Re: Re: Proposal for encrypting pg_shadow passwords - Mailing list pgsql-patches

From Peter Eisentraut
Subject Re: Re: Proposal for encrypting pg_shadow passwords
Date
Msg-id Pine.LNX.4.30.0108152326510.677-100000@peter.localdomain
Whole thread Raw
In response to Re: Re: Proposal for encrypting pg_shadow passwords  (Bruce Momjian <pgman@candle.pha.pa.us>)
Responses Re: Re: Proposal for encrypting pg_shadow passwords
List pgsql-patches
Bruce Momjian writes:

> > Hmm... I used to be able to build in a separate tree than my source
> > tree. Since this patch was applied, I can't. It seems that the
> > preprocessor is finding the system md5.h (i.e. /usr/include/md5.h)
> > rather than the PostgreSQL md5.h. Suggestions on how to fix?

I kind of doubt that, since the new code does #include "libpq/md5.h".

> Ewe, you have md5.h somewhere else.  Let me rename it to pgmd5.h.  Is
> that OK for everyone?  Or maybe I should move it all into crypt.h.  That
> seems to make more sense.

No, the problem is that you are not supposed to have include files in the
backend subtree.  I'm surprised the code builds at all.

--
Peter Eisentraut   peter_e@gmx.net   http://funkturm.homeip.net/~peter


pgsql-patches by date:

Previous
From: Gerhard Häring
Date:
Subject: Re: Fix for fetchone() and fetchmany() in Python interface
Next
From: Neil Padgett
Date:
Subject: Re: Fix for fetchone() and fetchmany() in Python interface