W3C home > Mailing lists > Public > www-xsl-fo@w3.org > July 2001


From: Arved Sandstrom <asandstrom@accesscable.net>
Date: Wed, 18 Jul 2001 20:39:51 -0300
Message-ID: <004d01c10fe2$f15cc960$721a8a18@accesscable.net>
To: <www-xsl-fo@w3.org>
I posted a long diatribe on this subject last night, from the wrong address,
so Max Froumentin gets it as list moderator. For all I know he'll
accidentally turf it, or he's away, so I'll distill my comments here and try

Maybe it's just me, but I've been dissecting the language of the spec as it
pertains to markers (fo:marker and fo:retrieve-marker), and I'm unhappy. I
understand the "intent" of the associated properties and I think I
understand how markers are intended to work, and if I'm willing to suspend
strict interpretation of what the spec actually says, I could proceed
further with FOP implementation, which is what I'm doing right now. But I
happen to be very rigid when it comes to specs, and so I'm stalled out.

Problem: there is a circularity in the discussion pertaining to
fo:retrieve-marker. According to this language - "The term "containing page"
is used here to mean the page that contains the first area generated or
returned by the children of the retrieved fo:marker." Also, "The fo:marker
whose children are retrieved is the one that is (conceptually) attached to
the area that is at the top of this hierarchy."

In other words, considering these statements in isolation, the "containing
page" is determined _once_ we have selected the "best" area in the hierarchy
(the set of qualifying areas) - that "best" area fixes the f:marker that is
to be used, and that fixes the "containing page".

My question: so how is it that when we are using properties on
fo:retrieve-marker like "retrieve-boundary" and "retrieve-position" that we
can refer to the "containing page"? The whole point of that exercise is to
find the "best" area - we don't necessarily even know the page yet. How can
we use a datum ("containing page") to help determine, ultimately, itself?
Answer: with great difficulty. In order to figure out what the "containing
page" is, we need to know the "containing page". That just won't cut it.

What's obviously implied in the whole line of discussion is the notion of
"current page", not just "containing page". The current page (the page with
the active fo:static-content and fo:retrieve-marker) is clearly significant,
but it doesn't get mentioned in this section of the spec at all, not once.

I'm looking for some clarification of the language, pretty much to confirm
what are already my working but unfounded assumptions, so I can proceed with
more confidence. It's not like I'm lost as to what the spec is trying to
achieve; I just wish the langauge was more solid and not, well, wrong. :-)
Thanks much.

Arved Sandstrom
Received on Wednesday, 18 July 2001 19:42:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:58:25 UTC