CLIM mail archive
[Prev][Next][Index][Thread]

Re: question on table-formatting + graphics

--> Date: Fri, 17 Jun 94 09:44:25 -0700
--> From: Marilyn Bunzo <marilynb@seahorse.arc.nasa.gov>
--> Subject: question on table-formatting + graphics
--> 
--> CLIM 1.1 ACL 4.1
--> I inherited an editor from a previous programmer.  The code was developed on
--> a Symbolics with CLIM 1.1 and showed the same problem that I have on ACL.
--> We have a workaround but I would prefer to find the solution or rewrite if
--> necessary.  
--> Our editor has two columns of information which the user can link together
--> by selecting an item in column one and one in column two.  A line is drawn
--> between the two items to show the link explicitly.  Unfortunately, the
--> _display_ of the lines interferes with the items in the column.  When you
--> run your mouse down the column, the lower items in the column will no longer
--> be selectable once the links are displayed.  We added a flag to control the
--> display of the links.  When the links are present but no drawn to the pane,
--> all items in the columns can be selected.  When the same links are drawn on
--> the pane (using draw-line*), the lower items in the list will once again
--> become inaccessable.  
--> We are using table formatting to build the columns.  I assume that we are
--> somehow interfering with the table by overlaying graphics.  Can anyone give
--> me any further information here?  I have included code for a simple test
--> case based on our editor below.  

we had exactly this same problem with CLIM 1.1 ACL 4.1. the simple
solution is just to NOT use formatting-table. I don't recall whether we
ever fixed the flaw or not, but it was just easier to step around it.
formatting-table is not very efficient anyway, doesn't do incremental
updating, etc., all things we found not tolerable.

i think we found format-graph-from-root to have a similar problem (my
guess is that it's exactly the same one down deep).

--> We are about to port our code to CLIM 2.0.  Is this problem likely to
--> persist in CLIM 2.0?  If this problem will go away in CLIM 2.0, I won't
--> worry about doing a rewrite now.  Our changeover to CLIM 2.0 isn't simple; 
--> we have to upgrade our operating system which also means that ACL 4.1 will
--> no longer work on the upgraded machines.  We also have major demos for our
--> project this month just to keep life interesting.
--> Thanks in advance for you help.
--> 
--> Marilyn

just draw the stuff yourself. it's easy enough to grab the bounding-box
size of the presentations, max the delta-x for column width, and max the
delta-y for row-heights, and output-record-set-position the individual
items wherever you want.

 -- clint
Main Index | Thread Index