Modularizing computation via Common Table Expressions (CTEs)
Karel has participated in all aspects of the solutions lifecycle from problem structuring, feasibility, solutions architecture selection, technology scanning, negotiating with vendors, implementation, deployment and user support.
These solutions entailed products for desktop, terminals, client/server, dynamic web content and multi-tier web applications addressing persistence using XBase libraries, RDMS’s (Oracle, MS, Sybase, Interbase), ORM’s (Object-Relational Mappers), NoSQL, XML-technologies and object stores.
Developed products ranged from simple CRUD systems with extensive reporting to decision support systems with high-value-addition processing.
Solutions were developed with degrees of tool-support ranging from little or none up to fully generated solutions.
Karel closely tracks processes and methods and their match to organizational capacity.
SQL allows for a terse expression of computation via the declarative style.
PostgreSQL lateral joins and set-returning functions remove the need for explicit iteration in computation.
SQL Common Table Expression (CTEs) - and some PostgreSQL-specific features - allow for new modes of modularisation of a computation.
This modularised computation still follows the economy of the declarative style, but with improved readability and testability.
Such a modularised, declarative computation resembles a functional pipeline process.
CTE's presents a modular source of existing data into such a functional pipeline.
Recursive CTEs presents a modular source of newly created data into such a functional pipeline.
- 2021 October 5 11:50 SAST
- 40 min
- South Africa 2021