Re: [SQL] Changing user passwords
> Is it possible to change a user's password without using the pg_passwd > command? > For example, how does a user changes his password from the interactive > monitor? See the following commands: CREATE USER ALTER USER CREATE GROUP ALTER GROUP For example: ALTER USER user1 WITH PASSWORD 'password1'; Please note thatuser passwords cannot be changed in a transaction. ( 7.0.2 ) Laci 1.0
Re: [SQL] Changing user passwords
A question about the ALTER USER command I couldn't find in the standard documentation: - How do I assign an 'undefined' password? ALTER USER nobody WITH PASSWORD ""; works, but doesn't assign an undefined password to nobody (like the `pg_passwd nobody` command would do). DROP USER nobody; CREATE USER nobody; doesn't do the job either. And why does the ALTER USER command operates on pg_shadow instead of pg_passwd? (or is this configurable?) rob. - "Nagy Laszlo Zsolt , KLTE TTK pm1" wrote: > > > Is it possible to change a user's password without using the pg_passwd > > command? > > For example, how does a user changes his password from the interactive > > monitor? > > See the following commands: > > CREATE USER > ALTER USER > CREATE GROUP > ALTER GROUP > > For example: > > ALTER USER user1 WITH PASSWORD 'password1'; > > Please note thatuser passwords cannot be changed in a transaction. ( 7.0.2 > ) > > Laci 1.0 -- Rob van der Leek E-mail: [EMAIL PROTECTED] FROG Navigation Systems b.v. Cartesiusweg 120 3534 BD Utrecht Tel. 030-2440550 Fax. 030-2440700 http://www.frog.nl
[SQL] returning a recordset with pl/pgsql
Hi everyone, I'm using postgresql with php4. I wanted to write a stored procedure to return a recordset (using pl/pgsql). From what I've read from the documentation it seems like you can only returned postgres types like strings, booleans, etc. Is it possible to return recordsets using pl/pgsql? If so, how do I declare the return type in the declaration? Thanks in advance. Cheers, Keith.
Re: [SQL] Changing user passwords
Rob van der Leek wrote: > A question about the ALTER USER command I couldn't find in the > standard documentation: > > - How do I assign an 'undefined' password? > > ALTER USER nobody WITH PASSWORD ""; > > works, but doesn't assign an undefined password to nobody (like the > `pg_passwd nobody` command would do). > > DROP USER nobody; CREATE USER nobody; > > doesn't do the job either. > > There is no space between: eg; createuser whoever; dropuser whoever; - Bernie begin:vcard n:Huang;Bernie tel;fax:(604)664-9195 tel;work:(604)664-9172 x-mozilla-html:TRUE org:Environment Canada;Standards and Technology Services adr:;;700-1200 West 73 Ave.;Vancouver;BC;V6P 6H9;Canada version:2.1 email;internet:[EMAIL PROTECTED] title:Programmer x-mozilla-cpt:;0 fn:Bernie Huang end:vcard
Re: [SQL] Functions too slow, even with iscachable?
On Mon, Aug 07, 2000 at 10:58:27AM -0400, Tom Lane wrote: > (I assume the lack of "survey_id =" here is just a cut-and-paste error?) Yup. :) > I think what you're getting bit by is that the optimizer doesn't > recognize "var = function(param)" as being a potential indexscan clause. > Does EXPLAIN show that the first query is producing an indexscan plan? It did. I'll try to make up a reproducible test case, if you need it. > I have not tried it, but I think you could get around this problem in > plpgsql, along the lines of > tmp1 = ticket2name($1); > tmp2 = ticket2survey_id($1); > SELECT passwd FROM ticket WHERE name = tmp1 AND survey_id = tmp2; > since the tmp vars will look like params to the optimizer and "var = param" > is indexable. Yup, it did work. Thanks! > Looks like we need to teach the optimizer that expressions involving > params can be treated like simple params for the purposes of > optimization. That'll be good. Anything to speed up the stored procedures are good: encourage people to put logic processing into the RDBMS where it should be.
