Hi everyone,

I am working on some tool. Now I want to store records in some
file.

1) Are there some examples I can look at ?
2) Is it better to write fast data access routines (and creating the file in code) instead of using a ODBC connection to some database ?

Thanks guys
Posted on 2002-03-04 16:52:36 by Jurgen
it depends on the complexity of actions you wish to perform on your data :)

do you need options that a database gives (e.g. searching for connections between data, replication, multiple clients, client -server,...) or will a flat sequential file do?

As for databases you can use ADO, it can ease your database programming a lot usually, but it won't necessarily be easier to use from asm then ODBC (given that it's COM).

So the question is: What kind of operations do you need to perform on your data and in what context?
Posted on 2002-03-04 16:57:52 by Hiroshimator
So the question is: What kind of operations do you need to perform on your data and in what context?

Hiroshimator,

The file is a local file. I would like to use some indexing system to read the file out. Because I want them in alphabetical order.
Posted on 2002-03-04 17:03:44 by Jurgen
will it be a lot of data?
will it be largely read-only with little writes?
is that the only operation you wish to perform on it?
Posted on 2002-03-04 17:25:47 by Hiroshimator
will it be a lot of data?
No, up to 100 records max.

1 record contains :
4 x a filename
1 x a text of 200 bytes

will it be largely read-only with little writes?
yes,
manually updating 1 by 1,
and reading the all the data out for generating .html from it.
Posted on 2002-03-04 17:39:51 by Jurgen
for the simple usage as you depict it I'd use a random access flat file.
Since it will be small, of specific size and limited in functionality it won't be hard to map or handle the records.

a memorymap will probably work best.
Posted on 2002-03-04 17:51:02 by Hiroshimator
OK,

Thanks for the advice Hiroshimator.
Posted on 2002-03-04 18:02:47 by Jurgen