Jean-Marc Lasgouttes a écrit :
"Abdelrazak" == Abdelrazak Younes <[EMAIL PROTECTED]> writes:

Abdelrazak> That's just an example. The development process would be
Abdelrazak> _much_ smoother and quicker if the whole tree didn't have
Abdelrazak> to be rebuilt each time you configure.
Do you really have to recompile at each reconfigure? I don't unless
config.h has actually been changed.

But that's my point, one should not have to recompile everything because of a single setting change that impact the compilation of a single file.

Abdelrazak> This is fixed at boost level. Code in BOOST_ASSERT will be
Abdelrazak> executed or not depending on the _BOOST_ policy. You
Abdelrazak> should not have to check for ENABLE_ASSERTION.

Hmm, and how do you tell boost whether we want the assertions or not.

That's exactly what I am saying, it's boost job. There's _no_ need to check for ENABLE_ASSERTION in LyX code. I am not saying that this macro is not useful but that changing it should not imply a complete rebuilding. If you want to enable assertions globally, "make clean" first.

To tell things differently, one _has_ to reconfigure all the object
files where assertions are used. This is how assertions work anyway.

I _know_.

How do we...?

Abdelrazak> Yes?

I am not going to do your homework.

I almost put a smiley here...

Abdelrazak> PS: Am I better now?

No. You have not only to be mean, but also to be right.

I think I am, we just don't have the same definition of truth.

Abdel.

Reply via email to