Leading international antiquarian booksellers
Foxclever rewrote their FoxPro for Windows system in Visual FoxPro (VFP) v9.0 and using SQL server based data. Systems include; Customer Relationship management, stock recording & cataloguing and management reporting.
Taking a well loved, well used, but ageing system originally written in 1993 using FoxPro for Windows 2.6, Foxclever rewrote the system using Visual FoxPro v9.0. At the same time the decision was taken to move the data from DBF files to SQL server.
The aim of the rewrite was to produce a system that had a more modern look and feel, was easier to use with an improved flow of work and most importantly included new functions requested by the business. Importantly the data would be migrated from DBF files to Microsoft SQL Server to achieve greater data integrity, stability and to cater for an increasing data volume.
A system specification was created for the new system outlining the requirements and design for each functional area. It was possible to keep the specification relatively brief as the existing system acted as a functional guide where the new system was not changing significantly. The specification was reviewed in detail with the customer and version 2 of the specification was produced, reviewed and approved a short time later.
The first step taken was to analyse the existing data and produce a new structure in SQL server. Improvements were implemented in relational data form and additional fields included as required. Once this was complete, data migration programs were written to move the data from their old DBF home to the new SQL server database.
Actual development progressed in functional business areas; customer management, stock management, ... Each area was developed and released in small working sections, demonstrated and given to the users for initial testing.
Final development was followed by system testing in a number of scenarios. Data migration routines developed at the beginning was used many times through the development cycle and allowed users to test using "real" data.