Working with data stored in DataFlex (any)

Topic

New Reply

I am embarking on a project to convert a large membership database (300K members with various related tables – some with as many as 10M records) that is currently stored in a DataFlex format into an already designed SQL Server database which will use Access and web pages as the front-end. I’ve gotten some conflicting advice on how DataFlex works, how files are stored and inter-relate and the like. For one thing, I’ve been told that the data file layouts are similar to dBase, but I’ve also been told that they are more similar to Clipper. I expect there will be some data corruption within the files as the past of this data is quite checkered, but one of the objectives is to import as much of the data as is possible – reentering data for that many people is not an option.

I guess my questions boil down to two – is there a way to get at the data at the file level and convert it into something that Access can easily deal with, such as delimited text files, and if the answer to that is no, then what experience have people had using ODBC to connect to DataFlex databases. In regard to the latter, I understand that there are a couple of ODBC drivers available, one called the DataFlex ODBC Connectivity Kit and one from FlexODBC. What experiences have people had with either, and what kind of performance do you get using an ODBC driver? We plan to cut the system over a weekend, and will need to do a final conversion of the DataFlex data during that time – if it takes more than 48 hours for the conversion, I have a problem. Thanks in advance for any and all advice and experiences.

Reply | Quote

Replies

Wendell

Looks like you may have drawn a blank here as far as others experience.

The options you have here are limited so you may have to go the FlexODBC way.
Once this driver is installed, then you will have access to the tables, I link mine as opposed to import, then a Make Table Query is used to do the rest.
The table structure after Import/Link is as Access, ie Header information.
Working with the data is really quite quick, but creating queries with relationships does seem to tax the speed.

The other alternative is to create a DOC or TXT file from within Dataflex using DFQuery !!.

If the system you are using has (Usually) F1 for help or a menu which gives you the opportunity to access the “Run” menu, you can then type DFQuery into this to access some of the setup files.

Hope this helps you in your quest.
Post back if I can help further.

Reply | Quote

Hi Dave,

Thanks for the reply – there are some useful concepts in there that I assumed must exist, but you given me enough background that I can at least throw around some of the terms. On a happy note (sometimes you get lucky), the individual who put this beast together a few years ago as agreed to help with the conversion, so he should be able to create a process such that we simply turn the crank to get the data into a form I can work with, and I can take it from there. I may be back asking questions however if he starts making it more difficult than it should be. Thanks again.

Reply | Quote

Hi Dave,

I’ve made great progress in importing the data into SQL Server. I ended up using the DFQuery tool that you mentioned, and it works very nicely except for one small problem. It apparently has an 8MB buffer, and if you try to dump more than that it wraps around and overwrites the buffer over and over again till it gets to the end. So I had to resort to dumping the data in small chunks – for the detail file (2.5M records) I end up with 26 text files. It’s easy enough to automate the import once I get the text files, but at this point I haven’t been able to figure out how to automate the dump process.

DFQuery has a facility to save the query properties as a report, but I can’t figure out how to invoke that file once it’s been saved. It would help of course if I had the Help file for DFQuery, but I didn’t end up with that. Since you have had some experience with it, can you enlighten me?

Reply | Quote