octave-bug-tracker
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #59806] SuiteSparse headers not found


From: anonymous
Subject: [Octave-bug-tracker] [bug #59806] SuiteSparse headers not found
Date: Mon, 4 Jan 2021 20:57:34 -0500 (EST)
User-agent: Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:84.0) Gecko/20100101 Firefox/84.0

Follow-up Comment #11, bug #59806 (project octave):

[comment #3 comment #3:]
> I see how changing the order in which OCTAVE_CHECK_LIB checks for headers
might make a difference.
> But if I understand that function correctly, it stops checking as soon as it
found the first header in the list:
>
https://hg.savannah.gnu.org/hgweb/octave/file/70ae19d4e67b/m4/acinclude.m4#l847
> 
> Does changing the order of the includes in oct-sparse.h make any
difference?
> 
> I'd agree that testing for the header without sub-folder first is probably a
good idea.
> I don't know how different distributions install these headers. So, I'm not
sure if changing the order in which the different sub-folders are checked
could be important.
> Is there a reason you changed those as well?

Changing the OCTAVE_CHECK_LIB order does the real magic; changing the order in
oct-sparse.h was cosmetic and meant to match how OCTAVE_CHECK_LIB searches. 
As long as nobody is hand-editing their config.h files and enabling more than
one HAVE_*_H option at a time, the oct-sparse.h change isn't strictly
necessary.

My only concern is if it happens to find some unrelated header with the same
name in a default search path, when previously it would have found the one in
a subdirectory first.  That seems like a more general problem of naming
conflicts between packages and ordering the CPPFLAGS correctly to avoid it if
it does happen.

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?59806>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]