Re: [xsl] XSL-FO and XSD Information request

Subject: Re: [xsl] XSL-FO and XSD Information request
From: Kamal Bhatt <kbhatt@xxxxxxxxx>
Date: Mon, 20 Mar 2006 18:47:00 +1100
Wendell Piez wrote:

Much of Mulberry's schema documentation is built this way, however, and we're not the only ones. The thing is, that's part of the special value we provide our clients, and not something we're able to develop and support for the world at large. So in general we (and many other schema designers will say the same, I think) can affirm that yes, it's doable, and in principle a good idea (assuming you know what you're doing); but we can't say much more beyond that than ask the classic question: "what are your requirements?". Whether it's hard or not depends on the task. It's not hard to do lots of useful stuff; but particular tasks can be quite challenging, depending on what kind of schema you have. Perhaps the biggest conceptual challenge is in developing a good sense of which tasks can be automated easily and usefully, and which ones are better left for sensible people to do.

OK, I was not too good at specifying what I wanted to do. I guess there are two difficult hurdles that I will need to jump. The first is having data is multiple files. I thought that maybe I could get away with something simple (such as looping through all the included xsd files) but then it occured to me that I can find (for example) a attribute group, but this attribute group may have a reference to a simple type which means either searching the file that has the attribute group or searching the imports of the file. This is doable I think by passing the document that called the template to the next template (if you follow). Unfortunately, I think that is a particularly bad idea from cocoon. That is probably a question for the cocoon mailing list.


Second problem is pretty printing XML examples. I was going to use some annotations to store example data and use the annotations to generate the date. Now, the file searching problem aside, we also have the problem of printing this nicely, and I am not talking about simply identing. Most people don't likes reading XML that looks like this:

<element attr1="xxxx" attr2="xxxx" attr3="xxxx" attr4="xxxx" attr5="xxxx" attr6="xxxx" attr7="xxxx" attr8="xxxx" attr9="xxxxx" attr10="xxxxxxxx" attr11="xxxxx" attr12="xxxxxxx" attr13="xxxxxx" attr14="xxxxx">
...
</element>
most people would prefer:


<element attr1="xxxx" attr2="xxxx" attr3="xxxx" attr4="xxxx" attr5="xxxx"
              attr6="xxxx" attr7="xxxx" attr8="xxxx" attr9="xxxxx"
              attr10="xxxxxxxx" attr11="xxxxx" attr12="xxxxxxx"
              attr13="xxxxxx" attr14="xxxxx">
...
</element>

Any ideas how to achieve this? I have none.


I've created a system that turns content stored in XML into PDF (through
XSLT to XSL-FO and then through FOP), Word (through XSLT to WordML), HTML
(throught XSLT), JavaHelp (through XSLT), and text (through XSLT).


So, what you have in mind can be done.

I was thinking that maybe I could use something to convert the XML into PDF, word and HTML inside of cocoon. I think there is support for FOP in cocoon, what about WordML?


Cheers.


Kamal.

--
Disclaimer: This email is confidential and may contain privileged information for the sole use of the person or business to which it is addressed. If you are not the intended recipient, please notify the sender by return e-mail or phone as you must not view, disseminate, distribute or copy this email without our consent. We do not accept any liability in connection with any computer virus, data corruption, incompleteness, or unauthorised amendment of this email. It is the sole responsibility of the receiver to scan for viruses before opening.

Current Thread