Lists Home |
Date Index |
I am not really at the FAQ level on this. I am an accountant with advanced technical
skills but far from a professional programmer type. I can say that I have studied and
understand most of the available XML technology and information theories behind it. Having
said that, the advice you provide is perfectly correct if I am interested in writing XSLT.
I would most admittedly need some help. :) However, your stylesheet example was really a
diversion from the central message ...the stylesheet was assumed in my "idea"(not that I
could easily write it).
Actually, I am saying that the stylesheet is so self-evident that we SHOULD not even need
it. We SHOULD (without anything application specific) be able to say to users: "Here is
what your files look like in this here <g><e/></g> thing... and that is ok to use as an
alternative with this or any other XML file you need to create.".
>>XML has (almost) no fixed names at all<<
Yes, I am simply reflecting on the idea that XML could have had something like <g><e/></g>
at the level of reserved words in a programming language. It doesn't go against the intent
of XML because it is not a vocabulary, it is a general condition of all flat xml in any
vocabulary. It is like where I can transpose a range in a spreadsheet, but xml has no
official syntax to "flip" between flat to hierarchy ... and back.
After trying this on some existing files, I realized that, from now on, I will seldom want
to directly use native flat formats, and will front them all, all such vocabularies, with
"<g><e/></g>" internally at some point. That "aha" is the only reason I posted this idea.
I will also be writing some scripts for this, but we can agree that XSLT fills this purpose
... and no doubt the particulars of XSLT is better taken up on a different forum.
Thanks much. I will check out Jeni's site ... have been there before and IIRC the same
grouping discussion is in her book.
David Carlisle wrote:
> > Going the other way might require reordering the groups of records to place all with
> > identical attributes together. Or the stylesheet could explicitly state the matching
> > attributes that cause a <g><e/></g> block and that would be pretty easy. What is
> > interesting is that users would do this grouping themselves ... they would do it as a
> > matter of convenience when creating the <g><e/></g> blocks.
> > The ultimate improvement of your example is an identity transform copying everything
> > verbatim except this expansion of the <g><e/></g>. That is the stylesheet that would
> > substantially implement what I am proposing for files preprocessed against the
> > stylesheet. How easy is that? :)
> Both of those (which would each only require a line or two of xslt) are
> day to day faq questions on xsl-list Perhaps you should move this there,
> or see the usual reference for grouping on Jeni's site
> Grouping and flattening of hierarchical data is just what XML is all
> about, so I'm not sure I understand this thread really. I think you are
> proposing that the element names be hard coded into applications but
> that would be going against the whole point of XML which is a meta
> language for defining vocabularies. Specfic vocabularies such as xhtml
> have fixed element names but XML has (almost) no fixed names at all.
> This e-mail has been scanned for all viruses by Star. The
> service is powered by MessageLabs. For more information on a proactive
> anti-virus service working around the clock, around the globe, visit:
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
> The list archives are at http://lists.xml.org/archives/xml-dev/
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://www.oasis-open.org/mlmanage/index.php>