Installing DBD::Oracle on Gentoo Linux fails with the following error:
x86_64-pc-linux-gnu-gcc: unrecognized option '-wchar-stdc++'
x86_64-pc-linux-gnu-gcc: unrecognized option '-cxxlib-gcc'
cc1: error: /ee/dev/bastring.h: No such file or directory
I tracked this down to the Oracle XE install pro
strings inside single quotes are evaluated by the shell. they're just passed
straight to the program.
i suspect the problem may be as you suggested: i'm using the old version 8
style sid instead of a service name, although ezconnect with the sid works
fine inside of a perl program. i'll see
John Scoles wrote:
>> [EMAIL PROTECTED] wrote:
>>> Anyone knows if DBD::Oracle has a function similar to the function
>>> $sth->rows that exist in DBD::Mysql?
>>
>> This is a DBI feature. It exists for all drivers.
>>
>>> This function give us the number o rows in a select command. So, we
>>> don
Why do you put your connect info in single quotes?
Here is my stuff.
First, sqlnet.ora:
$ cat sqlnet.ora
DISABLE_OOB = ON
NAMES.DEFAULT_DOMAIN = xx.com
#TRACE_LEVEL_CLIENT = on
#trace_level_client=16
#trace_file_client=client
#trace_directory_client=/tmp
#trace_level_client=admin
#trace_file_cl
hi ron,
thanks for the response. using sqlnet.ora just doesn't work for me. i agree
with you about hidden files -- i don't like them. that's why i'd like to
find a better way. here's what happens using sqlnet.ora:
-
[~] cat sqlnet.ora
NAMES.DIRECTORY_PATH=(ezconnect, tnsnames)
[~] sqlp
Well, this is not a Perl/DBI question, really. I personnaly do not use
TWO_TASK as it is an archaic environment variable that has been around
since version 4 of the DB.
If you are interested in using the Easy Connect method, you should read
the Oracle docs:
http://download-east.oracle.com/docs/cd
John Scoles wrote:
> Oracle doesn't have this feature so DBD::Oracle doesn't either.
> This has cropped up many times before So you can blame Oracle on that.
> Thier logic (as it was explained to me) is
>
> "How could you get the number of rows you are going to fetch without
> counting them all fi
reading man DBD::Oracle gives an in depth explanation on how the TWO_TASK
environment variable can be used. however, when i tried to use it in place
of .tnsnames.ora nothing worked. i believe that sql*plus does not use
TWO_TASK.
-
[~] cat .tnsnames.ora
infra=(description=(address=(protoco
On 3/25/07, Reidy, Ron <[EMAIL PROTECTED]> wrote:
This tells me there is something not set in your environment. Have you
tried setting the environment like I specified?
Here's a handy method for getting the environment correct
in a script (DBI or otherwise) to run from cron when all
else see
Oracle doesn't have this feature so DBD::Oracle doesn't either.
This has cropped up many times before So you can blame Oracle on that.
Thier logic (as it was explained to me) is
"How could you get the number of rows you are going to fetch without
counting them all first?"
or someting very co
[EMAIL PROTECTED] wrote:
> Anyone knows if DBD::Oracle has a function similar to the function
> $sth->rows that exist in DBD::Mysql?
This is a DBI feature. It exists for all drivers.
> This function give us the number o rows in a select command. So, we
> don´t have to run a fetchrow with a while
Thanks for the pointers. I was able to successfully compile and link
DBD::Oracle but the tests are failing. If I try to use the module, it
dumps core during connect.
-Original Message-
From: John Scoles [mailto:[EMAIL PROTECTED]
Sent: Wednesday, April 11, 2007 8:04 AM
To: dbi-users@
Well no exact answer but I might be able to point you in the right
direction. Seems Solaris has been a problem for a while.
Have a look at these links
http://www.zorranlabs.com/articles/step-by-step-installation-dbd-oracle-1.16-on-solaris-9-and-oracle-9-client.htm
http://research.imb.uq.edu.au/~
13 matches
Mail list logo
