Detect reserved keywords in specific generator languages

[hellovai]

When users add an enum or class field named True or None we run into issues in python that are very hard for the user to debug due to reserved word usage.

To address this we should instead preemptively detect this in our static analyzer.

[greptile-apps]

To detect reserved keywords in specific generator languages, extend the assert_is_not_a_reserved_scalar_type function in /engine/baml-lib/parser-database/src/names/mod.rs to include checks for reserved keywords in Python (e.g., True, None).

fn assert_is_not_a_reserved_scalar_type(ident: &ast::Identifier, ctx: &mut Context<'_>) {
    let reserved_keywords = ["True", "None"];
    if StaticType::try_from_str(ident.name()).is_some() || reserved_keywords.contains(&ident.name()) {
        ctx.push_error(DatamodelError::new_reserved_scalar_type_error(
            ident.name(),
            ident.span().clone(),
        ));
    }
}

This will preemptively detect and report errors for reserved keywords in Python during the static analysis phase.

References

/engine/baml-lib/parser-database/src/names/mod.rs
/engine/baml-lib/baml-core/src/configuration.rs

About Greptile

This response provides a starting point for your research, not a precise solution.

Help us improve! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

Ask Greptile · Edit Issue Bot Settings