Re: module cleanup [1/n]

[Eric Blake]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

According to Gary V. Vaughan on 9/14/2007 3:43 AM:
>> load(`nosuch', `silent') == load(`nosuch', `quiet') => like sinclude -
>> proceed without warning and without affecting exit status
> 
> Looks good.  Maybe similar extensions for include would be good too?

For include, maybe.  But recall that POSIX requires undivert(1,2) to
undivert from two locations, so our current behavior is undivert(a,b)
reads from two files.  So there is now a symmetry argument - do we make
include(a,silent) act like sinclude(a) (similar to the load proposal), or
like include(a)include(silent) (similar to undivert)?

>> Really?  I thought the whole point of adding the lt_dlinterface_id was to
>> allow multiple loaders.  Anyways, more on this in response to 3/n.
> 
> You're right.  What I meant was that this very system won't work if a
> module is written to meet two interfaces, because both libsnprintfv and
> libm4 could potentially load the same code through two different
> extension points.  When I wrote all the interface checking code it
> seemed like something no-one would want to do anyway, hence the assumption
> that no-one would try to do so...

Is that assumption clearly stated in the libltdl documentation?  Yes, it
simplifies other behavior if one can safely assume no module ever meets
multiple interfaces, but that seems like an unnecessary restriction, and
the GNU philosophy discourages artificial limitations.

- --
Don't work too hard, make some time for fun as well!

Eric Blake             address@hidden
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFG6nyQ84KuGfSFAYARAh2PAKDJGt9cM9n3wi375YepMyZEjN3m8wCfSVtl
scOG452koC/+R2t2Qdz59F0=
=vjdQ
-----END PGP SIGNATURE-----