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 "Peter B. West" <pb...@powerup.com.au> on 2003/02/28 02:16:09 UTC

alt.design TODO

Fopdevs,

Sorry about the delay in getting back to this.  I plan to tag HEAD and 
create a branch for alt.design integration.  There are a number of 
things that need to be done.

* Write or transmogrify the corresponding properties code.
The point at which this must occur is well-defined, I think, but I 
haven't got around to doing any of it.  I have no idea how much or 
little of the existing corresponding properties code will tranlate over.

* Complete the recognition of fo:marker elements on the FOs.  At the 
moment, only the parsing and storing of the fo:markers within the fo 
objects is being done, and that for only some of the FOs.  I just wanted 
to establish the method.  My understanding was always that I would have 
to copy the selected marker subtree, with its unevaluated expressions, 
into place under the fo:retrieve-marker, and process the subtree through 
the parser.  This latter part is only required as the area tree is being 
built.  There are a few questions about the appropriate expression 
environment for this operation, which I have asked of the editors. 
Yeah, ok.

* Replace the existing property handling regime.
This will involve the creation of any necessary additional interfaces in 
to the FO tree.  Because I am unfamiliar with the existing code base, 
this may take a while, but the compiler is going to be a big help here.

* Provide for caching of FO tree subtrees.
This process will be triggered by feedback from the layout engine about 
FO subtrees.  The possibilities are discard or cache.  I always 
envisaged this, but the scope has recently expanded.  See below.

* Modify the interaction between the FO tree builder and the layout engine.
This is a big and very interesting issue.  As I was considering the 
integration, I realised that FO tree expression parsing cannot be done 
sensibly without feedback from the area tree.  I have also asked the 
editors for clarification on this.  In order to implement it, I will 
need to vary my original idea of how the interaction between these 
components will work.  For the mud-map, see the attachment.

As I get these ideas a bit clearer, I will set up the wiki page(s) and 
get going with the integration.
-- 
Peter B. West  pbwest@powerup.com.au  http://www.powerup.com.au/~pbwest/
"Lord, to whom shall we go?"