what all we have to in order to properly write Asian 
languages/Japanese/Chinese to PDF's using Apache FOP. The version of FOP used 
in our application is 0.20.5, but i have used 0.95 version to generate the 
metrics file, will it create a problem? Can you please guide the appropriate 
versions to be used during the whole process.
 
Regards,
Srinivas