Skip Menu |

Preferred bug tracker

Please visit the preferred bug tracker to report your issue.

This queue is for tickets about the DBD-Oracle CPAN distribution.

Report information
The Basics
Id: 13865
Status: resolved
Priority: 0/
Queue: DBD-Oracle
People
Owner: champoux [...] pythian.com
Requestors: stefano [...] santoro.org
Cc:
AdminCc:
Bug Information
Severity: Normal
Broken in: 1.16
Fixed in: (no value)

History Show all quoted text
  Mon Jul 25 18:31:52 2005 Guest - Ticket created
Subject: DBI::db::column_info returns wrong column_size
Hi, I have run into this interesting SQL tidbit when I traced a $dbh->column_info call from my script. in the decode section that determines the column_size of a column of type NUMBER, if a NUMBER's data_precision is null then column_size gets a harcoded default of 38 In case of NUMBER would it not be better to use DATA_LENGTH which has the correct COLUMN_SIZE? Ciao Stefano
  Mon May 02 19:28:23 2011 champoux [...] pythian.com - Taken
  Mon May 02 19:30:58 2011 champoux [...] pythian.com - Correspondence added
The code responsible is in DBD/Oracle.pm at line 603 an following. I need to consult w/ Gwen and Jared to know if changing the value there makes sense.
  Mon May 02 19:30:59 2011 The RT System itself - Status changed from 'new' to 'open'
  Mon Sep 12 02:57:22 2011 DAMI [...] cpan.org - Correspondence added
Subject: DBI::db::column_info returns wrong column_size (also for VARCHAR)
Le Lun 25 Jui 2005 18:31:52, guest a écrit : Show quoted text
> in the decode section that determines the column_size of a column of > type NUMBER, if a NUMBER's data_precision is null then column_size > gets a harcoded default of 38
Column size is also wrong on VARCHAR2 data types. According to the DBI spec, it should return the maximum length _in characters_, so DBD::Oracle should return tc.CHAR_LENGTH instead of tc.DATA_LENGTH
  Wed Oct 12 14:57:54 2011 champoux [...] pythian.com - Correspondence added
For the length of INTEGER, the value of 38 seems to be correct, according to what I was able to find around (including http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:1145132537055) For the VARCHAR2 size, though, I think you're are right. I have a patch ready, it'll be merged in the trunk in a few minutes.
  Thu Oct 13 13:27:02 2011 champoux [...] pythian.com - Correspondence added
On Wed Oct 12 14:57:54 2011, PYTHIAN wrote: Show quoted text
> For the length of INTEGER, the value of 38 seems to be correct, > according to what I was able to find around (including >
http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:1145132537055) Show quoted text
> > For the VARCHAR2 size, though, I think you're are right. I have a > patch > ready, it'll be merged in the trunk in a few minutes.
Committed into master: https://github.com/yanick/DBD-Oracle/commit/e27fa49e42f52d635448207ee51689bd1fc513c6 Should come out with 1.33_00 (dev release of next week).
  Thu Oct 13 13:27:04 2011 champoux [...] pythian.com - Status changed from 'open' to 'patched'
  Mon Oct 31 16:42:03 2011 champoux [...] pythian.com - Status changed from 'patched' to 'resolved'