* kajaa wrote:
> now, another problem occured,
> when building sitemap.xml, mod/index.xml, mod/directives.xml,
>
> -generate.html:
> [xslt] Processing F:\ami\mod\index.xml.zh-cn to
> F:\ami\mod\index.html.zh-cn.gb2312
> [xslt] Loading stylesheet F:\ami\style\manual.zh-cn.xsl
> [xslt] : Fatal Error! java.io.UTFDataFormatException: Invalid byte 1 of
> 1-byte UTF-8 sequence. Cause: java.io.UTFDataFormatException: Invalid byte 1
> of 1-byte UTF-8 sequence.
> [xslt] Failed to process F:\ami\mod\index.xml.zh-cn
That means, there's an encoding problem (obviously ;-). The new parser
version recognizes wrong byte sequences and throws an error, the former
didn't and converted malformed characters to just more garbage.
Hmm.
You'll need to find the wrong character. It may help to exclude the
autogenerated files from the build process. This can be done by modifying
the build.xml. In the helper.html.generic target, the filset within the
foreach loop is:
<fileset dir="../" includes="**/*${input_ext}">
<patternset refid="excludes" />
</fileset>
Just change it:
<fileset dir="../" includes="**/*${input_ext}"
excludes="mod/index.xml.zh-cn sitemap.xml.zh-cn">
etc. And let's see in which file it really occurs.
By the way, there's another change that you should notice. We've
re-introduced the allmodules.xml files in the mod directory. These contain
the list of modulefiles for the particular language. That way the module
lists are moved out of sitemap. Therefore all internal references to the
sitemap (like &sitemap; in mod/index.xml) have to be changed. Just have a
look at
http://cvs.apache.org/viewcvs.cgi/httpd-docs-2.0/manual/mod/index.xml?rev=1.4.2.3&content-type=text/vnd.viewcvs-markup
for an example.
nd
--
package Hacker::Perl::Another::Just;print
[EMAIL PROTECTED] split/::/ =>__PACKAGE__]}~;
# Andr� Malo # http://www.perlig.de #
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
