Bug report

Errrgh... jolly(at)cs(dot)berkeley(dot)edu seems to be unexistent :(

Your name : Denis N. Stepanov
Your email address : D(dot)N(dot)Stepanov(at)inp(dot)nsk(dot)su

System Configuration
---------------------
Architecture (example: Intel Pentium) : Intel Pentium

Operating System (example: Linux 2.0.26 ELF) : Linux 2.0.37 ELF

PostgreSQL version (example: PostgreSQL-6.5.3): PostgreSQL-6.5.1

Compiler used (example: gcc 2.8.0) : gcc 2.7.2.3

Tcl/Tk version : 8.0

Please enter a FULL description of your problem:
------------------------------------------------

An error statement issued when transactions rolls back is not seen
in pgtclsh/pgtksh scripts.

Please describe a way to repeat the problem. Please try to provide a
concise reproducible example, if at all possible:
----------------------------------------------------------------------

Generally say, I don't think that this problem depends on particular
architecture, OS or even compiler.

1. Create test database:

$> createdb test
$>

2. Enter into it and create test table:

$> psql test
Welcome to the POSTGRESQL interactive sql monitor:
Please read the file COPYRIGHT for copyright terms of POSTGRESQL
[PostgreSQL 6.5.1 on i586-pc-linux-gnu, compiled by gcc 2.7.2.3]

type \? for help on slash commands
type \q to quit
type \g or terminate with semicolon to execute query
You are currently connected to the database: test

test=> CREATE TABLE test (value int2);
CREATE
test=> \q
$>

3. Create test script with invalid insertion:

$> cat >test.pgtclsh
set DBhandle [pg_connect test]
pg_result [pg_exec $DBhandle "insert into test values (100000)"] -clear
pg_disconnect $DBhandle
^D
$>

4. Execute it:

$> pgtclsh test.pgtclsh
$>

As you will see, this script executes succesfully, though no lines will be
inserted actually. Invoking the same insert statement directly in psql causes
the following message:

ERROR: i4toi2: '100000' causes int2 overflow

So, inside the script I can't determine whether operation succeded or not.

If you know how this problem might be fixed, list the solution below:
---------------------------------------------------------------------

The latest distribution of Postgres (6.5.3) behaves similarly. But
pgtclsh/pgtksh in 6.3.2 distribution in exactly the same situation fails with
the following error message:

ERROR: i4toi2: '100000' causes int2 overflow

while executing
"pg_exec $DBhandle "insert into test values (100000)""
invoked from within
"pg_result [pg_exec $DBhandle "insert into test values (100000)"]..."
(file "test.pgtclsh" line 2)

Even if it is not a bug, I don't like such difference between distributions,
and this is the thing that keeps me from upgrading to 6.5.x. All my program
stuff is used to catch such errors from a database.

Please, send me a reply for my awareness of your opinion.

Sorry for English mistakes, if any.

Sincerely,
Denis N. Stepanov
BINP SB RAS, Novosibirsk, Russia.