I've been working on field enhancement for OpenOffice.org Writer for quite a while and today I finished my proof-of-concept hacking:
OpenOffice.org Writer has a lot of shortcommings wrt. to fields which I tried to address:
- Allow for in-place editing of input field (turn off pop-up)
- allow use of different fonts within one field
- Nested Fields support on Write and allow nested conditions when using conditional fields
- ...
In my proof-of-concept I was able to enhance the Writer core such that these issues are addressed. (That's the good news!)
Unfortionately my proof-of-concept still needs a lot of love. First thing is to clean up the prototype and generate patches for ooo-build.
However I'm happy since this is my first major work on the OpenOffice.org Writer layout and the field support is an issue in OpenOffice.org Writer for quite a while...
12 comments:
Wahou, that sounds nice. Thanks for working on that old issue!
Great news, indeed encouraging! Are there plans to write a spec about that?
Hi Pierre,
thanks for the feedback. So --- have you tried it and does it work for you?
Wrt. to the specification issue. At the moment I'm working full time on solving the issue in the Writer core. The issue is out there for years for some reason: It needs lots of changing in the core.
So I would need some help in writing a spec....
Oh -- and thanks Peter too.
At the moment my spec is: Make the hard core changes.
But I'm definetly willing to work together on a spec.
Florian: what would be the ODF behind this? The new text:meta-field?
In any case, I like what I see (can't test it since I'm on the Mac).
Hi Bruce,
as you know I had my use case stated in the ODF Metadata SC:
Use meta data to map "MS fields" to OpenDocument [http://wiki.oasis-open.org/office/metadata_use_cases]
Unfortunately they where not picked up: http://wiki.oasis-open.org/office/OpenDocumet_Metadata_Use_Cases_and_Requirements
So as you know. “text:meta-field” fields it not an option.
Additionally I addressed the problem in the main TC:
* enhance field support by introducing a <text:field-start/> and a <text:field-end/> element to which metadata can be attached. [http://www.oasis-open.org/archives/office/200611/msg00048.html]
Not loved by the TC [http://florianreuter.blogspot.com/2007/08/status-of-my-suggested-enhancements-for.html]. Did my best.
But luckily there are solutions which do not require a file format change…
~Florian
You're simplifying the story here. :-)
Most of what you disccuss in this post (nested fields, formatting within fields, etc.) is the intended use case for text:meta-field. I also am interested in being able to map between text:meta-field and Word fields. The only example in your screenshot that is not explicitly intended is the one where the display content spans multiple paragraphs.
So then we'll have two different ways to do (almost) the exact same thing?
Is there a way to structure the work you do here such that it ccould be easily extended to support the new field (e.g. so that when the new RDF framework is in place, it can tie into some generic field API)?
Well --- as you said --- the point is that the text:meta-field is not sufficient to encode everything.
My problem is that I have several sample documents which need this feature.
So I can't use it.
The sad thing is that I really tried to convince the ODF Metadata SC to make the text:meta-field more powerfull. Really no idea why they refused to add a <text:meta-field-start/> and a <text:meta-field-end/> similar to bookmarks.
So --- Yes. I'd love to use the meta fields. I even started to implement them.
However since I can't use them I need to find an alternative.
So --- oversimplified --- you voted against the start-/end-meta-fields which made them unusable for encoding Word fields.
Sorry for the above statement. But its the truth.
Wrt to the specification. I made a request the the UX project: http://ux.openoffice.org/servlets/ReadMsg?list=request&msgNo=4
I hope the UI people have some input here too.
Restated the problem at the ODF TC: http://lists.oasis-open.org/archives/office/200710/msg00066.html.
I don't think we ever voted on the idea, did we? I know I was uneasy about it, but I can only find reference to a hint of that position earlier on when I was favoring a different approach to encoding the field in XML. I think with the way we ended up the in the proposal, it seems to make sense to adopt your suggestion.
But in any case, I posted a followup to your's on the TC list ;-)
Post a Comment