Is it possible to use non-DBF databases in Alaska+Express?

[Eugene Lutsenko]

Is it possible to use non-DBF databases in Alaska+Express? I would like to abandon DBF and use databases of another standard that does not have such strict size restrictions (only 2 GB) and the number of fields (really no more than 1500).

[rdonnay]

You would need to search for a SQL database that allows the number of records and fields that you want.

Something like that would not be free.

Look at PostGreSQL and check their specs.

[Auge_Ohr]

hi Eugene,

FOXDBE (DATA-Komponente) does support 2038 FIELD(s)
PostgreSQL (Server need install) can max. have 1600 Column, depend of Type.

but you are still limited at 2^32 under Xbase++ ...

have you "test-drive" CLASS TBinFile()

[Tom]

Check out the ISAM-emulation/PGDBE. Meanwhile, this works really excellent (some few problems like with complex filters left) and has a great performance, and only few code changes are necessary. You can mix record-navigation-based models and pure SQL in one application, there are no storing limits and you can optimize your application step by step moving to native SQL.

[Auge_Ohr]

there are no storing limits and you can optimize your application step by step moving to native SQL.

did PGDBF "translate" Pointer of "Tuple" > 2°32

some other "Limit"
Excel Total number of rows and columns on a worksheet 1,048,576 rows by 16,384 columns
MySQL has hard limit of 4096 columns per table,

[Tom]

Jim, there are almost no limits that have anything to do with file sizes. Besides this, everything (in the world) is limited somehow, but if Eugene thinks about moving to FOX to exceed the DBF limits, he should take a very close look at the excellent (and very long lasting) work Alaska did with the brillant PGDBE. Upsizing is just a little monkey work, and the other changes needed are minimal. PostGreSQL is for free.

[skiman]

Hi Tom,

Are you using the PGDBE in your application? I'm interested in it, but up to now I didn't find much positive comments on it. Here it looks as you are enthousiastic about it.

[Tom]

Hi, Chris.

Yes, and it's almost complete. Our major application is really huge, and it works excellent (>98%) with the PGDBE. It's a little more work than Alaska says, especially if the file-based-version works with a multi-folder-structure for different customer situations and stuff like "SET DEFAULT" and "SET PATH" for finding tables in the data hierarchy. Since the ISAM-emulation doesn't support different schemas in PG (which supports a similar mechanism), a connection must be established for everything that was in a folder before. And complex filters are still a problem - the PGDBE tries to translate every filter into a SELECT statement, which leads to application locks. But Alaska is working on something the ADSDBE is able to do aswell - falling back to local filters if server filters can't be executed. We changed lots of filters into conditions evaluated in loops, where possible. The app is designed to support all database models (DBF, ADS, PG) in the same code.

Besides that, this is the best Alaska invented since 1998. It's really smooth, it runs fast, it works good. It's much better than ADS. There are much more possibilites, and you can mix ISAM and standard SQL wherever you want.

[rdonnay]

Yes, and it's almost complete. Our major application is really huge, and it works excellent (>98%) with the PGDBE.

This is the first I have heard about success with PGDBE.

Bobby and I were talking about this yesterday and wondering if it is ready for prime time.

[Tom]

Bobby and I were talking about this yesterday and wondering if it is ready for prime time.

It is.