I must first excuse myself. I haven't been working on Rye for a while. It's not that I forgot about it, far from it, but my regular projects took over for a while. I needed to make major updates at the end of the year at some project, and also re-started a SaaS project for compact support teams that I created in Rebol more than 10 years ago, used ever since, but haven't found time to finalize and offer to others yet.
I was also struck a little by ChatGPT demos, and took some time to familiarize myself with GPT, BERT and similar techologies. This was all Python based, so I had no chance to inject it with Rye.
This period is unfortunately not over yet, but last week I needed to make a simple script on one of the servers, to de-duplicate some data. Users entered the same contacts multiple times and assigned tasks to them.
Script needed to find the duplicates (triplicates, etc.), assign all tasks from the older one to the newer one, and remove the older contact. For the simplicity and debug-ability I made it interactive, using the Rye shell itself ...
It's simple, mostly uses the SQL dialect, which creates prepared SQL statements (no string concatenation) out of pure Rye blocks. Query returns the Rye specific Spreadsheet datatype that I wrote about here not long ago.
Follow us on Github. I'll be back! B)
Komentarji
Objavite komentar