-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
My bet:
param.xsl:<xsl:param name="draft.watermark.image"
select="'http://docbook.sourceforge.net/release/images/draft.png'"/>
Chris Chiasson a 嶰rit :
> after further diagnosis using tcpdump (adding filters to view less traffic)
>
> tcpdump -e ip and not host (my workstation's ip)
>
> I found this in the output when I ran my ant script and java forked to
> saxon:
>
> 04:32:13.623349 00:13:d4:cc:94:92 (oui Unknown) > 00:12:43:a9:bc:75
> (oui Unknown), ethertype IPv4 (0x0800), length 74:
> wsip-70-182-153-37.br.br.cox.net.49543 >
> projects.sourceforge.net.http: S 2632400219:2632400219(0) win 5840
> <mss 1460,sackOK,timestamp 16154975 0,nop,wscale 4>
>
> So, I am reasonably sure that it is trying to load something from
> source forge. I would guess that this means I have a problem with my
> catalogs under java. However, I am unsure how to fix that. Any
> suggestions?
>
> On 9/7/06, Chris Chiasson <> wrote:
>> In trying to diagnose this problem, I switched to saxon. I am having
>> trouble with it as well (both the saxon 6 and saxon 8 versions). It
>> hangs when processing the stylesheet, though saxon 8 does at least
>> print one warning that I am processing an XSLT 1 document with an XSLT
>> 2 processor. I have saxon, xml-commons, xml-commons-resolver, and
>> xml-commons-external-1.3 in the classpath (it's dynamically generated
>> from package.env files with these names). I think the hang is coming
>> from the processor or parser trying access the internet, but I wasn't
>> able to verify this using tcpdump, and I couldn't find an option to
>> have saxon disable net access. I will say that xsltproc's --nonet
>> option works just fine with this combination of xml source and
>> stylesheet, so perhaps an internet delay isn't the problem in saxon's
>> case. The reason why I think it might be an internet delay is that XEP
>> was exhibiting the same behavior (thought it produced a lot more
>> console text first) until I turned off draft.mode.
>>
>> Anyway, that's my update so far. I would appreciate any ideas you may
>> have on this issue. It's pretty hard to make good documentation
>> without any lists.
>>
>> On 9/6/06, Chris Chiasson <> wrote:
>> > I have a docbook 5 document where adding orderedlist or itemized list
>> > (as a child of chapter or para, so far) causes the following output:
>> >
>> > to-fo:
>> > [exec] Stripping NS from DocBook 5/NG document.
>> > [exec] Processing stripped document.
>> > [exec] Making portrait pages on USletter paper (8.5inx11in)
>> > [exec] XPath error : Invalid context position
>> > [exec] xmlXPathCompiledEval: evaluation failed
>> > [exec] Result: 10
>> >
>> >
>> > Then sections of the result document are missing, etc. Everything
>> > works fine when I don't have a list. What might be the cause of
>> > something like that?
>> >
>> > --
>> > http://chris.chiasson.name/
>> >
>>
>>
>> --
>> http://chris.chiasson.name/
>>
>
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFE/+vtjv9P65BfOUMRAuFiAKCFAUFYQPU+YXEWqyLyLb8qd2FJnQCcDu30
e+R8SoGg81/bFKmJLnJ7uw8=
=4eME
-----END PGP SIGNATURE-----
