This is something done a lot based on my experience. Here
we're okay for the most part: We mostly need to extract
spreadsheet data row by row and either use it as Oracle query
arguments or add to it with row by row queries.
You can read row by row like that with Spreadsheet::ParseExcel
and create new spreadsheets with Spreadsheet::WriteExcel. The
Spreadsheet::ParseExcel author posted the other day that he's
working on a DBI interface using the same base (OLE::Storage_Lite).
Both of these modules have saved me lots of time.
Since most of what I'd need is based on joins with our Oracle
data, I'm not sure a DBI module would help if I need to do anything
more complicated. It would still be two data sources and I can't
join across different DB's that I know of (maybe I'm also missing
something there). But using Spreadsheet::ParseExcel and DBI, I can
read the data and load it into Oracle tables that I can then join
with existing Oracle data. So as with everything Perl, There's
More Than One Way To Do It. 8-)
Another side of this whole sort of thing: We also periodically
have a need to run server side Perl apps. that pull data out of
Access. For that, I'd guess ODBC would be the best choice --
I've done this job on NT using Perl ODBC and DBI modules (read
using ODBC, write using DBI). But again, there is not a full set
of hooks to ODBC on all UNIX platforms. That may just be a matter
of finding a UNIX based ODBC driver. I looked into this for only
a very short time since it's a very small part of everything else
I do.
If anyone has knowledge here and free time (I don't have enough of
either!) it would be great to have a consolidated web resource that
explained the various ways to integrate Microsoft world data with
UNIX based databases. I get by here but I'm sure there are many
things I'm missing ...
----
Steve Sapovits
Global Sports Interactive
Work Email: [EMAIL PROTECTED]
Home Email: [EMAIL PROTECTED]
Work Phone: 610-491-7087
Cell: 610-574-7706
Pager: 877-239-4003
> -----Original Message-----
> From: Sterin, Ilya [SMTP:[EMAIL PROTECTED]]
> Sent: Wednesday, May 02, 2001 8:56 AM
> To: 'Simon Oliver '
> Cc: '''[EMAIL PROTECTED]' ' '
> Subject: RE: Extract data from MS Excel Spreadsheets. Can it be done?
>
> Also I would think that this is a big enough issue to address, since most
> of
> the servers are not NT and most of the clients are win32, therefore if
> clients are working on excel which is later uploaded to the server (Unix),
> it should be processed on the server side, rather than having users save
> as
> CSV.
>
> Ilya Sterin
>
> -----Original Message-----
> From: Simon Oliver
> Cc: ''[EMAIL PROTECTED]' '
> Sent: 05/02/2001 2:27 AM
> Subject: Re: Extract data from MS Excel Spreadsheets. Can it be done?
>
> Steve Sapovits wrote:
>
> > If I'm wrong about this and someone's used either DBD::ADO or
> > DBD::ODBC on Sun/Solaris, I'd like to hear how you did it.
> One can't use DBD::ADO because that uses Win32::OLE which needs a Win32
> Operating System.
>
> --
> Simon Oliver
