bug? - Mailing list pgsql-hackers

From Joe Conway
Subject bug?
Date
Msg-id 3D7AEBE7.9090200@joeconway.com
Whole thread Raw
Responses Re: bug?  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
I found the following while poking around. RangeVarGetRelid takes a 
second parameter that is intended to allow it to not fail, returning 
InvalidOid instead. However it calls LookupExplicitNamespace, which does 
not honor any such request, and happily generates an error on a bad 
namespace name:

/* * RangeVarGetRelid *     Given a RangeVar describing an existing relation, *     select the proper namespace and
lookup the relation OID. * * If the relation is not found, return InvalidOid if failOK = true, * otherwise raise an
error.*/
 
Oid
RangeVarGetRelid(const RangeVar *relation, bool failOK)
{
[...]  if (relation->schemaname)  {    /* use exact schema given */    namespaceId =
LookupExplicitNamespace(relation->schemaname);   relId = get_relname_relid(relation->relname, namespaceId);  }
 
[...]
}

Oid
LookupExplicitNamespace(const char *nspname)
{
[...]  namespaceId = GetSysCacheOid(NAMESPACENAME,                  CStringGetDatum(nspname),0, 0, 0);  if
(!OidIsValid(namespaceId))   elog(ERROR, "Namespace \"%s\" does not exist", nspname);
 
[...]
}

Shouldn't LookupExplicitNamespace be changed to allow the same second 
parameter? All uses of LookupExplicitNamespace, besides in 
RangeVarGetRelid, would have the parameter set to false.

Comments?

Joe



pgsql-hackers by date:

Previous
From: "Matthew T. O'Connor"
Date:
Subject: Re: --with-maxbackends
Next
From: Michael Meskes
Date:
Subject: 7.3beta and ecpg