git bisect
I fixed two bugs in Datasette using - a tool which lets you run an automated binary search against a commit log to find the source of a bug.
Notes on software engineering
521–540 of 547 posts, newest first.
I fixed two bugs in Datasette using - a tool which lets you run an automated binary search against a commit log to find the source of a bug.
Today I decided to try a slightly different approach, using JSON to store changes to a table in an audit log.
I guess I have to learn TypeScript now. Here's how I got started in as few steps as possible, with the help of Get Started With Typescript in 2019 by Robert Cooper.
For sqlite-utils issue 464 I implemented a fix to a tiny bug in a dependency in my own fork on GitHub.
sqlite-utils provides a decorator for registering custom Python functions that looks like this: I wanted to add an optional deterministic=True parameter to the decorator.
I was trying to use to glue together some column values into a stiched together Markdown document. My first attempt looked like this: This attempt didn't work, because the order of the elements…
Useful for seeing what's available for conditions (see context and expression syntax). I copied this example action from here and deployed it here. Here's an example run.
I wanted the ability to test my Python software against specific version of SQLite on macOS. I found a way to do that using pysqlite3.
I've been writing my Fly logs to S3 in newline-delimited JSON format using the recipe described in Writing Fly logs to S3. I recently needed to run a search against those logs.
From this example I learned that you can set environment variables such that they will be available in ALL jobs once at the top of a workflow: I had previously been using this much more verbose…
I finally figured out a process that works for me for hacking on Python CLI utilities using uv to manage my development environment, thanks to a little bit of help from Charlie Marsh.
I use pytest-httpx in a bunch of my projects. Occasionally I run into test failures like this one, which can sometimes be really hard to figure out: Today I figured out a pattern for investigating…
I wrote the tests using playwright-pytest, which lets you write tests in Python using Microsoft's Playwright browser automation library.
A popular nightmare scenario for AI is giving it access to tools, so it can make API calls and execute its own code and generally break free of the constraints of its initial environment.
WizardLM is "an Instruction-following LLM Using Evol-Instruct". It's a fine-tuned model on top of Meta's LLaMA.
I needed this for VIAL issue 722. I decided to disable form submissions for two seconds after they are submitted, to protect against accidental double submissions without risk of unexpected issues…
Uvicorn silently ignores exceptions that occur during startup against the ASGI lifespan protocol - see starlette/issues/486.
This is almost really simple. I used electron/update-electron-app for it, the instructions for which are: - Add it to with - Make sure your field in that file points to your GitHub repository - Use…
Facebook's LLaMA is a "collection of foundation language models ranging from 7B to 65B parameters", released on February 24th 2023. It claims to be small enough to run on consumer hardware.
I had some data with weird co-ordinates in it: Note the and in there. I know that this data refers to somewhere in Houston, and a few people on Twitter pointed out that it's actually really simple…