Re: [ADodb-general] Re: adodb execute method

[Darius <da...@m2...>]

Hi Alfredo,

my code looks like this:

$sql =3D "Select * from table order by x"
$sqlCount =3D "Select count(*) from table" (without order and limit)

But it'is not good for performance.

Best regards,
Darius


  ----- Original Message -----=20
  From: Alfredo Yong=20
  To: Darius=20
  Cc: ado...@li...=20
  Sent: 2006 m. kovo 2 d. 14:28
  Subject: Re: [ADodb-general] Re: adodb execute method


  Hi Darious,

  I use a code like this:

  $SQL =3D "select * from table1";
  $SQLCount =3D "select count(*) from ($SQL)";

  Then I use $SQLCount to know the row count, and $SQL with a =
SelectLimit

  But Im afraid the cost of this two queries is more than double the =
original ($SQL) query

  Any better idea?

  greetings,
  =20
  On 3/2/06, Darius <da...@m2...> wrote:=20
    Hi all,

    we have also the same problem.
    With selectLimit(..) I retrieve only needed rows. But there need to =
know count of all rows.=20
    In MySQL I can do this with SQL_CALC_FOUND_ROWS. What is solution in =
AdoDB ?

    Best regards,
    Darius


    Marcin Szkudlarek wrote:

    > Thanks for reply. You are right - I want to implement paging but =
it"s=20
    > not that simple. I want to select 100 rows from 50k table but I =
don"t
    > know how much rows I have to fetch to get the 100 .. it could be =
in
    > the WORST case the last 100 from the whole 50k table. I"t because =
I"m=20
    > checking access rights to every row in php and then diplaying it =
or
    > not. So your example with selecting only first 100 rows from the =
table
    > woudn"t work here. If there isn"t a way to fetch the result row by =
row=20
    > I"ll have to reimplement some ugly code.

    > You have just described the classic poorly designed application.  =
;-)

    >If there"s something in the data that you can look at and determind
    >access rights in PHP, then you should move that logic to your SQL =
query
    >and select only the rows you"re interested in.

    >
    > Best regards,
    >
    > Marcin Szkudlarek
    >
    > On Fri, 24 Jun 2005, Emil Zegers wrote:=20
    >
    >> What do you exactly mean by "fetching the rows"?
    >>
    >> If you want to do a row count then your query could be something =
like
    >> "SELECT COUNT(*) FROM TABLENAME"=20
    >>
    >> If you want to get only a certain amount of rows you can do =
something
    >> like "SELECT TOP 100 * FROM TABLENAME".
    >>
    >> Also take a look at paging options.
    >>=20
    >> If you do want to get the data in a recordset for manipulation =
you
    >> really have to fetch it.
    >>
    >> You can limit the data (and thus the query time) by selecting =
only the
    >> columns needed.=20
    >>
    >> Regards,
    >>
    >> Emil
    >>
    >> -----Original Message-----
    >> From: adodb-general-admin@li...
    >> [mailto:adodb-general-admin@li ...] On Behalf Of Marcin
    >> Szkudlarek
    >> Sent: vrijdag 24 juni 2005 11:49
    >> To: adodb-general@li...
    >> Subject: [ADodb-general] adodb execute method
    >>
    >> Hi!
    >>
    >> I"m using ado 4.62 with oracle 8.17. My question:
    >> is there a way to execute select query without getting all the =
results?
    >> I have a table with about 50k records and calling only the =
execute=20
    >> method takes about 1,8s. What I want to achieve is fetching the =
rows
    >> without having to get all the data at once.
    >>
    >> Thanks.
    >>
    >> Marcin Szkudlarek



  --=20
  Alfredo Yong
  Sistemas web