On Mon, Sep 03, 2001 at 11:58:01PM +0200, Karsten Jensen wrote:
>
> One way is to do something like this for the headerfiles:
>
> AC_CONFIG_FILES([include/Motif-1.2/Xm/Xm.h.tmp:include/Motif-1.2/Xm/Xm.h.in],
> [if cmp -s include/Motif-1.2/Xm/Xm.h.tmp
> include/Motif-1.2/Xm/Xm.h
Just realized that our source distributions are a too big,
they carry redundant data.
Biggest contribution is probably from the headers. For whatever reasons
we currently seem to link and distribute old headers
(1.2, 2.0) in each subdirectory {1.2, 2.0, 2.1}.
Even the links are useless, since we a
If we get rid of the links we'll probably have to resort to some sort of magic
during the install phase, since a 2.x install still needs the 1.2 headers, or
at least most of them.
On 4-Sep-01 at 08:38, Alexander Mai ([EMAIL PROTECTED]) wrote:
> Just realized that our source distributions are a t
On Tue, Sep 04, 2001 at 08:42:06AM -0400, Rick Scott wrote:
> If we get rid of the links we'll probably have to resort to some sort of magic
> during the install phase, since a 2.x install still needs the 1.2 headers, or
> at least most of them.
Yes/no.
The approach is that we install from each s
Sounds fine. I consider this magic :)
On 4-Sep-01 at 08:53, Alexander Mai ([EMAIL PROTECTED]) wrote:
> On Tue, Sep 04, 2001 at 08:42:06AM -0400, Rick Scott wrote:
> > If we get rid of the links we'll probably have to resort to some sort of
> > magic during the install phase, since a 2.x instal
On Tue, Sep 04, 2001 at 09:08:42AM -0400, Rick Scott wrote:
> Sounds fine. I consider this magic :)
Hmm, while working on it I got an idea of a drawback:
now you can not simply open up a header while editing in
e.g. include/Motif-2.1/Xm.
But I don't think this is a major problem, is it?!
--
Al
Sometimes the LessTif sources lack some helpful comments.
e.g. why do we have _XmClearIconCompatibility() interface which
does nothing and is nowhere documented?
If anyone adds some interface to enhance binary compatibility
this should be documented.
Hmm, as cvs annotate shows this came in in the
Don't you just hate CVS sometimes :) That file was possibly generated with
misc/test14 (I think), or the stubs are the result of an nm. I started that
file a _long_ time ago, then stopped because I didn't know what a Trait was...
On 4-Sep-01 at 15:19, Alexander Mai ([EMAIL PROTECTED]) wrote:
>
How do I get a list of selected cells? I tried the following, but I
keep getting core dump:
int i, j;
int rows, columns;
Boolean **selectedCells;
XtVaGetValues( w, XmNrows, &rows, XmNcolumns, &columns, NULL );
selectedCells = (Boolean **) XtMalloc( sizeof(Boolean *) * rows
