:
Some command-line SQL tools use statement terminators, like a
semicolon, to indicate the end of a statement. Such terminators
should not normally be used with the DBI.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
ocesses on
the machine.
I think that ora_connect_with_default_signals is misguided.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
a way of specifying a schema name different
from
the user name, it needs fixing.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
BI::errstr\n";
> $sth->bind_param_inout(":p_Target",\$p_Target, 20) or die
> "bind(target) err is $DBI::errstr"; # Should I specify a bind type and if
> so, what type?
> $sth->bind_param(":p_ID",$i) or die "bind(ID) err i
by doing
NLS_LANG="GERMAN_GERMANY.**AL32UTF8" ./your-test-script
greetings
hermann
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
emo-rdbms.mk for 32-bit compilations in 32-bit
environments.
> Are you setting the 'LD_LIBRARY_PATH' ENV value??
I have never found this to be necessary except with Instant Clients.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, F
gt; Alternatively you can use Oracle Instant Client.
>
> In the unlikely event that a suitable *.mk file is installed
> somewhere non-standard you can specify where it is using the -m
> option:
> perl Makefile.PL -m /path/to/your.mk
>
>
irs_h,"$Config::Config{archlib}/CORE"]);
# force addition of __restrict into keywords (asap)
diff -ur Oracle-OCI-0.06.clean/utility.c Oracle-OCI-0.06.patched/utility.c
--- Oracle-OCI-0.06.clean/utility.c 2001-09-01 01:03:05.0 +0100
+++ Oracle-OCI-0.06.patched/utility.
ll you please stop spamming
this list.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
tarts.
> $ENV{TNS_ADMIN} ||= "$ENV{ORACLE_HOME}/network/admin";
>
> my $dbh = DBI->connect("dbi:Oracle:", [user], [password])
> || die "Couldn't connect to Oracle: $!";
> #
>
> # more stuff here that never executes
>
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
ent $dbh of the statement handle.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
hat a crash.
John, does this give you enough to allow you to spot the bug?
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
*** glibc detected *** /usr/local/bin/perl: free(): invalid size:
0x00968500 ***
===
Solaris 10. I can now run some of my production scripts, which
use objects, on the sparc.
I recommend that this patch be incorporated into version 1.28.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
diff -r -u DBD-Oracle
nt.
I see your problem if I use the Oracle 11.2 instant client.
The versions of everything else are the same.
I am afraid that I have very little time to work on this, so a solution
may not come quickly.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
eserved word. If I substitute 'dual' for 'table', I get the
expected results.
Can you let us know whether your problem can be reproduced with 'dual' as the
FROM clause. If not, can you tell us what, if anything, is special about the
actual FROM clause in your tes
this second problem. I would like to know
either way.
I have no big endian hardware on which to test these patches.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
diff -r -u DBD-Oracle-1.24/dbdimp.h DBD-Oracle-1.24.patc
On 22/04/10 14:54, Charles Jardine wrote:
> On 21/04/10 18:56, Everett, Timothy S wrote:
>> Thank you for the patch. This fixed the ORA-24334 problem.
>
> Thank you very much for testing the patch. I will try to ensure that
> a similar fix gets into the next release of DBD:
On 22/04/10 14:54, Charles Jardine wrote:
> On 21/04/10 18:56, Everett, Timothy S wrote:
>> Thank you for the patch. This fixed the ORA-24334 problem.
>
> Thank you very much for testing the patch. I will try to ensure that
> a similar fix gets into the next release of DBD:
need to bother the list
with the output.)
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
#!perl -w
use DBI;
use DBD::Oracle qw(ORA_RSET SQLCS_NCHAR);
use strict;
use Data::Dumper;
DBI->trace(9);
use Test::More tes
1072
> fields = 10094d960
The field_count above should be 2. It is 2 * 2^16.
Could you please try the attached patch. I think it will fix the problem,
but I don't have any SPARC hardware to test it on.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.a
has received no maintenance since August 2001.
I have attached a patch I have used to make a version which
can be built against a full Oracle 10.2 client.
I am offering this patch *completely unsupported*. I simply
do not have the time to maintain Oracle::OCI.
--
Charles Jardine - Computing Service,
ORACLE_SID and using
a tns alias.
On Thu, 2009-10-29 at 16:51 +, Charles Jardine wrote:
On 29/10/09 15:51, jeff wrote:
External connections Orcale 8:
If the remoter server has a username that matches the username of the
remote client it will still authenticate. This work in our setup. You
don&#
ular it is
usne to find the server instance.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
cess. Clashing linker names in the libraries
will cause segfaults etc..
You have to find a way of establishing your two connections
in different processes.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
cal/lib/oracle/instantclient/lib
... the above is wrong. Lose the final '/lib'
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
than happy to test release candidates for 1.24.
Charles Jardine wrote:
On 07/10/09 11:47, Martin Evans wrote:
Charles Jardine wrote:
I have encountered a problem with the 'extended' support for
objects in DBD::Oracle. Specifically, if $dbh->{ora_objects}
is true, any attempt to fetc
On 07/10/09 11:47, Martin Evans wrote:
Charles Jardine wrote:
I have encountered a problem with the 'extended' support for
objects in DBD::Oracle. Specifically, if $dbh->{ora_objects}
is true, any attempt to fetch a nested table of objects
causes perl to crash with a segme
y $dbh = DBI->connect('dbi:Oracle:', , {RaiseError=>1});
$dbh->{ora_objects} = 1;
my $sth = $dbh->prepare("select * from groups"); $sth->execute;
print("about to fetch\n");
$sth->fetch;
print("fetched\n");
$sth->finish; $dbh->disconn
oenloaded DBI-1.40 and DBD-Oracle-1,19 from CPAN and
searched both distributions for the word 'Aborted', calls of
abort(), and mentions of 'ABRT'. I have found nothing.
I am suspicious of the package perl-DBI-1.40-8.x86_64.rpm. Where
did it come from? Why is the version so old? DB
servers must be irrelevant in this
context.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
it
by adding
bequeath_detach = yes
to your sqlnet.ora file.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
On 08/05/09 15:26, E R wrote:
On Fri, May 8, 2009 at 5:40 AM, Charles Jardine wrote:
You do not say which perl driver your are using to talk to Oracle.
The DBD::Oracle driver binds all placeholders as strings by default.
Are you using DBD::Oracle?
Yes, I am using DBD::Oracle.
What other
942
Is there a way to ensure that the parameter is passed to Oracle as a string?
You do not say which perl driver your are using to talk to Oracle.
The DBD::Oracle driver binds all placeholders as strings by default.
Are you using DBD::Oracle?
--
Charles Jardine - Computing Service, University o
ave used a password when running 'make test'.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
--- DBD-Oracle-1.23/t/58object.t2009-04-29 16:04:18.0 +0100
+++ DBD-Oracle-1.23.patched/t/58object.t
On 26/02/09 16:31, Charles Jardine wrote:
I have built a 64 bit version of Perl 5.10.0 under SLES 10 (x86-64).
I am now installing DBI and DBD::Oracle.
Both of these modules produce alarming warnings at compile time.
E.g.
DBI.xs: In function ‘dbih_setup_fbav’:
DBI.xs:1549: warning: format
x86-64) RC3
passes all its tests both with and without the patch. With the patch,
it compiles without warnings - *except* those which come from the two
DBI macros mentioned above.
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223
ll try to get a patch out today.
Charles Jardine wrote:
I have built a 64 bit version of Perl 5.10.0 under SLES 10 (x86-64).
I am now installing DBI and DBD::Oracle.
Both of these modules produce alarming warnings at compile time.
E.g.
DBI.xs: In function ‘dbih_setup_fbav’:
DBI.xs:1549: warni
L
dbiproxy
Extracted dbiproxy from dbiproxy.PL with variable substitutions.
dbiproxy syntax OK
cp dbiproxy blib/script/dbiproxy
/usr/local/perl/5.10.0-B/bin/perl "-MExtUtils::MY" -e "MY->fixin(shift)"
blib/script/dbiproxy
/usr/local/perl/5.10.0-B/bin/perl "-Iblib/arch" "-Iblib/lib" dbilogstrip.PL
dbilogstrip
Extracted dbilogstrip from dbilogstrip.PL with variable substitutions.
dbilogstrip syntax OK
cp dbilogstrip blib/script/dbilogstrip
/usr/local/perl/5.10.0-B/bin/perl "-MExtUtils::MY" -e "MY->fixin(shift)"
blib/script/dbilogstrip
--
Charles Jardine - Computing Service, University of Cambridge
c...@cam.ac.ukTel: +44 1223 334506, Fax: +44 1223 334679
racle cannot cope
with some of the columns in the actual cursor returned. Specifically,
it is telling you that the types of the 9th, 18th, 19th and 20th columns
are user-defined Object types. DBD::Oracle does not support these types.
--
Charles Jardine - Computing Service, University of Cambridg
#x27;Commit','Other','Application',
'Configuration','Concurrency','Cluster','Administrative')
AND ash.event_id(+) = evt.event_id
GROUP BY evt.wait_class
ORDER BY count(ash.event_id) DESC &qu
l
/usr/local/cm58/lib/perl5/5.8.7/sun4-solaris-64
/usr/local/cm58/lib/perl5/5.8.7
/usr/local/cm58/lib/perl5/site_perl/5.8.7/sun4-solaris-64
/usr/local/cm58/lib/perl5/site_perl/5.8.7
/usr/local/cm58/lib/perl5/site_perl
.
Database Name is uclde
$rc [256]
$! []
$rc [0]
$! []
end of script
I was wondering if anyone knows the cause of why the test in
't/10general.t' failed.
Albert Tom
2600 Camino Ramon
San Ramon, Ca. 94583
925-823-9881
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
data found" exception.
Oracle support have insisted that it is not a bug. It is part
of the same mechanism by which
select (select * from dual where 0=1) from dual
returns a null value rather than an exception.
I have been severely burnt buy this, and now avoid CALL altogether.
--
Charles Jard
antec suggested I add $SIG{ CHLD } = 'DEFAULT'; after
the Oracle connect to restore the default sig handler; this also corrects
it.
I wouldn't do that. Oralce must have some use for the SIGCHLD
handler.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]
using the bequeather?
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
/2005/08/msg27335.html
As you will see, the thread never reached a satisfactory conclusion.
Sadly, I have not been able to find the time to work on anything
DBI related since that posting.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506
Bruce Sears wrote:
Charles Jardine wrote:
Most versions of Unix impose restrictions on run time linking to
prevent trojan horse attacks on privileged programs. The following
extract from the man page to ld.so.1(1) on Solaris describes one
version of this as follows:
Secure processes
crifice
security to gain performance.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
low the CGI user to search the directory?
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
it 0;') is failing.
I am afraid I do not have the time to investigate this myself at
the moment, but it does need investigating. A good start would be
to modify the test to print the value of $! after the failure.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
riable
TWO_TASK to the same value. This is usually appropriate for
connecting to remote databases.
If you want to connect directly to a database on the local
computer, the simplest way is to use the ORACLE_SID environment
variable. You can set this within the perl program if you
wish. Vis:
$ENV{ORACLE_S
ors.
2. Does you result set have LONG or LONG RAW columns? Caching is
disabled if you have these.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
{AutoCommit=>0, RaiseError=>1, PrintError=>0, ShowErrorStatement=>1});
};
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
necessary is to set ORACLE_HOME to point as the same installation
of Oracle as was used to compile and link DBD::Oracle.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
OP's problem is that the Oracle
client installation he is using has been upgraded since his
DBD::Oracle was compiled. Other possibilities are that the
DBD::Oracle was compiled using a different ORACLE_HOME than
tha used at run-time, or that dynamic linking is being done
wrongly.
--
Charles Ja
wrong on this point,
but I have checked the 9i and 10g OCI manuals, and I
am depressingly sure that I am right.
So - my point remains. DBD::Oracle, as designed, cannot
implement the suggested feature.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44
rt it.
This seems to me to be a sufficient reason not to attempt
to extend the DBI to provide this facility. Perhaps it would
be better if the authors of DBDs which can support it could
be persuaded to do it using a special bind value, as DBD:Pg
apparently does does.
--
Charles Jardine - Computing Se
There is a PL/SQL example of using rowid instead of WHERE CURRENT
OF in the PL/SQL manual. Look for "Fetching Across Commits" in
the index.
--
Charles Jardine - Computing Service, University of Cambridge
Tim Bunce wrote, on 17/08/2005 21:23:
On Wed, Aug 17, 2005 at 06:01:03PM +0100, Charles Jardine wrote:
Tim Bunce wrote, on 15/08/2005 14:36:
On Mon, Aug 15, 2005 at 12:07:03PM +0100, Charles Jardine wrote:
This what I am proposing for prepare. It already happens for values
sent of returned
Tim Bunce wrote, on 15/08/2005 14:36:
On Mon, Aug 15, 2005 at 12:07:03PM +0100, Charles Jardine wrote:
Peter J. Holzer wrote, on 14/08/2005 09:23:
However, I am sure there exist tons of code which expect that perl
strings are byte arrays in the client character set. This code would
then
Peter J. Holzer wrote, on 14/08/2005 09:23:
On 2005-08-12 16:45:50 +0100, Charles Jardine wrote:
The method $dbh->prepare($stmt) of DBD::Oracle ignores the
state of the utf8 flag in the SV for $stmt.
[ snip ]
This is clearly a bug. It can affect any SQL statement which
contains a
Tim Bunce wrote, on 12/08/2005 22:21:
On Fri, Aug 12, 2005 at 04:45:50PM +0100, Charles Jardine wrote:
The method $dbh->prepare($stmt) of DBD::Oracle ignores the
state of the utf8 flag in the SV for $stmt.
[ details snipped ]
Can anyone suggest a way forward?
This change to Driver.
nces.
1. The bug must be present in all drivers which use Driver.xst :=(
2. I can't see at all how to fix it for DBD::Oracle. It seems that
a fix _requires_ an incompatible change to Driver.xst. The
bug cannot be fixed without change to the prototype for
dbd_st_prepare. Howeve
is deemed
acceptable in principle, I will be happy to re-work it against
the latest development version.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
diff -r -u -N DBD-Oracle-1.16/Oracle.pm DBD-Oracle-1.16-nested
Michael A Chase wrote, on 27/04/2005 13:42:
On 04/27/2005 03:01 AM, Charles Jardine said:
Here is a patch to the pod documentation of DBD::Oracle-1.16.
The patch alters the section headed 'Binding Cursors'.
> diff -ur DBD-Oracle-1.16.cursor-docs/Oracle.pm DBD-Oracle-1.16/Oracle.pm
ite
because a proper test requires access to views, including
V$OPEN_CURSOR, which ordinary users cannot see.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
diff -ur DBD-Oracle-1.16.cursor-docs/Oracle.pm DBD-Oracle-1.
Michael Styer wrote:
On Thu, 21 Apr 2005 18:12:08 +0100, "Charles Jardine" <[EMAIL PROTECTED]>
said:
I am sure that Oracle can close the cursors automatically in Oracle
9.2..6. I would be very grateful if you could find out for me if the
same is true in your Oracle 8i database.
Michael Styer wrote, on 21/04/2005 13:17:
Charles Jardine said:
I am still convinced that Michael should not need explicitly to
close his cursors. I would be interested to know if tracing shows
his cursors being destroyed in a timely fashion, since I suspect
that something may be maintaining
works if the bind is changed to match the test, rather than
the documentation.
I am still convinced that Michael should not need explicitly to
close his cursors. I would be interested to know if tracing shows
his cursors being destroyed in a timely fashion, since I suspect
that something may be mainta
Michael Styler wrote:
So now I'm really confused. In this thread
(http://www.mail-archive.com/dbi-users@perl.org/msg24109.html)
Charles Jardine suggests that when the variable goes out of scope the
cursor is closed, but you're saying they stay in existence.
and Ron Reidy replied:
W
Michael A Chase tech wrote:
On 03/16/2005 08:25 AM, Charles Jardine said:
perldoc DBD::Oracle says, under "Binding Cursors":
To close the cursor you (currently) need to do this:
$sth3 = $dbh->prepare("BEGIN CLOSE :cursor; END;");
$sth3->bind_param_inout("
give
the reasonable error message
> Field 2 has an Oracle type (116) which is not explicitly supported
Is anyone thinking about the implementation of the fetching of
nested result sets? If not, I might look at it myself.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTEC
oes anyone know why this was once thought to be necessary?
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
on.
I have attached a patch. I hope this helps someone.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
diff -u -r Oracle-OCI-0.06/boot Oracle-OCI-0.06.patched/boot
--- Oracle-OCI-0.06/bootThu Aug 30 16:49:
to the run-time path of the Oracle libraries themselves. This
binding took place at install time. (This is why some Oracle
documentation says that soft links to ORACLE_HOME directories
are not supported).
This may be better with the Oracle 10g "Instant Client", but I
haven't checke
Tim Bunce wrote:
On Thu, Feb 05, 2004 at 10:50:46AM +, Charles Jardine wrote:
If not, would it be useful if I re-worked the patch for
version 1.15.
That would be great. Thanks.
A re-worked patch to make placeholder binding in DBD::Oracle
case insensitive is attached.
--
Charles Jardine
Tim Bunce wrote:
Thanks for the info and the patch!
Tim.
On Fri, Oct 17, 2003 at 02:07:46PM +0100, Charles Jardine wrote:
The problem:
---
When Oracle matches placeholder names passed to OCIBindByName()
with the actual placeholders in the SQL or PL/SQL, it uses
a case-insensitive
Tony Adolph wrote:
Hi Garry,
Install "the lot" has fixed the problem, thanks.
But, and there always a but. We dont want to put the full Oracle 9.2 on
the clients. Do you know if it is possible to unistall the rdbms packages
and still have the DBI / DBD working?
I can't answer for Red Hat. I am
a number */
@@ -1329,7 +1329,13 @@
if (SvGMAGICAL(ph_namesv)) /* eg if from tainted expression */
mg_get(ph_namesv);
if (!SvNIOKp(ph_namesv)) {
+ int i;
name = SvPV(ph_namesv, name_len);
+ if (name_len>31)
+ croak("Placeholder name too long (%s)", neatsvpv(ph_namesv,0));
+ for (i=0; i
===
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
f the required patch is in any 2.x version of mod_perl.
--
Charles Jardine - Computing Service, University of Cambridge
[EMAIL PROTECTED]Tel: +44 1223 334506, Fax: +44 1223 334679
file. Despite this DBD::Oracle passes
all tests except t.long #10. (I will collect more information
about this test failure and post another message.)
Should the warnings be ignored? Is some systematic change required
in DBD::Oracle to make it compatible with PerlIO?
--
Charles Jardine
80 matches
Mail list logo
