Re: [platform-text-dev] Support for record-oriented document styles

[David Dykstal <david_dykstal@xxxxxxxxxx>]

Kai and the team --

We do, in fact, already have a separate plugin.  The component itself is
mature.  But I think you're talking about the "maturity" of any derived or
new extensions we would want to provide, right?

What we'd like to do is redistribute the function in this component so that
the eclipse editor framework can take advantage of it and offer it to
others.  We believe such function would be generally useful -- block
selections and column specific operations, for example, have shown up as
requests in bugzilla.  Eclipse would gain a more complete framework for
editing these kinds of files.  We would gain a greater integration into
eclipse.  What we don't want to do is place function into the framework
that would be useful to us alone, but enable us and others to exploit
functions that are unique to our components.  We need your help in striking
the balance.  I suggest that we begin by outlining the proposals for each
function we think we need in bugzilla and carry on the discussions there
with pointers to the major items and revisions from the mailing list.

Our needs include some rendering changes (eg. block selection, selection
beyond text borders), and interaction behavior (eg. cursor skipping over
"control" characters) that are currently part of the StyledText widget.  Do
we talk about that stuff here or in another mailing list?

I know that everyone (us too!) is really busy with the upcoming 3.0
release, but I want to make sure that these design discussions don't begin
too late and miss the window of opportunity for anything that could make it
into the release after that.  So we'll probably keep things at a fairly low
(but still perceptible) level of activity until everyone has more time to
look.

Our team here understand the critical nature of binary compatibility and
stable APIs very well.  Anything we can do to delay this, but still make
progress, is a good thing.  So the separate plugin idea might make a lot of
sense, but it would have to be one that integrates into the current scheme
in some fashion -- allowing us to exploit IDocument and its extensions.
Without that, we have no incentive to do this work.

We would be more than happy to show the platform-text folks the kinds of
function we are looking for and how we addressed it.  What's a good way to
do that?  Screenshots?  Movies?  Personal visit to Zurich <grin>?

As for the PMC goals -- I believe those are yet to be established for
releases after 3.0.  I will see what I can do to take the case for
something like an "enhanced server affinity" item to them.
_______________________
David Dykstal
david_dykstal@xxxxxxxxxx