Re: [SQL] cursor and for update

2006-03-28 Thread Maciej Piekielniak
Hello Wiebe, Tuesday, March 28, 2006, 2:42:20 AM, you wrote: >> WC> Fist, what's the error you get? ERROR: missing ".." at end of SQL expression CONTEXT: compile of PL/pgSQL function "uporzadkuj_reguly" near line 7 WC> Second, what does "rekord" do? You don't seem to use it. Yes, i don't nee

Re: [SQL] cursor and for update

2006-03-28 Thread Wiebe Cazemier
On 03/28/06 11:13, Maciej Piekielniak wrote: >create or replace function uporzadkuj_reguly(text,text) RETURNS VOID AS >' >DECLARE > licznik integer:=1; > > reguly CURSOR FOR SELECT * from firewall ORDER BY id_firewall WHERE > tabela=$1 and lancuch=$2 for UPDATE; >BEGIN > for i in reguly LOOP > U

[SQL] Update question

2006-03-28 Thread Fay Du
Hi all:   I have a table like this:   Table cost Id  edge_id    cost 1   30 101.4 2   30  null 3   40  500.2 4   40   null 5    45  300.7 6    45   null   I want to set cost value with same edge_id by same

Re: [SQL] Update question

2006-03-28 Thread Markus Schaber
Hi, Fay, Fay Du wrote: > Table cost > > Id edge_idcost > 1 30 101.4 > 2 30 null > 3 40 500.2 > 4 40 null > 545 300.7 > 645 null > I want to set cost value with same edge_id by same value. The exis

Re: [SQL] Update question

2006-03-28 Thread Gábriel Ákos
Markus Schaber wrote: Hi, Fay, Fay Du wrote: Table cost Id edge_idcost 1 30 101.4 2 30 null 3 40 500.2 4 40 null 545 300.7 645 null I want to set cost value with same edge_id by same value. T

Re: [SQL] Update question

2006-03-28 Thread Fay Du
Markus and Gábriel: Thank you very much for the help. Yes, the table is redundant. What I am trying to do is using pgdijstra module from postGIS which is on the top of postgreSQL. And the module asks for entire network is directional or non-directional. But my network part is directional and other

Re: [SQL] Bitfields always atomic? Other way to store attributes?

2006-03-28 Thread TJ O'Donnell
> If I have two threads modifying the same "bit" field: > thread1=> update table set bf=bf | '01000' > thread2=> update table set bf=bf | '1' > Will this operation always be safe (e.g. result in bf='11000')? Or Won't this always result in bf='11xxx', depending on the original values o

Re: [SQL] Bitfields always atomic? Other way to store attributes?

2006-03-28 Thread Owen Jacobson
TJ O'Donnell wrote: > > If I have two threads modifying the same "bit" field: > > thread1=> update table set bf=bf | '01000' > > thread2=> update table set bf=bf | '1' > > Will this operation always be safe (e.g. result in bf='11000')? Or > > Won't this always result in bf='11xxx', d