Hi
If I link to a CSV ( comma seperated value) file does access think of this 
as a table with records and behave the same way.
I was wondering what would happen if you had say 5000 records in a text file 
and each record had 50 fields Would it be slower than if the 5000 records 
were in an Access Table. Can anybody give me some info on problems or limits 
on this type of action.
An example of potential problem maybe and could be if the computer over 
writes the text file with updated data while the database is reading records 
from it.
Any comments most welcome
Steve - From a land down under

On Jan 4, 9:50 am, Steve <St...@discussions.microsoft.com> wrote:
> Hi
> If I link to a CSV ( comma seperated value) file does access think of this
> as a table with records and behave the same way.
> I was wondering what would happen if you had say 5000 records in a text file
> and each record had 50 fields Would it be slower than if the 5000 records
> were in an Access Table. Can anybody give me some info on problems or limits
> on this type of action.
> An example of potential problem maybe and could be if the computer over
> writes the text file with updated data while the database is reading records
> from it.
> Any comments most welcome
> Steve - From a land down under

Access makes the CSV file look like a table but it's not that way
under the covers and doesn't offer anything like the performance of a
'real' table. You can get into all sorts of file locking issues,
format problems and other maintenance nasties - I used this method for
a few months on a CSV directory of names/addresses for a firm that
totaled 40,000 rows and it was a nightmare.

Instead, I imported the CSV and handled it from within Access, which
turned out to be much more stable. Is there any reason you want to
stay within the CSV?

-- James

the usual thing of course I have a customer with an accounts package that 
will only export names and address in CSV Text and the customer wants the 
address to use in a Access Quote package that he has created.
So in your opinion its better to update and append files from a CSV to an 
Access back end table and let the front ends link to the access back end 
table.
Steve - from a land down under

"Minton M" wrote:

> On Jan 4, 9:50 am, Steve <St...@discussions.microsoft.com> wrote:
> > Hi
> > If I link to a CSV ( comma seperated value) file does access think of this
> > as a table with records and behave the same way.
> > I was wondering what would happen if you had say 5000 records in a text file
> > and each record had 50 fields Would it be slower than if the 5000 records
> > were in an Access Table. Can anybody give me some info on problems or limits
> > on this type of action.
> > An example of potential problem maybe and could be if the computer over
> > writes the text file with updated data while the database is reading records
> > from it.
> > Any comments most welcome
> > Steve - From a land down under
> 
> Access makes the CSV file look like a table but it's not that way
> under the covers and doesn't offer anything like the performance of a
> 'real' table. You can get into all sorts of file locking issues,
> format problems and other maintenance nasties - I used this method for
> a few months on a CSV directory of names/addresses for a firm that
> totaled 40,000 rows and it was a nightmare.
> 
> Instead, I imported the CSV and handled it from within Access, which
> turned out to be much more stable. Is there any reason you want to
> stay within the CSV?
> 
> -- James
> 

How you do this will be determined by how much activity there is with the csv 
file.
You can overcome formatting issues by creating in import spec.  You do that 
be manually creating the link (Files, Get External Date, etc).  When you get 
to the imort wizard, click Advanced.  You will be able to define your field 
names and data types there.  Then click Save As and give it a name.  Use the 
name in the TransferText.  I would do this whether I were going to link or 
import the file.
-- 
Dave Hargis, Microsoft Access MVP


"Steve" wrote:

> the usual thing of course I have a customer with an accounts package that 
> will only export names and address in CSV Text and the customer wants the 
> address to use in a Access Quote package that he has created.
> So in your opinion its better to update and append files from a CSV to an 
> Access back end table and let the front ends link to the access back end 
> table.
> Steve - from a land down under
> 
> "Minton M" wrote:
> 
> > On Jan 4, 9:50 am, Steve <St...@discussions.microsoft.com> wrote:
> > > Hi
> > > If I link to a CSV ( comma seperated value) file does access think of this
> > > as a table with records and behave the same way.
> > > I was wondering what would happen if you had say 5000 records in a text file
> > > and each record had 50 fields Would it be slower than if the 5000 records
> > > were in an Access Table. Can anybody give me some info on problems or limits
> > > on this type of action.
> > > An example of potential problem maybe and could be if the computer over
> > > writes the text file with updated data while the database is reading records
> > > from it.
> > > Any comments most welcome
> > > Steve - From a land down under
> > 
> > Access makes the CSV file look like a table but it's not that way
> > under the covers and doesn't offer anything like the performance of a
> > 'real' table. You can get into all sorts of file locking issues,
> > format problems and other maintenance nasties - I used this method for
> > a few months on a CSV directory of names/addresses for a firm that
> > totaled 40,000 rows and it was a nightmare.
> > 
> > Instead, I imported the CSV and handled it from within Access, which
> > turned out to be much more stable. Is there any reason you want to
> > stay within the CSV?
> > 
> > -- James
> >