According to Ralf Wildenhues on 3/1/2010 11:42 PM:
>>> -[m4_ifvaln([$2], [{ AS_SET_STATUS([$2])])]dnl
>>> -[as_fn_error "_AS_QUOTE([$1])"m4_ifval(AS_MESSAGE_LOG_FD,
>>> -  [ "$LINENO" AS_MESSAGE_LOG_FD])[]m4_ifval([$2], [; }])])
>>> +[as_fn_error m4_default([$2], [$?]) 
>>> "_AS_QUOTE([$1])"m4_ifval(AS_MESSAGE_LOG_FD,
>>> +  [ "$LINENO" AS_MESSAGE_LOG_FD])])
>>
>> This can be simplified, to calling as_fn_error $? "_AS_QUOTE([$1])"...
>>
>> Why?  Because we just called AS_SET_STATUS([$2]) the line before, so $? is
>> guaranteed to be the correct value.  And one less m4_default is that much
>> less m4 effort.
> 
> Erm, the patch just removed that AS_SET_STATUS.  If weighing m4 against
> sh time, I usually try to go for lower sh time.

Phooey - I misread the patch.  Indeed, you are correct: with the removal
of AS_SET_STATUS, your patch is correct as-is, and doesn't need any tweaking.

Hmm, that means that as_fn_error no longer requires that we drag in the
definitions used in AS_SET_STATUS; maybe there's another round of cleanup
patches we can do to separate those two, but it is not as high of a priority.

-- 
Eric Blake   [email protected]    +1-801-349-2682
Libvirt virtualization library http://libvirt.org

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to