Best Practices For Insta Filtering/Redactions

[NicholasLYang]

Hi, we're trying out Insta in turborepo. We currently use Prysk, which lets you write inline regexes into your snapshots (example). This is pretty convenient for stuff like normalizing paths across platforms.

I'm wondering what the equivalent solution would look like in Insta. I tried doing a filter that just replaces \ with /, but that isn't great because not all occurrences of \ are in paths. I can try to manually normalize the data before it reaches Insta, but that's a little tedious tbh. I could also split snapshots across OS, so I get a turbo_test_windows.snap and a turbo_test_mac.snap, but that makes updating snapshots difficult, especially across operating systems.

To be clear, this isn't a feature request. I'm just curious what the idiomatic "Insta-y" solution would be.

Thanks!

[max-sixty]

Possibly dynamic redactions? https://insta.rs/docs/redactions/ Then we can use a regex / any custom logic...

[NicholasLYang]

Hmm okay. That makes sense for structured data, but perhaps not for larger blocks of text. I'll give it a shot though!