search

fgardt / factorio-scanner

Factorio tools to render and work with blueprint strings, (de)serialize prototype dumps and more
MIT License
13 stars 2 forks source link

Support different application and user data directories and improve path inference #140

Closed mkfmnn closed 2 months ago

mkfmnn commented 2 months ago

On my Mac this lets me run the scanner with just the command scanner --out x.png file bp.txt

mkfmnn commented 2 months ago

only small issue I have is the string error type

I initially did try using ScannerError but the CLI UX felt messy (no reason to be showing the source location of the error) so I more preferred emulating the clap errors—though I stopped short of reworking the command-line parsing to give me access to the Command to hook into clap's error reporting.

Let me know if you have any references/reading for the pull request workflow or commit/coding conventions you'd like me to follow for the future. I tried to emulate what it looked like you've been doing and it feels like a standard process that other devs use too but I don't know if there's an official standard for it somewhere.

fgardt commented 2 months ago

I initially did try using ScannerError but the CLI UX felt messy (no reason to be showing the source location of the error) so I more preferred emulating the clap errors—though I stopped short of reworking the command-line parsing to give me access to the Command to hook into clap's error reporting.

I want to get rid of error-stack and just switch to thiserror in the scanner crate aswell. I've never hooked into clap's error reporting but if thats a thing then that could be quite nice to have for the cases where more args are required yeah.

Let me know if you have any references/reading for the pull request workflow or commit/coding conventions you'd like me to follow for the future. I tried to emulate what it looked like you've been doing and it feels like a standard process that other devs use too but I don't know if there's an official standard for it somewhere.

I don't have a contributing.md or anything along those lines. I mostly just try to stick to conventional commits and thats about it in terms of procedure. Branch naming is up to you and your commits while working on a PR are also not too important since they'll be squashed before merging anyways.

also: I have no idea why the code scanning CI job is taking forever...