Recent changes to feature-requests

Support non-UTF-8 encodings (e.g., UTF-16LE)

David L. Chandler — Tue, 26 Jul 2005 00:12:04 -0000

The converters are UTF-8 centric. Not everybody groks
iconv, though, so make it friendly.

TMW->Unicode should handle font size wisely

David L. Chandler — Sun, 27 Feb 2005 00:36:26 -0000

When you do a TMW to Unicode(RTF) conversion, the font
sizes are preserved. This is better than dropping all
info, but better yet is to have a smart mapping between
TMW font sizes and Tibetan Machine Uni's font sizes,
which are not the same.

Make ICU support Tibetan, specifically TMUni

David L. Chandler — Sat, 26 Feb 2005 23:07:00 -0000

ICU (icu.sourceforge.net) may already support shaping
Tibetan Unicode, but we should make it do so if it
doesn't. (It's X licensed so everyone can change ICU
and everyone can benefit from the changes (though they
need not contribute their own changes back).)

Test by making Jskad support TMUni. Package TMUni with
Jskad and make using it, and storing things as Unicode,
the default.

Gray out the copy/cut/paste functions when appropriate

David L. Chandler — Sat, 26 Feb 2005 21:50:25 -0000

If the clipboard is empty, you can't paste, now can
you? If the selection is empty, you can't copy or cut,
can you? if the duffpane is not editable, you can't
cut or paste, can you?

Gray out the menu items when appropriate.

WW: optimize out font changes

David Chapman — Fri, 25 Feb 2005 02:20:24 -0000

Experiments with Cathy Cantwell's font-fixer show that
performance dramatically improves if font changes are
done on maximal contiguous ranges, rather than
character-at-a-time.

I'm not sure what fraction of WW's overhead is due to
this effect, but it might be substantial.

WW: all Tibetan menu items should have accellerators

David Chapman — Thu, 27 Jan 2005 12:18:08 -0000

It's a dumb oversight that they don't.

WW: take non-T font from nearest?

David Chapman — Wed, 19 Jan 2005 01:10:04 -0000

Plausibly it would be better if when you did a ] or
ctrl-T, the font (and size and other text properties)
would be taken not from the Options dialog (as
currently) but rather from the textually most recent
bit of non-T text. I.e. should scan back from the
insertion cursor to find the first non-T character and
format the next non-T text that way

WW: Tibetan spell-checking?

David Chapman — Wed, 19 Jan 2005 00:55:26 -0000

Since Tibetan is especially prone to spelling errors,
it would be really good if there were a spell-check
dictionary for it in Word. I have no idea what is
involved in creating one -- it would probably be
trivial if the API were documented, and a major pain if
not.

Maybe Microsoft is going to do this?

WW: interlineal: horizontal whitespace option

David Chapman — Wed, 19 Jan 2005 00:54:04 -0000

Cathy Cantwell said one of the reasons she didn't like
interlineal mode was that it puts in too much
horizontal whitespace. There should be an option to
select how much.

WW: preserve footnotes in conversion

David Chapman — Fri, 24 Sep 2004 19:39:20 -0000

Cathy Cantwell
<catherine.cantwell@oriental-institute.oxford.ac.uk>
writes:

"As you may know, Rob and I are working on rNying ma'i
rgyud 'bum critical editions on one of our Oxford
projects. We would like to be able to make Tibetan
versions of them, but WylieWord as it is at present
can't manage to convert even tiny files since it can't
deal with footnotes! Obviously, with critical
editions, one has footnotes on every other syllable.
Is there any chance of improving this?"

This would require changing the iteration over the
selection string in ConvertWylie to iterate over some
sort of object other than characters. I'm not sure
what the underlying data type is, but presumably you
can iterate over some sort of hairier object and
recognize footnote objects as such.

I'm not sure if she wants conversion to occur in the
footnotes, or to leave them alone, or what.