Minor refactor and cleanup #716
Conversation
| use sqlx::{query, SqliteExecutor}; | ||
|
|
||
| use crate::errors::MbtResult; |
There was a problem hiding this comment.
@upsicleclown do you mean why it has been moved to a separate line separated by a blank line? We have a few config options already set - https://github.com/maplibre/martin/blob/main/rustfmt.toml - but I don't want to enable it "by default" because the stable rustfmt throws an error if rustfmt has some settings it does not support. So keeping it commented out.
Note that reorder_imports_in_group seem to have been removed - try adding it to rustfmt and running cargo +nightly fmt - and it will complain.
I added a fmt2 just target to do it from CLI
This will simplify some planned work
martin/src/file_config.rs
Outdated
| pub fn extract_file_config(&mut self) -> FileConfig { | ||
| match self { | ||
| FileConfigEnum::Path(path) => FileConfig { | ||
| paths: Some(OneOrMany::One(mem::take(path))), |
There was a problem hiding this comment.
nit: This looked cleaner when One and Many were imported directly
There was a problem hiding this comment.
fixed. This one is kinda tricky because often when you see One, it might not give enough context - so i have seen a lot of things like io::Result when things are not extra clear.
|
|
||
| impl Sources { | ||
| pub fn insert(&mut self, id: String, source: Box<dyn Source>) { | ||
| self.0.insert(id, source); |
There was a problem hiding this comment.
nit: It might be worth giving the HashMap in Sources a name, so that the accesses are clearer
There was a problem hiding this comment.
this is a fairly common pattern - esp when there is only one value wrapped. I would rather not change it to a named struct
IdResolverto a separate filejust fmt2build target to format code using nightly modeData<AppState>into individual state objects, e.g.Data<Sources>(allows other source types, separate fromSourcestype)Sourcesstruct (a simple wrapper over a HashMap)