Re: SOLVED (was: Weird SELECT bug in 5.0.21)

2006-08-31 Thread Dan Nelson
In the last episode (Aug 31), Renald Buter said: > On 12:27 Thu 31 Aug , Duncan Hill wrote: > > On Thursday 31 August 2006 12:21, Renald Buter wrote: > > > The problem is that a simple 1-table query shows different > > > answers depending on whether you select 1 or 2 columns. > > *blush* > >

SOLVED (was: Weird SELECT bug in 5.0.21)

2006-08-31 Thread Renald Buter
On 12:27 Thu 31 Aug , Duncan Hill wrote: > On Thursday 31 August 2006 12:21, Renald Buter wrote: > > > The problem is that a simple 1-table query shows different answers > > depending on whether you select 1 or 2 columns. *blush* Of course. I see. How stupid. Thanks and sorry to have bother

Re: Weird SELECT bug in 5.0.21

2006-08-31 Thread Duncan Hill
On Thursday 31 August 2006 12:21, Renald Buter wrote: > The problem is that a simple 1-table query shows different answers > depending on whether you select 1 or 2 columns. Relational databases are founded on mathematical set theory. Unless you specify an ORDER BY stanza in your query, the data

Re: Weird SELECT bug in 5.0.21

2006-08-31 Thread Renald Buter
On 11:34 Thu 31 Aug , Renato Golin wrote: > Renald Buter wrote: > >Odd, eh? But what's worse, the JOIN between this column and other > >columns *also* uses this truncated values and the result is bogus. > > I wouldn't say odd, as you didn't specified any order I wouldn't rely on > the order o

Re: Weird SELECT bug in 5.0.21

2006-08-31 Thread Renato Golin
Renald Buter wrote: Odd, eh? But what's worse, the JOIN between this column and other columns *also* uses this truncated values and the result is bogus. I wouldn't say odd, as you didn't specified any order I wouldn't rely on the order of the output. Try ordering things for what you want and

Weird SELECT bug in 5.0.21

2006-08-31 Thread Renald Buter
Hello list, I've found this strange select bug in retrieving rows from a table. I can best illustrate this with an output of two queries: mysql> select id,jn from paper_2001 limit 10; +--+---+ | id | jn| +--+---+ | 19360350 | 6165 | | 19360351

RE: (simple select) bug? version problem? or just bad SQL?

2004-03-22 Thread Jennifer Horne
them. If you're looking for names where -any- of them can have 'ja' in them, use the OR. Jennifer Horne Panda Voice Systems Inc. 1.888.767.2632 ext 23 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Sent: Monday, March 22, 2004 9:41 AM To: [EMAIL PRO

(simple select) bug? version problem? or just bad SQL?

2004-03-22 Thread ryan
Hi, I am running this (simple?) query on mysql 3.23 : select * from clients_tb where active='on' AND first_name LIKE '%ja%' AND middle_name LIKE '%ja%' AND last_name LIKE '%ja%'"; I know there are people in the table with: first name James last name James middle name Jacob etc but it does not

RE: Create ... select bug

2002-07-16 Thread Jay Blanchard
[snip] CREATE TABLE order_cross SELECT date_format(data_iniz_num,'%y') as anno,ord_tmp.soc,ord_tmp.ml, ord_tmp.cod_agente, ord_tmp.linea, CASE WHEN month(data_ordine) = 1 then qty else 0 END as jan, CASE WHEN month(data_ordine) = 2 then qty else 0 END as feb, CASE W

Create ... select bug

2002-07-16 Thread Andrea Forghieri
Dear Dirs, I think I' ve found a bug. Please run the following query (I know you don' t have my db, try something similar) CREATE TABLE order_cross SELECT date_format(data_iniz_num,'%y') as anno,ord_tmp.soc,ord_tmp.ml, ord_tmp.cod_agente, ord_tmp.linea, CASE WHEN

Create ... select bug

2002-07-16 Thread Andrea Forghieri
Dear Dirs, I think I' ve found a bug. Please run the following query (I know you don' t have my db, try something similar) CREATE TABLE order_cross SELECT date_format(data_iniz_num,'%y') as anno,ord_tmp.soc,ord_tmp.ml, ord_tmp.cod_agente, ord_tmp.linea, CASE WHEN

create table ... select bug

2001-06-14 Thread amackey
>Description: create table ... with a select statement at the end seems to think there are duplicate entries: >How-To-Repeat: mysql> create table nr_current (id int unsigned not null primary key) TYPE=HEAP > select distinct protein_id from nr where current = 1; ERROR 1062: Duplicate entry

Re: Select bug?

2001-02-28 Thread Peter Zaitsev
Hello Mike, Wednesday, February 28, 2001, 3:55:35 AM, you wrote: You should know what you can't do direct comparasions between float number and float const as the number really stored in a database can be slightly different i.e 355.619 M

Select bug?

2001-02-27 Thread Mike Machado
There seems to be a matching bug in MySQL. Why will it match if I do a > of a number smaller that is in the custom_price column, or if I do a strcmp(custom_price, '355.62'), but will NOT match if I do "WHERE custom_price = '355.62'" OR if I try "WHERE custom_price = 355.62"? This happens in b

patch: select bug concerning decimal datatypes and negative values

2001-01-15 Thread Michael Widenius
Hi! > "root" == root <[EMAIL PROTECTED]> writes: >> Description: root> In certain cases the select fails (ie (a) does not return required rows, or (b) returns non-wanted rows (cannot repeat (b))) root> The problem seems to be related to negative values in decimal type fields and in query

Re: select bug concerning decimal datatypes and negative values

2001-01-15 Thread Sinisa Milivojevic
root writes: > >Description: > > In certain cases the select fails (ie (a) does not return required rows, or (b) >returns non-wanted rows (cannot repeat (b))) > The problem seems to be related to negative values in decimal type fields and in >query against those fields. > the problem is

select bug concerning decimal datatypes and negative values

2001-01-15 Thread root
>Description: In certain cases the select fails (ie (a) does not return required rows, or (b) returns non-wanted rows (cannot repeat (b))) The problem seems to be related to negative values in decimal type fields and in query against those fields. the problem is applicable to several 3.22.* ver