On 11/23/2010 22:48, Andrew Pitonyak wrote:
On Tue, 23 Nov 2010 22:42:07 -0500, TJ Frazier<[email protected]>
wrote:
Short take: Banding works fine, borders poorly.
...
*Mysterious border problems*
1) Some tables work fine. Some, only the rows after a page break work.
Some, only the last row works. Some show no borders at all.
Investigation of the rows that work shows the cell properties have been
set to the table-border properties (how? I didn't do that!). This may be
somehow related to the AutoFormat facility, which does this kind of
thing. I have not yet investigated the object "User defined properties"
in the tables. So far, I am unable to account for any of this behavior.
*Testing*
I have been testing with the published v3.2 "0219WG3 - Keyboard
shortcuts" which has lots of tables, but frequently loads badly. The
Math Guide, 0800MG3, is a lot easier to work with. The "Attributes"
table shows the no-border problem; all other tables work fine. It also
has tables without headers, and a table that page-breaks on an odd row.
For anyone who wants the code, I will email it privately; it's only
about a 4 KB text file, of type ".bas", and loads with the "Insert Basic
Source" icon in the IDE.
I wrote a macro some time back that does this sort of thing. For all I
know, your macro may have started from mine. I do believe that I set the
background using a hard coded value rather than relying on a style (and
that was changed in your latest version if I remember correctly).
The code I'm working with came from Michele Zarri; where he got it, I
have no idea, but it shows strong traces of Iannz's coding style from
the Cookbook on the wiki; I have maintained that.
I occasionally see issues with borders. I never did figure out precisely
why, but, I think that I am more likely to see the problem if the table
crosses a table boundary than if it does not. Usually, running the macro a
second time fixes the issue. Because I can fix it by a second execution, I
never tried to figure out the precise problem.
Thanks for the hint. The problem is definitely related to crossing page
boundaries. I'm waiting until 3.3 escapes (pardon me, "is released") to
download the SDK and the MRI debugging tool. There is definitely a bug
in Writer, related to repeating headers.
My experience with repeating macro execution is that I always get the
same result the second time, too. If I run the macro on a short table
that is not split across pages, it displays borders. If I shove that
table down so that it now splits, the borders are still displayed—until
I run the macro again, and the borders disappear from the first page.
This is the same result I get if I force the table to split before
running the macro the first time. Running the macro definitely
aggravates something.
Andrew Pitonyak
--
/tj/
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
