First of all, my apologies for the huge patch. After coming out of my
cocoon I noticed the thread concerning the submission of a large patch.
While I understand the reasoning behind your reluctance to accept such
patches, I am hoping that you'll accept it as it stands though for a few
reasons.
It doesn't touch even one line of actual FLTK code.
A considerable amount of care was put into simplifying and regrouping
the script code into clearly marked sections with the intent of making
the CMake code easier to maintain, whether by myself or someone else.
In other words, it should be even easier to fix the new scripts than the
old ones.
In my review of the STRs relating to CMake a comment was made by
AlbrechtS concerning the fact that CMake was a feature "most of the core
developers don't use or have available". This being the case I can
foresee a group of smaller patches being allowed to stagnate even longer
than one big patch. This feeling seems to be re-enforced by STR 2088,
which seems to be an earlier effort to do, on a more limited scale, what
I have just submitted.
Still, I can break my patch up if that's what needs to be done.
Now, onto my observations.
Only after submitting the patch did I think of reviewing the other CMake
related patches. Here are some observations of those STRs relative to
my patch.
2299 - fltk 1.3 6970 do not compile with cmake on osx 10.6
Manlo suggests compiling Fl.cxx with the -x objective-c++ compile flag.
This could easily be done with the set_source_file_properties command.
Everything else seems to have been done already.
2244 - This is already all but closed.
2088 - [CMake] update to make it work again
What seems to be an earlier effort at updating the CMake build. One
could say that my STR is for the 1.3 branch what this is for the 1.4 branch.
Interestingly this is where adding the -DWIN32_LEAN_AND_MEAN flag was
first suggested. This is what was eventually patched in with STR 2244.
The submitter, MaikBeckmann, also used the cmake feature #cmakedefine01
which was already implemented at 2.4 but undocumented until CMake 2.8.
Even though I run 2.8, I was looking at the 2.6 documentation, so I
missed it. It definitely would have simplified my scripts.
So should I wait and see if the currently submitted patch gets accepted
and then submit a new patch to implement Manlo's suggestion on STR 2299
and yet another to simplify my code?
Alternatively I could apply those changes and re-submit, either as one
big patch or a series of patches. Personally, this would be my second
choice.
Mike
_______________________________________________
fltk-dev mailing list
[email protected]
http://lists.easysw.com/mailman/listinfo/fltk-dev
