FreeTDS Development Group

[robin nayathodan <rnayathodan AT yahoo.com>]

Here is my question...
I have a linux distribution which connects to the sql
server through freetds.
I have a table which has a unique identifier field
type
say pop table
uid unique identifier(may be sql 7.0 has a new
feature)

whenever i query this table i get some records which
is not similar to the table values?
the source program is in perl (using dbi)

Anyway thanks for an early reply.....

Awaiting for a reply.....
Robin....

--- TDS Development Group digest
<freetds AT franklin.oit.unc.edu> wrote:
> FREETDS Digest for Sunday, October 01, 2000.
> 
> 1. Can anybody help me out with this unique id
> problem
> 2. Re: Can anybody help me out with this unique id
> problem
> 3. Re: Delay closing resultsets
> 4. Re: Delay closing resultsets
> 
>
----------------------------------------------------------------------
> 
> Subject: Can anybody help me out with this unique id
> problem
> From: robin nayathodan <rnayathodan AT yahoo.com>
> Date: Sun, 1 Oct 2000 01:35:27 -0700 (PDT)
> X-Message-Number: 1
> 
> Hi All,
> Anybody please help me this unique identifier field
> Whenever i query this field i get the values
> different
> from the table which i cannot use later on for
> further
> queries depending on this field.
> If anybody knows any soln for this problem please
> mail
> me at rnayathodan AT yahoo.com
> Waiting for an early reply from someone,
> Robin.
> 
> 
> > From: "Brandon M. Reynolds" <bmr AT comtime.com>
> > Date: Sat, 30 Sep 2000 12:01:16 -0400
> > X-Message-Number: 1
> > 
> > Fernando Lozano wrote:
> > > The only workaround I got so far is to call
> > > Statement.setMaxRows(), but this way the user
> > won't be able to see
> > > all results, and this is not desirable.
> > > 
> > > Any hints will be very apreciated.
> > 
> > I don't know whether or not there is bug here, but
> I
> > wrote a search
> > engine and I had to address a similar issue. 
> Since
> > I'm using HTML
> > to display my results, I just display a certain
> > number of results
> > at one time.  I use 'SELECT TOP n name...' where n
> > keeps growing as
> > the user proceeds to look at more results.  I just
> > skip the rows that
> > they have already seen by fetching them but
> tossing
> > them.  Of course,
> > the pages will get slightly slower to load as the
> > user proceeds.  But
> > because my search is weighted, I'm assuming
> they're
> > not going to want
> > to look at any page beyond 2 or 3 most of the
> time. 
> > Normally they will
> > refine they search criteria.
> > 
> > If you are using a more windows-type setting, then
> > maybe you could load
> > more results as the scrollbar advances?
> > 
> > I came up with this technique just on my own, so
> I'm
> > definitely willing
> > to hear if anyone else has a good solution.  But
> > remember the result
> > rows
> > I'm dealing with are not able to be indexed in any
> > way because the
> > weight
> > of each result is the sorting criterion and that
> has
> > to be calculated
> > based
> > on the search query, i.e. I can't gain any real
> > improvement by adding
> > 'WHERE SUM(weight) <= target_weight'.
> > 
> > -- 
> > Brandon Reynolds                  Phone: (330)
> > 644-3059
> > Systems Engineer                    Fax: (330)
> > 644-8110
> > Commercial Timesharing Inc.      E-mail:
> > bmr AT comtime.com
> > 
> > 
> > 
> > ---
> > 
> > END OF DIGEST
> > 
> > ---
> > You are currently subscribed to freetds as:
> > [rnayathodan AT yahoo.com]
> > To unsubscribe, forward this message to
> $subst('Email.Unsub')
> 
> __________________________________________________
> Do You Yahoo!?
> Yahoo! Photos - 35mm Quality Prints, Now Get 15
> Free!
> http://photos.yahoo.com/
> 
>
----------------------------------------------------------------------
> 
> Subject: Re: Can anybody help me out with this
> unique id problem
> From: Bob Kline <bkline AT rksystems.com>
> Date: Sun, 1 Oct 2000 09:18:42 -0400 (EDT)
> X-Message-Number: 2
> 
> On Sun, 1 Oct 2000, robin nayathodan wrote:
> 
> > Anybody please help me this unique identifier
> field Whenever i query
> > this field i get the values different from the
> table which i cannot
> > use later on for further queries depending on this
> field.
> > 
> [Lengthy quote of apparently unrelated messages
> omitted.] 
> 
> I'm afraid I don't understand your question.  Could
> you elaborate with
> more information, please?
> 
> -- 
> Bob Kline
> mailto:bkline AT rksystems.com
> http://www.rksystems.com
> 
> 
>
----------------------------------------------------------------------
> 
> Subject: Re: Delay closing resultsets
> From: Stephanie Lewis <slewis AT thinweb.com>
> Date: Sun, 01 Oct 2000 11:38:15 -0300
> X-Message-Number: 3
> 
>     When you query the database, the result set is
> returned in a DataInputStream as
> packets that are processed as needed (ie: calls to
> next()). If you close the result
> set before the end of the data is returned, the
> DataInputStream may still have data
> from the query left in it. The driver uses
> connection pooling to improve the
> efficiency of the driver. So , after you've called
> close() on the result set, the
> driver wants to put the this connection (really just
> a socket right), back into the
> pool to be used by another query. Now if you perform
> another query using the same
> socket, when you start reading the results, you are
> actually going to start reading
> what was left from the last query before you get to
> the data you want. It is for this
> reason that the DataInputStream is "flushed" when a
> result set is closed. This is what
> you've observed as :   "it looks like the freetds
> jdbc driver continues to fetch all
> rows when I close it" .
>     That's it in a nutshell. I would like for it not
> to do this as well, but can't
> think of a way yet to solve it without getting rid
> of the pooling and using a new
> connection every time. Then you could just close the
> socket on close(). I think this
> would be just trading one performance problem for
> another though. Establishing a
> connection is costly, hence the pooling. I haven't
> tried it yet though.
>     I may have missed some glaring issue, so if I've
> misunderstood the process, input
> from anyone interested would be appreciated.
> 
> Cheers,
>         Stephanie
> 
> 
> 
> 
> Fernando Lozano wrote:
> 
> > Hi there!
> >
> > I am using the freetds jdbc driver for an
> application that queries
> 
=== message truncated ===


__________________________________________________
Do You Yahoo!?
Yahoo! Photos - 35mm Quality Prints, Now Get 15 Free!
http://photos.yahoo.com/