Confusing question or should I address it some where else?

I posted a question here on 12/6/05 about importing data from .csv file and
no one addressed it.
Here is the link
http://www.microsoft.com/technet/co...ogramming&fltr=
And the title is
Subject: import data properly from csv file.
Should I post it some where else or the question was not clear?
Please let me know if there is a better design for this.
I would really appreciate it.
ThanksActually, this is the direct link to your original quiestion:
http://msdn.microsoft.com/newsgroup...86f3&sloc=en-us
I don't really see any need for a global temporary table here. Why can't you
simply query the file directly? You're building stairs in the football field
.
If for some reason you do need a temporary table, create a local one (#name
instead of ##name). Of course local temporary tables are scope-specific, so
you'd have to create it in a top level procedure that starts the whole
process (creates the table, calls a sub procedure to import data, then does
the rest of transformations).
ML
http://milambda.blogspot.com/|||Assuming you know the structure of the resulting table, which would
be likely if you have other code to process it, how about
create table #temptbl (
thisCol thisType,
thatCol thatType
)
insert into #temptbl exec (
'SELECT * into ##temptbl FROM '+
@.linked_server + '...['+@.file + '#' +
@.extension + ']')
-- process as before
drop table #temptbl
If this is all in a procedure, you don't have to explicitly drop
#temptbl, but it's not a bad idea to do so.
Steve Kass
Drew University
sqlster wrote:

>I posted a question here on 12/6/05 about importing data from .csv file and
>no one addressed it.
>Here is the link
>http://www.microsoft.com/technet/co...ogramming&fltr=
>And the title is
>Subject: import data properly from csv file.
>Should I post it some where else or the question was not clear?
>Please let me know if there is a better design for this.
>I would really appreciate it.
>Thanks
>
>|||ML and Steve,
Thank you very much...
"Steve Kass" wrote:

> Assuming you know the structure of the resulting table, which would
> be likely if you have other code to process it, how about
> create table #temptbl (
> thisCol thisType,
> thatCol thatType
> )
> insert into #temptbl exec (
> 'SELECT * into ##temptbl FROM '+
> @.linked_server + '...['+@.file + '#' +
> @.extension + ']')
> -- process as before
> drop table #temptbl
> If this is all in a procedure, you don't have to explicitly drop
> #temptbl, but it's not a bad idea to do so.
> Steve Kass
> Drew University
> sqlster wrote:
>
>