[Jprogramming] (no subject)

[John Baker]

If you are fetching fixed width fields (not VARCHAR VARBINARY ) or other
variable length types use ddfch.  It is much faster when your data is
all fixed width.

On 3/14/07, eiwco at tiscali.it <eiwco at tiscali.it> wrote:
>
> Hi everybody , I'm a J newbie and I' m really amazed by its
> mathematical
> attitudes, so I started using it to analyze data from some databases
> using odbc extraction code like:
>
> ch=: ddcon 'dsn=firebird_pivot;uid=SYSDBA;pwd=masterkey'
> sh =. 'select * from PIVOT_VENDITE_SUM' ddsel ch
> r=. ddfet sh,_1
> r2=. ddfch sh, _1
> ...
>
> I've found that:
>
> a) with Firebird database engine ddfet is very slow and causes J to
> stop working if the number of records I read is too great (over 5000).
> b) with Firebird ddfch hasn't got such problem but fields seem broken
> (for example in a column which contains
> 2005
> 2006
> 2007
> 2008
> ....
>
> the query gives
>
> 2002
> 0020
> 0720
> ...
> Does it depends by using varchar fields ?
>
> c) with Mysql or Oracle ddfet doesn't seem to have the above problems,
> but queries are very slow if compared (for example) to the same queries
> executed with perl-dbi (dbd-odbc)
>
> Is there anybody who can suggest me anything ?
>
> Best regards
>
> Alessio Costantini
>
>
> Naviga e telefona senza limiti con Tiscali
> Scopri le promozioni Tiscali adsl: navighi e telefoni senza canone Telecom
>
> http://abbonati.tiscali.it/adsl/
>
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>