[xsl] RE: [saxon] OASIS-Open/CALS *big* table conversion problems

Subject: [xsl] RE: [saxon] OASIS-Open/CALS *big* table conversion problems
From: Michael Müller-Hillebrand <info@xxxxxxxxxxxxx>
Date: Thu, 13 May 2004 16:49:22 +0200
It seems as the solution mainly by Andrew Welch which I posted in March 2004 with the subject "Table Normalization"


addresses some of your issues, especially retrieving the correct column widths from a table with colspan and rowspan attributes. Unfortunately the archive display is truncated, so I send you the posting in a separate mail.

- Michael

On 13.05.2004 (9:56 Uhr -0400), Mullee, John wrote:

>(Sorry, OT, Cross posting to XSL-LIST; let's resume there ...)
>Jakob's case doesn't /require/ recursion (AFAICT), but
>I had issues with the other people's code when processing
>rowspans and colspans.
>There was a thread about this in July 2001, 'table formatting challenge',
>where a few solutions were posted, but none of them handled the mixed
>rowspan/colspan stuff I had in ~ 100MB xml with < 10^4 cals tables.
>The recursive technique was the only one I could come up with,
>having to refer, for row/col span reasons, to computed output from
>previous context. Recursion allowed storage of intermediate results
>in an RTF, which was made available to subsequent calls.
>I don't know why NW's code didn't handle Jakob's 2000-row table -
>perhaps it's head-recursive and blew stack.
>I'm new to XSLT in particular and even functional languages
>in general, so NW's code was a bit opaque to me ...
>NW posted some corner cases I sucessfully tested my code against (see below)
>Does your code handle it okay? like to post ? :)

Michael M|ller-Hillebrand, Dipl.-Ing.    <http://cap-studio.de/>
    FrameMaker, FrameScript, XML/XSL,... Publishing-Lvsungen

Current Thread