You are viewing a plain text version of this content. The canonical link for it is here.
Posted to fop-dev@xmlgraphics.apache.org by Manuel Mall <mm...@arcus.com.au> on 2007/06/30 03:38:08 UTC
Re: [Xmlgraphics-fop Wiki] Update of "FontSelectionStrategy" by AndreasDelmelle
On Saturday 30 June 2007 03:00, Apache Wiki wrote:
> + === Alternative Implementation ===
> +
> + ==== Rationale ====
> +
> + Given the above noted dependencies, there seems to be a third
> alternative: create multiple FOText instances in the FOTree, which
> would lead to two separate TextLMs being instantiated as a part of
> FOP's "natural" flow. The different, separate TextLMs will, by
> design, trigger the correct relative alignment, and maybe solve other
> of the above mentioned complications of implemententing it in the
> layout-engine. +
Andreas,
have you considered the case of fo:marker elements where the fonts
available for selection may not be known at fo tree construction time?
IIRC we have been 'bitten' before by doing things during the fo tree
construction phase only to find out later that it is broken when it
comes to markers. May be its not an issue here as I haven't thought
this through in detail.
Cheers
Manuel
Re: [Xmlgraphics-fop Wiki] Update of "FontSelectionStrategy" by AndreasDelmelle
Posted by Andreas L Delmelle <a_...@pandora.be>.
On Jun 30, 2007, at 03:38, Manuel Mall wrote:
Hi Manuel,
> have you considered the case of fo:marker elements where the fonts
> available for selection may not be known at fo tree construction time?
Good that you point it out now, since I indeed did not explicitly
consider markers yet.
OTOH, I don't immediately see it as a blocker, since a great deal of
what is done for FOs during the tree construction, is in fact
deferred for markers until they are retrieved, exactly because of the
fact that inherited property values are not known, and markers could
be retrieved multiple times in different contexts.
For marker-descendants, this extra treatment of the text/characters
can be folded into the marker-cloning instead of performing it on the
original FOs. Just like property resolution and whitespace-handling
are skipped for the marker-descendants when they are performed for
all other FOs.
Cheers,
Andreas