Enum ToolValidation 

pub enum ToolValidation {
    Unsupported,
    Attempted(Vec<ToolOutcome>),
}

Every checker’s verdict for one generated file.

Variants

Unsupported

No tool-based validator exists for this backend’s language.

Distinct from “the tools are missing”: Java, C#, Elixir and Rust have no validator written for them at all, and reporting that as a skip would suggest installing something would help.

Attempted(Vec<ToolOutcome>)

A validator exists; one entry per checker it tried to run.

Implementations

impl ToolValidation

pub fn errors(&self) -> Vec<&str>

Findings from the checkers that actually ran.

pub fn missing_tools(&self) -> Vec<&'static str>

Checkers that were not installed.

pub fn tools_run(&self) -> Vec<&'static str>

Checkers that ran to completion.

pub fn fully_checked(&self) -> bool

Whether every checker this validator knows about actually ran.

False for Self::Unsupported: nothing checked the code, so claiming it was fully checked would be the same lie in a different shape.

pub fn into_result(self) -> Result<(), Vec<String>>

Collapse to pass/fail under the current strict_mode_enabled policy.

Outside strict mode a missing tool is tolerated. Inside it, a missing tool is a failure – that is what stops a checker from quietly falling out of CI and taking its coverage with it.

pub fn into_result_with_strictness( self, strict: bool, ) -> Result<(), Vec<String>>

Self::into_result with the policy passed explicitly, so tests can exercise both modes without mutating process-global environment state (which races under cargo test’s thread-per-test execution).

Self::Unsupported is deliberately not a strict-mode failure. Strict mode exists to catch a checker that fell out of the CI image, and every one of its failures should be fixable by installing something. A language with no validator written for it is a real gap – Java, C#, Elixir and Rust are all in it – but no CI configuration closes it, so failing here would leave a permanently red build with no action attached. That gap is tracked by an explicit inventory test (backends_with_no_tool_validator_are_a_known_and_shrinking_set in tests/tool_validation.rs) which fails when the set changes in either direction, rather than by a signal nobody can act on.

Trait Implementations

impl Clone for ToolValidation

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ToolValidation

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Eq for ToolValidation

impl PartialEq for ToolValidation

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for ToolValidation