[jsword-devel] MapTable

[Joe Walker]

Hi,

> Earlier I had worked on the MapTable used to display Sword Book conf
> files. One of the problems is that some of the lines are too long and
> should wrap. I had wrapping at one point, but took it out because it was
>   ugly when the window was resized. It is not really a big deal but it
> bothers me. So I want to fix it.
>
> I have been pleased with how JTextPane handles tables and I am thinking
> that would be a good solution for displaying tabular data. Any thoughts
> on whether it would be a good, neutral or bad idea?

Thoughts:
- I think HTML has more of a future than Swing.
- A Swing table is more resizable and (with extra effort) sortable, etc.
- Swing would probably fit better with the look of the rest of the app
- HTML can display much more than just a table

On a quick scan BookMetaData.getProperties is only used by this display so
it is tempting to replace it with:
  BookData getInfo();
Which would return an OSIS doc, containing whatever, but defaulting to a
table displaying the properties of the book.

> Further with the conf files, I have presented the rows in the order that
> they are in the file, with the synthetic rows at the bottom. In trying
> to figure out what Bibles would represent a good data test set
> (basically a one-of-each encoding approach) I was scanning the conf's in
> the Book installer window and found the lack of a consistent ordering a
> little bit frustrating. Is there a value in having the order consistent
> with that of the file? (Earlier it was random.) I would like to change
> it to make it more consistent. My initial thought is to order it based
> upon importance to the application (what it uses and what it does not)
> and importance to the user (i.e. some reasonable approximation of the
> order of the other keys) I guess I could go overboard and let it be
> fully customizable, but I was thinking that a resource file with ordered
> keys would be a good starting place. Maybe represent the Map in xml and
> use xslt to present it.

See comments above.

My one word of caution would be that I'm not going to get too involved
because I want to focus on fixing more important bugs, and that I don't
want to remove the current method until this is better than it.

Joe.