0 byte index.html on chunked docbook output

[loge]

Hi,

i have a weird problem over here. I have the xml below to generate my documentation. If i am creating PDFs, everything is fine, my Xincludes seem to be resolved and voila.

If i am transforming to chunked HTML with the default chunk.xsl supplied with docbook. Everything is created correctly, but not my main index.html, this main file is 0 bytes. All the other chapter HTMLs are correct.

I tried several XSLT tranformators in the Oxygen Settings without any difference.

Code: Select all

<?xml version="1.0" encoding="UTF-8"?>

<!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.3//EN"

                      "http://www.docbook.org/xml/4.3/docbookx.dtd" [

                      <!ENTITY % xinclude SYSTEM "file:/C:/Programme/Oxygen%205.1/frameworks/docbook/dtd/xinclude.mod">

                      <!ENTITY preface        "preface.xml">

                      <!ENTITY intro              "chap_intro.xml">

                      <!ENTITY installation  "chap_installation.xml">

                      %xinclude;

                      ]>

<book lang="de">

    <title>VersysNG - Benutzerhandbuch</title>

    <subtitle>Versandsystem </subtitle>

    <bookinfo>

        <author>

            <affiliation>

                <orgname>Logentis e.K.</orgname>

            </affiliation>

        </author>

        <edition>0.1</edition>

        <copyright>

            <year>2005</year>

            <holder>Logentis e.K.</holder>

        </copyright>

    </bookinfo>

    

    <xi:include href="&preface;" xmlns:xi="http://www.w3.org/2001/XInclude"/>

    <xi:include href="&intro;" xmlns:xi="http://www.w3.org/2001/XInclude"/>

    <xi:include href="&installation;" xmlns:xi="http://www.w3.org/2001/XInclude"/>

</book>

Thanks for any info on this.

[loge]

I ve done some more research on this one. The index.html is created ok, but one second before the whole tranformation process ends, it gets overwriten with 0 bytes. I dont know whats causing this, but its weird.

[sorin_ristache]

Hello,

Sorry it works for me. The index.html result file is created correctly from your XML document and it is not truncated to 0 bytes. The XInclude parts are included and linked correctly. The same happens if I set other name for the main HTML file of the chunked result of the transformations - using the XSLT stylesheet parameter root.filename set in the dialog Edit transformation scenario -> Configure parameters. Are you sure you use [oXygen-install-dir]/frameworks/docbook/xsl/html/chunk.xsl ? Do you get an error message in the message panel at the bottom of the <oXygen/> window ?

I hope this helps,
Sorin

[loge]

ups, no i am using the original stylesheets coming with docbook 4.4, i will try use the original ones. Thanks for the hint.

I will report if this was the error.

marc

[loge]

No, doesnt make a different, can you tell me what transformator you have used and what output filename (i am using: ${cfd}\${cfn}.html). THis is really weird. Sometimes and i dont know when and why, my results were placed on the desktop (thats also strange) and then index.html is correct. I know this sounds like a fairy tale.

How can i see what the command line is when executing the transformation? I want to check the docs for the specified XSLT parsers and do it manually to see whats wrong.

marc

[loge]

Ok, i didnt have used "root.filename" before. I have set this to "start" and now i can create the main page. Of course i still have the 0 bytes index.html and also cant define "index" as root.filename because then i have the same problem as before. The solution is not really a solution, but a good workaround. If i will have time, i will investigate into that further.

But i really would like to know how i can see what Oxygen is doing in terms of commandline call in the background when doing transformation. This info would help me debugging this problem.

THanks

Marc

[sorin_ristache]

Hello,

I finally figured out what is wrong. The DocBook chunk XSL names the main HTML file of the result document index.html and it does not take into account the value that you set in the field Save As in the Edit scenario dialog, Output tab. This is the way the DocBook chunk XSL works. You can configure the output file name for the chunk XSL by using the param root.filename as I explained in my previous post. In this particular case (DocBook chunk stylesheet) the name that you set in the Save As field of the dialog will be always the name of a 0 length file. In your case probably your source document is named index.xml so the name of the result file after expanding the macros ${cfd} and ${cfn} that you used in the field Save As will be index.html in the same directory as index.xml. This index.html will be the empty file that overwrites the good file with the same name created by the chunk XSL. The solution is to rename your XML source document or use the root.filename parameter and the output of the stylesheet should not be truncated to 0 bytes any more.

Best regards,
Sorin

[sorin_ristache]

Hello again,

<oXygen/>'s Debugger perspective and the error and warning messages displayed in the Results panel at the bottom of the <oXygen/> window should be enough for debugging a transformation/the XSLT stylesheet. Not in your case because this is the way that the DocBook chunk XSLT stylesheet works but this is a known problem of this styleshet and questions related to it were already answered on this forum: http://www.oxygenxml.com/forum/viewtopic.php?p=910 You can't see something like a command line call when the transformation is running because there is no such call - what goes on during a transformation is more than simply running a command line containing the name of a XSLT transformer.

I hope this helps,
Sorin

[loge]

YES!

Thanks for the link. I left out the "save as" and added another XSL paramter "useIdAsFilename" and modified the paramter "root.filename" with value "index". Perhaps it would have been enough to just leave out the "save as" filename, i dont know.

To me it seems the XSL chunked stylesheet allready generates the index.html and oxygen does something afterwards when you dont leave "save as" empty.

This was tricky.

Marc