To comment on the following update, log in, then open the issue:
http://www.openoffice.org/issues/show_bug.cgi?id=60594





------- Additional comments from [EMAIL PROTECTED] Tue Apr  8 07:37:48 +0000 
2008 -------
@hdu: some additional thoughts

I made some more experiments. I spare you the patches until they will be more
useful or you explicitly request to get them. But I think I have some important
hints to find a reasonable approach.

- It seems that - no matter if writer will respect a minimal kashida width in
its justification or not - the approach taken in my previous patch remains
valid. Probably due to rounding errors from logical units to device units and
due to a tendency of ScriptApplyLocigalWidths to be just off by one pixel in its
results, I still observed the gaps breaking the words. Less frequently though.

- There is another problem with vowel placement on a stretched letter. This also
seems to require some extra manipulation of the justified glyph widths. (not
implemented in the previous patch yet)

- I saw, that there is a GenericSalLayout::KashidaJustify method, which also
respects a minimal kashida width. Does this work? Or does this require
debugging? This issue was reported on all platforms! Anyhow, the minimal kashida
width used over there seems to be much smaller than the one returned by usp.

- When I moved the code to correct the character widths to writer, I discovered
that it's pretty useless to pad the blanks as I did in winlayout.cxx. As said
before, the rounding errors force me to do it again over there anyhow, and so
there is not much to gain. After looking at ScriptJustify once more, I switched
to reduce the number of kashida insertion points in the line, and recalculated
the space to add for each kashida. With this, the justification looks really
clean now. The fixing in winlayout.cxx has not a lot to do, except for some rare
cases (e.g. very short lines). But I encountered one problem: The cursor
placement is affected, and I have a "chicken and the egg" problem there. Cursor
placement relies on the correct number and identification of kashida insertion
points. But I manipulated there, so the cursor placement is off in some cases.
So there is some work to be done there, but I left that for the moment... 



---------------------------------------------------------------------
Please do not reply to this automatically generated notification from
Issue Tracker. Please log onto the website and enter your comments.
http://qa.openoffice.org/issue_handling/project_issues.html#notification

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to