It's a valid discussion here (although better on -advocacy), because it helps
me have the right facts to present to clients about whether they should stay
with a legacy database in MySQL vs upgrading to a modern PostgreSQL.
For all its flaws, MySQL is catching on quick and has a very active
com
> "John" == John Meyer <[EMAIL PROTECTED]> writes:
John> I'd say fine, but why discuss the flaws of MySQL on a PostgreSQL list?
John> If you want to correct it, why not put that flaw on a MySQL list. And
John> yes, I agree, there is a difference between pointing out a legitimate
John> flaw an
I'd say fine, but why discuss the flaws of MySQL on a PostgreSQL list?
If you want to correct it, why not put that flaw on a MySQL list. And
yes, I agree, there is a difference between pointing out a legitimate
flaw and simply bashing for bashing's sake.
Joshua D. Drake wrote:
> John Meyer wrote:
John Meyer wrote:
> What I think bothers me is this whole concept that if PostgreSQL is to
> flourish, MySQL has to be beaten down. Folks, both products are free,
> both can be used in the same shop (maybe not on the same computer if
> your running them in production). Putting down MySQL will not
What I think bothers me is this whole concept that if PostgreSQL is to
flourish, MySQL has to be beaten down. Folks, both products are free,
both can be used in the same shop (maybe not on the same computer if
your running them in production). Putting down MySQL will not make
PostgreSQL any bette
"Michael Nolan" <[EMAIL PROTECTED]> writes:
> select count(*) from memmast where memid in (select plr_rated_memid from
> tnmt_plr where plr_eventid in ('200607163681');
> This query takes about a second on PostgreSQL but takes OVER SEVEN MINUTES
> on MySQL!
Yeah, and we probably would have sucked
Ron Johnson <[EMAIL PROTECTED]> writes:
> On 01/20/07 16:52, Michael Nolan wrote:
>> select plr_rated_memid from tnmt_plr where plr_eventid in ('200607163681');
> Is this query created by an application? I.e, there might be a list
> of PLR_EVENTIDs?
> If so, I understand why it is like it is. O
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
It's exactly what I would do...
On 01/20/07 17:09, Michael Nolan wrote:
> This is a generated query in a web form where there could be a series of 12
> digit event IDs input by the user, hence using the 'in' form. This is
> slightly lazy programming
This is a generated query in a web form where there could be a series of 12
digit event IDs input by the user, hence using the 'in' form. This is
slightly lazy programming on my part, but it makes little difference in
either PostgreSQL or MySQL whether I use = or 'in'.
--
Mike Nolan
On 1/20/07,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 01/20/07 16:52, Michael Nolan wrote:
> I have a MySQL table on our public website that is populated from a similar
> table on our internal site, which runs PostgreSQL.
>
> Recently I was trying to enhance one of our website queries and ran across
>
I have a MySQL table on our public website that is populated from a similar
table on our internal site, which runs PostgreSQL.
Recently I was trying to enhance one of our website queries and ran across
an interesting phenomenon:
The following query runs very quickly in both PostgreSQL (8.1.3) an
From: "Kelly Burkhart" <[EMAIL PROTECTED]>
> On 1/20/07, Shoaib Mir <[EMAIL PROTECTED]> wrote:
> > Should help --> ALTER TABLE tablename ALTER columname TYPE text;
>
> I was looking for a way to alter a column from varchar(n) to text
> without using the alter command and consequently touching ever
I haven't used it this way, anyone else who did might be able to comment on
it. Why will you not like to use the ALTER table command? I think a text
should be use in case you don't know the limit for characters (much faster
too in that case I guess) in a column but if you know the limits then you
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 01/20/07 10:52, Bruce Momjian wrote:
> Ron Johnson wrote:
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>>
>> On 01/19/07 18:21, Bruce Momjian wrote:
>>> Sorry, I know of no way to get a status bar that shows how far the an
>>> INSERT or COPY
On 1/20/07, Shoaib Mir <[EMAIL PROTECTED]> wrote:
Should help --> ALTER TABLE tablename ALTER columname TYPE text;
I was looking for a way to alter a column from varchar(n) to text
without using the alter command and consequently touching every single
row. Below is sql which seems to work, but
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Ron Johnson wrote:
>> How hard *would* it be to patch the \copy code to print out a "#"
>> every N number of input records?
> Yea, probably, but considering they probably want a graphic progress
> bar, the big question is what API would you use to allow
Ron Johnson wrote:
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> On 01/19/07 18:21, Bruce Momjian wrote:
> > Sorry, I know of no way to get a status bar that shows how far the an
> > INSERT or COPY has progressed. People have asked for it, but no one has
> > any idea how to implement it.
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 01/19/07 18:21, Bruce Momjian wrote:
> Sorry, I know of no way to get a status bar that shows how far the an
> INSERT or COPY has progressed. People have asked for it, but no one has
> any idea how to implement it.
How hard *would* it be to patch
Bruce Momjian schrieb:
Sorry, I know of no way to get a status bar that shows how far the an
INSERT or COPY has progressed. People have asked for it, but no one has
any idea how to implement it.
Well but thats aparently not the case here anyway. Fileupload is
something different from looking h
Richard P. Welty wrote:
> running 8.1 on a fedora core 5 linux box, up to date so far as
> i know.
>
> this page:
>
> http://www.postgresql.org/docs/8.1/interactive/manage-ag-templatedbs.html
>
> says the following:
>
> Note: template1 and template0 do not have any special status beyond
>
I think you can use the binary string functions
(http://www.postgresql.org/docs/8.2/interactive/functions-binarystring.html) to
achieve the same as ReadText and UpdateText.
/Mikael
> -Original Message-
> From: Benedict Faria [mailto:[EMAIL PROTECTED]
> Sent: den 19 januari 2007 14:02
>
21 matches
Mail list logo
