Обсуждение: OS X and ossp-uuid, next chapter

We already know that the uuid-ossp extension doesn't build OS X unless a
small patch is applied.

This has now gotten slightly worse after the Autoconf upgrade, because
it will now fail if a header is present but cannot be compiled.
(Previous versions would only warn.  This is part of a decade-long
transition process Autoconf is doing.)

So now you get:

checking ossp/uuid.h usability... no
checking ossp/uuid.h presence... no
checking for ossp/uuid.h... no
checking uuid.h usability... no
checking uuid.h presence... yes
configure: WARNING: uuid.h: present but cannot be compiled
configure: WARNING: uuid.h:     check for missing prerequisite headers?
configure: WARNING: uuid.h: see the Autoconf documentation
configure: WARNING: uuid.h:     section "Present But Cannot Be Compiled"
configure: WARNING: uuid.h: proceeding with the compiler's result
configure: WARNING:     ## ---------------------------------------- ##
configure: WARNING:     ## Report this to pgsql-bugs@postgresql.org ##
configure: WARNING:     ## ---------------------------------------- ##
checking for uuid.h... no
configure: error: header file <ossp/uuid.h> or <uuid.h> is required for
OSSP-UUID

A possible patch is to hack up the uuid.h check to revert to the old
behavior; see attached.

I don't necessarily want to apply that, because it's an OS-specific hack
and it doesn't even work by itself unless we also patch the place where
the header is used (previously discussed).  But I'll put it on record
here for future reporters and for the benefit of packagers.

On Sat, Feb 15, 2014 at 4:17 PM, Peter Eisentraut <peter_e@gmx.net> wrote:
> We already know that the uuid-ossp extension doesn't build OS X unless a
> small patch is applied.
>
> This has now gotten slightly worse after the Autoconf upgrade, because
> it will now fail if a header is present but cannot be compiled.
> (Previous versions would only warn.  This is part of a decade-long
> transition process Autoconf is doing.)
>
> So now you get:
>
> checking ossp/uuid.h usability... no
> checking ossp/uuid.h presence... no
> checking for ossp/uuid.h... no
> checking uuid.h usability... no
> checking uuid.h presence... yes
> configure: WARNING: uuid.h: present but cannot be compiled
> configure: WARNING: uuid.h:     check for missing prerequisite headers?
> configure: WARNING: uuid.h: see the Autoconf documentation
> configure: WARNING: uuid.h:     section "Present But Cannot Be Compiled"
> configure: WARNING: uuid.h: proceeding with the compiler's result
> configure: WARNING:     ## ---------------------------------------- ##
> configure: WARNING:     ## Report this to pgsql-bugs@postgresql.org ##
> configure: WARNING:     ## ---------------------------------------- ##
> checking for uuid.h... no
> configure: error: header file <ossp/uuid.h> or <uuid.h> is required for
> OSSP-UUID
>
> A possible patch is to hack up the uuid.h check to revert to the old
> behavior; see attached.
>
> I don't necessarily want to apply that, because it's an OS-specific hack
> and it doesn't even work by itself unless we also patch the place where
> the header is used (previously discussed).  But I'll put it on record
> here for future reporters and for the benefit of packagers.

Given commit e6170126fc201052b0ec5fc92177eb181d602d26, I think we
should just remove uuid-ossp from our tree.  Somebody can maintain it
on PGXN if they feel the urge.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Robert Haas <robertmhaas@gmail.com> writes:
> On Sat, Feb 15, 2014 at 4:17 PM, Peter Eisentraut <peter_e@gmx.net> wrote:
>> We already know that the uuid-ossp extension doesn't build OS X unless a
>> small patch is applied.
>> 
>> This has now gotten slightly worse after the Autoconf upgrade, because
>> it will now fail if a header is present but cannot be compiled.

> Given commit e6170126fc201052b0ec5fc92177eb181d602d26, I think we
> should just remove uuid-ossp from our tree.  Somebody can maintain it
> on PGXN if they feel the urge.

I think that's premature, since that commit hasn't shipped yet.  Give
people a release or two to switch over.

I'd be good with putting a notice on uuid-ossp saying that it's deprecated
and will be removed soon.  Of course, our track record for enforcing such
things is bad, but we gotta start somewhere ...
        regards, tom lane