So finally the solution is that i should have a PL/SQL bloch which will
retrive the columns at runtime and depends on to that i need to find out the
duplicate rows and delete them, is that so?
-- 
Thanks and Regards,
Niraj Singh Parihar



On Tue, Mar 30, 2010 at 1:28 PM, sonty <[email protected]> wrote:

> Hi Andrej,
>
> Thanks for your response, I had a misunderstanding of count funtion.
>
> Regards,
> Sonty
>
> On Mar 30, 10:44 am, Andrej Hopko <[email protected]> wrote:
> > Hi Sonty,
> >
> >      in other discussion in here we already discus the COUNT(*) and the
> > fact, that oracle this way counts ROWIDs, not a real data (and ROWID is
> > unique - at leas for table, possibly for higher volume of data)
> >
> > Andrej Hopko
> >
> > On 30. 3. 2010 6:58, sonty wrote:
> >
> > > Hi David,
> >
> > > Just a small question, "select count(*) from any_table" does mean
> > > "select count(*) from any_table group by *" (the later is offcourse
> > > wrong), Isn't it??
> >
> > > BR
> > > Sonty
> >
> > > On Mar 29, 6:25 pm, Andrej Hopko<[email protected]>  wrote:
> >
> > >> As David said - it seems lazy when you don't want to write so much
> > >> columns (they also may change)
> > >>       I say - where am I lazy in PL/SQL, there is place for some
> thinking
> > >> and EXECUTE IMMEDIATE (powerful thing, but thinking is quite
> necessary)
> >
> > >> so you can query to USER_COLUMNS for column names, and build a
> VARCHAR2
> > >> variable with your query inside (example of such query is already in
> > >> previous mail)
> > >>       just look for some examples of EXECUTE IMMEDIATE with clause
> INTO
> > >> or clause USING
> >
> > >> good luck
> >
> > >> Andrej Hopko
> >
> > >> On 29. 3. 2010 14:30, ddf wrote:
> >
> > >>> On Mar 29, 5:00 am, Niraj Parihar<[email protected]>
> > >>> wrote:
> >
> > >>>> But here the column name is given and if i have 100 columns then in
> that
> > >>>> case it is not fieaseble to use this kind on query as i need the
> query which
> > >>>> works independent of the columns defined in the table.
> > >>>> --
> > >>>> Thanks and Regards,
> > >>>> Niraj Singh Parihar.
> >
> > >>>> On Mon, Mar 29, 2010 at 1:06 PM, rajesh dhumal<
> [email protected]>wrote:
> >
> > >>>>> Try this
> >
> > >>>>> SELECT COL_TO_CHECK, COUNT(COL_TO_CHECK)
> > >>>>> FROM TABLE_NAME
> > >>>>> GROUP BY COL_TO_CHECK
> > >>>>> HAVING COUNT(COL_TO_CHECK)>    1
> >
> > >>>>> delete from test a
> > >>>>> where rowid<>    ( select max(rowid)
> > >>>>> from test b
> > >>>>> where a.sno = b.sno
> > >>>>> and a.sname = b.sname )
> >
> > >>>>> This querry will find duplicate rows and delete those.
> >
> > >>>>> On Mon, Mar 29, 2010 at 11:55 AM, Niraj Parihar<
> > >>>>> [email protected]>    wrote:
> >
> > >>>>>> Can any body help me on a query where i have to find out the
> number of
> > >>>>>> duplicate records in a table independent of the columns defined in
> a table.
> > >>>>>> --
> > >>>>>> Thanks and Regards,
> > >>>>>> Niraj Singh Parihar
> >
> > >>>>>>    --
> > >>>>>> You received this message because you are subscribed to the Google
> > >>>>>> Groups "Oracle PL/SQL" group.
> > >>>>>> To post to this group, send email to
> [email protected]
> > >>>>>> To unsubscribe from this group, send email to
> > >>>>>> [email protected]
> > >>>>>> For more options, visit this group at
> > >>>>>>http://groups.google.com/group/Oracle-PLSQL?hl=en
> >
> > >>>>>> To unsubscribe from this group, send email to oracle-plsql+
> > >>>>>> unsubscribegooglegroups.com or reply to this email with the words
> "REMOVE
> > >>>>>> ME" as the subject.
> >
> > >>>>> --
> > >>>>> Rajesh V.Dhumal- Hide quoted text -
> >
> > >>>> - Show quoted text -
> >
> > >>> If you won't specify the columns for the table how do you propose to
> > >>> find 'duplicate' records?  I believe I understand your desire (find
> > >>> completely duplicate rows in the table) however you cannot do that
> > >>> without specifying column names as group by doesn't accept * as an
> > >>> operand.  You don't 'need' to find such a query, you simply want to
> > >>> find one to avoid typing which appears lazy to me.
> >
> > >>> David Fitzjarrell- Hide quoted text -
> >
> > >> - Show quoted text -
> >
> >
>
> --
> You received this message because you are subscribed to the Google
> Groups "Oracle PL/SQL" group.
> To post to this group, send email to [email protected]
> To unsubscribe from this group, send email to
> [email protected]
> For more options, visit this group at
> http://groups.google.com/group/Oracle-PLSQL?hl=en
>
> To unsubscribe from this group, send email to oracle-plsql+
> unsubscribegooglegroups.com or reply to this email with the words "REMOVE
> ME" as the subject.
>

-- 
You received this message because you are subscribed to the Google
Groups "Oracle PL/SQL" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/Oracle-PLSQL?hl=en

To unsubscribe from this group, send email to 
oracle-plsql+unsubscribegooglegroups.com or reply to this email with the words 
"REMOVE ME" as the subject.

Reply via email to