more on large objects using postgresql.snapshot from today (March 11)

This is a very urgent problem for me to resolve, if there is anything I can do to help figure this out,
please let me know.

Peter has my older test code that failed out 6.3.2 and 6.4.2.
I have compiled my postgres with --enable-cassert.

all of these are done through the jdbc driver.

1) I have two problems with the jdbc driver.
1) the linux (blackdown) jdk 1.1.6 has a version string
of Linux_JDK_1.1.6_v1, not 1.1.6, so the jdbc driver
tries to load the 1.2 driver which of course isn't present..
Minor fix to add a if || getProperty( "java.version").startsWith( LinuxJDK_1.1 )
to the 1.1 test

2) the LARGE OBJECT selector in database metadata is
{"LARGE OBJECT", "(relkind='r' and relname ~ '^xinv')"},

and for this version of postgres appears to need to be
{"LARGE OBJECT", "(relkind='l' and relname ~ '^xinv')"},

I haven't checked into the data handling stuff that is so
problematic...

1) I can't fetch large objects except within transactions, this
is new, It used to be that it didn't matter
2) I can't store more that 1 large object in a transaction,
if i have multiple lo's i have to store them using
autocommit(false).

3) after uploading 280 lo's in in autocommit(true) mode
I get the following message in the postmaster log file

NOTICE: DateStyle is Postgres with US (NonEuropean) conventionsN
OTICE: SIReadEntryData: cache state reset
TRAP: Failed Assertion("!(RelationNameCache->hctl->nkeys == 10):", File: "relcache.c", Line: 1467)

!(RelationNameCache->hctl->nkeys == 10) (0) [No such file or directory]
NOTICE: Message from PostgreSQL backend:
The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory.
I have rolled back the current transaction and am going to terminate your database system connection and exit.
Please reconnect to the database system and repeat your query.
NOTICE: Message from PostgreSQL backend:
The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory.
I have rolled back the current transaction and am going to terminate your database system connection and exit.
Please reconnect to the database system and repeat your query.

-- the message I get if I try to do multiple lo's inside of a transaction is:
NOTICE: DateStyle is Postgres with US (NonEuropean) conventions
TRAP: Bad Argument to Function Call("!(AllocSetContains(set, pointer)):", File: "aset.c", Line: 292)

!(AllocSetContains(set, pointer)) (0) [No such file or directory]
NOTICE: Message from PostgreSQL backend:
The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory.
I have rolled back the current transaction and am going to terminate your database system connection and exit.
Please reconnect to the database system and repeat your query.
NOTICE: Message from PostgreSQL backend:
The Postmaster has informed me that some other backend died abnormally and possibly corrupted shared memory.
I have rolled back the current transaction and am going to terminate your database system connection and exit.
Please reconnect to the database system and repeat your query.