Обсуждение: Insert fails when it shouldn't
The table:
CREATE TABLE DILOSIS (
DL_AEM INT NOT NULL,
DL_CODE VARCHAR(6) NOT NULL,
DL_YEAR INT NOT NULL,
DL_GRADE FLOAT(3) NOT NULL,
DL_LESTYPE INT NOT NULL,
DL_SEMESTER INT NOT NULL,
DL_DM INT NOT NULL,
DL_COEF FLOAT(3), --Syntelestis ptyxiou
DL_CAT_NORM BOOLEAN NOT NULL, --TRUE NORMAL, FALSE APALLAGI
DL_TEI INT NOT NULL,
DL_THERINO BOOLEAN NOT NULL
);
ALTER TABLE DILOSIS ADD CONSTRAINT pk_dilosis PRIMARY KEY (DL_AEM, DL_CODE, DL_YEAR, DL_TEI, DL_SEMESTER);
The table is empty and statements
stored in a file are executed with psql.
===== Statements ===
\encoding iso_8859_7
delete from dilosis;
INSERT INTO DILOSIS VALUES(1, 'ΓΕ0100', 2000, 0.00, 1, 1, 5, 5.00, true, 1, true);
INSERT INTO DILOSIS VALUES(1, 'ΗΥ0100', 2000, 0.00, 1, 1, 5, 5.00, true, 1, true);
The second insert fails with an error message:
psql:a.sql:4: ERROR: duplicate key violates unique constraint "pk_dilosis"
The second insert is identical to the first one except
the DL_CODE value.
1:ΓΕ0100 (Gamma Epsilon ..)
2:ΗΥ0100 (Heta Ypsilon ..)
I think what is hapening is that the two first
letters are ignored and the two values are found
identical since the postfix is 0100 for bouth of them.
What can be done so the second instert does not fail?
Anakreon
--
Three words describe our society:homo homini lupus
What encoding did you initdb with? On Thu, Mar 02, 2006 at 04:59:39PM +0200, Anakreon Mendis wrote: > The table: > CREATE TABLE DILOSIS ( > DL_AEM INT NOT NULL, > DL_CODE VARCHAR(6) NOT NULL, > DL_YEAR INT NOT NULL, > DL_GRADE FLOAT(3) NOT NULL, > DL_LESTYPE INT NOT NULL, > DL_SEMESTER INT NOT NULL, > DL_DM INT NOT NULL, > DL_COEF FLOAT(3), --Syntelestis ptyxiou > DL_CAT_NORM BOOLEAN NOT NULL, --TRUE NORMAL, FALSE APALLAGI > DL_TEI INT NOT NULL, > DL_THERINO BOOLEAN NOT NULL > ); > ALTER TABLE DILOSIS ADD CONSTRAINT pk_dilosis PRIMARY KEY (DL_AEM, DL_CODE, DL_YEAR, DL_TEI, DL_SEMESTER); > > The table is empty and statements > stored in a file are executed with psql. > ===== Statements === > \encoding iso_8859_7 > delete from dilosis; > INSERT INTO DILOSIS VALUES(1, '??0100', 2000, 0.00, 1, 1, 5, 5.00, true, 1, true); > INSERT INTO DILOSIS VALUES(1, '??0100', 2000, 0.00, 1, 1, 5, 5.00, true, 1, true); > > The second insert fails with an error message: > psql:a.sql:4: ERROR: duplicate key violates unique constraint "pk_dilosis" > > The second insert is identical to the first one except > the DL_CODE value. > 1:??0100 (Gamma Epsilon ..) > 2:??0100 (Heta Ypsilon ..) > > I think what is hapening is that the two first > letters are ignored and the two values are found > identical since the postfix is 0100 for bouth of them. > > What can be done so the second instert does not fail? > > Anakreon > -- > Three words describe our society:homo homini lupus > > > ---------------------------(end of broadcast)--------------------------- > TIP 4: Have you searched our list archives? > > http://archives.postgresql.org > -- Jim C. Nasby, Sr. Engineering Consultant jnasby@pervasive.com Pervasive Software http://pervasive.com work: 512-231-6117 vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
Adding -general back to the distribution list. Please do a reply-all. Even if you tell psql what encoding to use, if the database is created with the wrong encoding it might not be able to differentiatebetween the two strings. -----Original Message----- From: anakreonmejdi@yahoo.gr [mailto:anakreonmejdi@yahoo.gr] Sent: Sat 3/4/2006 9:46 AM To: Jim Nasby Cc: Subject: Re: [GENERAL] Insert fails when it shouldn't To execute the insert statements I use psql. From the command line the command \encoding writes:ISO_8859_7 which is the right one since characters are greek strings. -- Three words describe our society:homo homini lupus
I use this script to create the database: DROP DATABASE tei; CREATE DATABASE tei WITH ENCODING='iso_8859_7' OWNER=anakreon TEMPLATE=template1 The encoding of the database is iso_8859_7. -- Three words describe our society:homo homini lupus