Hi,
   I haveinformation that PostGre will keep directory information of one 
Table. In my case, my system is Multi Company, them we have the same Table 
(Ex Sales.dbf) in more than one folder (directory).

   I understood that each file table had a file descriptor that indicate 
original directory e physical table name. And we can continue to open a dbf 
file in Postgre using directory specification that it translate correctly to 
correct name of postgre Table.

  But I convert it, but to open, it does not accept path on USE command.

  Is that true ?

  Thanks
  Osvaldo L AOki

hi,

PostgreSQL Table are in "Catalog" so for a different Customer you have to 
define a new "Catalog" for each Customer.

that also mean you have to setup a Connection for each Customer to access 
right Table ... don't know if it is possible with PgDBE USE Command.

Osvaldo,

>I haveinformation that PostGre will keep directory information of one 
>Table. In my case, my system is Multi Company, them we have the same Table 
>(Ex Sales.dbf) in more than one folder (directory).

Actually, the Alaska's PGDBE will keep the (assumed) directory information 
for tables that are created manually with DbCreate() or through Upsizing in 
a special table that the PGDBE creates, named "alaska-software.isam.tables".

That table has a column named "unc_name" in which the "original" file name 
is stored. That file name is either based on the active Set Default value 
when DbCreate() is used, or on the "dbf" attribute in the *.upsize file. All 
values in that column must be unique, but it doesn't seem to be used for 
anything, at least as far as I can tell.

>I understood that each file table had a file descriptor that indicate 
>original directory e physical table name. And we can continue to open a dbf 
>file in Postgre using directory specification that it translate correctly to 
>correct name of postgre Table.
>But I convert it, but to open, it does not accept path on USE command.

That might or might not work -- I don't know for sure, as I have never tried 
this. But I also haven't read anywhere in the documentation that this would 
or should work, so I'm quite skeptical that this would be even possible.

For once, you can't have two tables with the exact same name in the same 
schema, so the table names from the separate directories would have to be 
changed to be put into the same SQL schema.

A better approach would be to either create separate schemas for separate 
users or to prefix the name of the (originally identically named) tables 
with the directory name that they came from, replacing any backslash "\" 
with an underscore "_" character.

If you "upsize" your data, you will have to manually change the "*.upsize" 
configuration file to update/correct the table names, I suppose. 

So, how does your original directory structure, your "*.upsize" file, and 
your resulting PGSQL table structure and "alaska-software.isam.tables" look 
like?

If you post that information here, maybe it is possible to determine what 
the best (or easiest) way would be to make it work for you.

Hope that helps,

Andreas

Andreas Gehrs-Pahl
Absolute Software, LLC

phone: (989) 723-9927
email: Andreas@AbsoluteSoftwareLLC.com
web:   http://www.AbsoluteSoftwareLLC.com
[F]:   https://www.facebook.com/AbsoluteSoftwareLLC

Andreas Gehrs-Pahl wrote:

> A better approach would be to either create separate schemas for separate 
> users or to prefix the name of the (originally identically named) tables 

I have been told that the Xbase++ implementation of SQL does not support
schemas. Is this right?

Peter

Peter,

>I have been told that the Xbase++ implementation of SQL does not support
>schemas. Is this right?

Nope. There is no reason why schemas shouldn't be supported with either the 
PGDBE (in ISAM mode, Universal SQL, or Pass-Through SQL) or the ODBCDBE. You 
simply create one connection for each individual schema, if you want to use 
ISAM mode or Universal SQL. With Pass-Through SQL, you can be more creative.

I use separate schemas for individual customers in the same database on the 
same server, allowing customer-specific data as well as data-sharing between 
schemas. The user that is specified in the connection string determines the 
default schema for the connection, similar in scope to specifying a default 
directory with "Set Default".

There are a few problems with programmatically creating schemas, as the 
PGDBE doesn't always create the entire environment required for ISAM mode, 
but that can be manually fixed with some extra code to create the required 
triggers, etc.

Andreas

Andreas Gehrs-Pahl
Absolute Software, LLC

phone: (989) 723-9927
email: Andreas@AbsoluteSoftwareLLC.com
web:   http://www.AbsoluteSoftwareLLC.com
[F]:   https://www.facebook.com/AbsoluteSoftwareLLC