[Bug 599097] Review Request: libgexiv2 - Gexiv2 is a GObject-based wrapper around the Exiv2 library

bugzilla at redhat.com bugzilla at redhat.com
Sat Jun 5 07:40:04 UTC 2010


Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=599097

--- Comment #13 from Martin Gieseking <martin.gieseking at uos.de> 2010-06-05 03:39:58 EDT ---
(In reply to comment #12)
> PS : I dont understand how including the make portion(the new hack) works any
> different from the earlier hack you had mentioned.    

It has to do with the way the shell handles variables. Each shell gets its own
environment so that variables defined there are usually local and unknown
outside. Adding "export" makes a variable accessible in the current shell S and
also in all its sub-shells, but after returning to the parent of S they are
removed. 
So if I, for instance, define CFLAGS in a shell script "build.tmp" and use it
in a spec file, CFLAGS is known inside the shell calling "build.tmp" but not in
the parent (rpm) environment, i.e. all variable assignments are lost when
returning to the parent (rpm) shell. Thus, the following "make" process doesn't
know anything about the previously set CFLAGS variable. But when placing the
make statement inside "build.tmp", the variable is still accessible. 
I hope that's understandable. :)

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.



More information about the package-review mailing list