<div dir="ltr">2015-04-27 13:27 GMT+02:00 Philip Taylor <span dir="ltr"><<a href="mailto:P.Taylor@rhul.ac.uk" target="_blank">P.Taylor@rhul.ac.uk</a>></span>:<br><div class="gmail_extra"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class=""><br>
<br>
Zdenek Wagner wrote:<br>
<br>
> if you use <para indentation="none" looseness="1">Something...</para><br>
> you are not using XML properly. The purpose of XML is to describe the<br>
> structure of a document, not its appearance.<br>
<br>
</span>Yes, the element describes the structure; attributes such as those<br>
you quote in the example above convey hints about its intended<br>
appearance. There is no conflict -- software wishing to ascertain the<br>
structure interrogates the elements; software wishing to depict the<br>
structure visually can make use of the formatting attributes if it so<br>
chooses.<br></blockquote><div><br></div><div>Formatting advices should better be present in processing instructions. You can type <?TeX something?> <?HTML something else ?> etc. <br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class=""><br>
> Generally you will print the information from the XML file in a<br>
</span>> different order ...<br>
<br>
Indeed we do; notes are taken out of the flow and re-set as end-notes.<br>
This behaviour is unaffected by the use of formatting attributes.<br>
<span class=""><br>
> As the second step you will run XSLT in order to extract the elements<br>
> that have to be printed and finally XSL-FO in order to format the<br>
> output.<br>
<br>
</span>That is /a/ methodology; there is nothing written in tablets of stone<br>
that says that one is required to follow it. The XML sources to which I<br>
refer /could/ be processed using such a methodology, because the XML is<br>
well-formed; in practice, it is far far simpler to implement an XML<br>
parser and formatter using XeTeX, which is what I have elected to do.<br></blockquote><div><br></div><div>Such parsers already exist in luatex and ConTeXt. I only know about them, I have never used them. And FO may be processed by passiveTeX. And there are, of course, numerous other tools.<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
** Phil.<br>
</blockquote></div><br></div></div>