<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<META NAME="Generator" CONTENT="MS Exchange Server version 6.5.7652.24">
<TITLE>RE: [postgis-devel] Do you think we should put a note about 2D and3Don each function?</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->
<BR>
<P><FONT SIZE=2>I gave this a tad bit more thought. Maybe we should just repurpose existing tags. E.g. is there a name property we can flagg all mm, 3d things with?<BR>
<BR>
I guess what I was thinking was 1 xsl for each that would generate a doc book xml index (e.g. listing all mm, 3d functions, agg functions) which would then be processed by regular docbook xsl. That way we just need one set for all output types.<BR>
<BR>
<BR>
-----Original Message-----<BR>
From: postgis-devel-bounces@postgis.refractions.net on behalf of Kevin Neufeld<BR>
Sent: Fri 9/19/2008 1:40 PM<BR>
To: PostGIS Development Discussion<BR>
Subject: Re: [postgis-devel] Do you think we should put a note about 2D and3Don each function?<BR>
<BR>
I agree. Having our own extra xsl would cleanup the sqlmm and ogc<BR>
compliant tagging. Give 'er a try.<BR>
<BR>
How would this work with the different conversions we do (html, chunked<BR>
html, and pdf all use different docbook stylesheets)? Would we have to<BR>
create three different xsl wrappers for html/docbook.xsl,<BR>
html/chunk.xsl, and fo/docbook.xsl?<BR>
-- Kevin<BR>
<BR>
<BR>
Mark Cave-Ayland wrote:<BR>
> Obe, Regina wrote:<BR>
>> Is there an easy way to define a new tag or something for this. I was<BR>
>> thinking if we had a separate tag for it, it would be easier to pull out<BR>
>> those functions that have that tag later.<BR>
>><BR>
>> I guess we have similar issues with MM. Anyrate we can still do it if<BR>
>> we stick to the same terminology, but I figured it would be less prone<BR>
>> to error if we had a tag.<BR>
>><BR>
>> I am feeling a bit more confident about my XSLT abilities now that I<BR>
>> have a couple of unrelated projects that involved me having to write xsl<BR>
>> files to transform the data into another format. I was hoping to tackle<BR>
>> generating the function alter comments statements in the next week or so<BR>
>> trying to stick with just using a custom .xsl file for minimum increase<BR>
>> in dependencies.<BR>
>><BR>
>> Which brings me to the topic of if we create a custom xsl file for this,<BR>
>> where should we put this in the source tree or is this direction<BR>
>> completely wrong?<BR>
>><BR>
>> Thanks,<BR>
>> Regina<BR>
><BR>
> I'd say post a proof-of-concept to the list. As long as the paths within<BR>
> the docbook source are all relative, I see no reason why we can't<BR>
> include an extra .xsl file if required.<BR>
><BR>
><BR>
> ATB,<BR>
><BR>
> Mark.<BR>
><BR>
_______________________________________________<BR>
postgis-devel mailing list<BR>
postgis-devel@postgis.refractions.net<BR>
<A HREF="http://postgis.refractions.net/mailman/listinfo/postgis-devel">http://postgis.refractions.net/mailman/listinfo/postgis-devel</A><BR>
<BR>
</FONT>
</P>
</BODY>
</HTML>
<HTML><BODY><P><hr size=1></P>
<P><STRONG>
The substance of this message, including any attachments, may be confidential, legally privileged and/or exempt from disclosure pursuant to Massachusetts law. It is intended solely for the addressee. If you received this in error, please contact the sender and delete the material from any computer.
</STRONG></P></BODY></HTML>
<P><hr size=1></P>
<P><STRONG><font size="2" color="339900"> Help make the earth a greener place. If at all possible resist printing this email and join us in saving paper. </p> <p> </font></STRONG></P>