Struct Answer 

pub struct Answer {

    pub kind: QuestionType,
    pub probabilities: IndexMap<String, f64>,
    pub confidence: Option<f64>,
    pub choice: Option<String>,
    pub noul: Option<f64>,
    pub score: Option<f64>,
    pub legend: Option<IndexMap<String, String>>,
    pub selected: Option<Vec<String>>,
    pub threshold: Option<f64>,
    pub set_confidence: Option<f64>,
    pub action: Option<String>,
    pub answer: Option<String>,
    pub expected_costs: Option<IndexMap<String, f64>>,
    pub calibration: Option<String>,
    pub refused: Option<String>,
    pub max_error: Option<String>,
    pub evidence: Option<Vec<EvidenceSpan>>,
    pub objective: Option<MultiObjective>,
    pub expected_score: Option<f64>,
    pub constraints_applied: Option<Value>,
    pub tool: Option<String>,
    pub tools: Option<Vec<String>>,
    pub call: Option<bool>,
    pub multi: Option<bool>,
    pub ranking: Option<Vec<ToolRanking>>,
    pub deferred: Option<bool>,
    pub experimental: Option<bool>,
    pub label: Option<String>,
    pub top_label: Option<String>,
    pub labels: Option<Vec<String>>,
    pub scores: Option<IndexMap<String, f64>>,
    pub automated: Option<Automation>,
    pub claims: Option<IndexMap<String, ClaimResult>>,
    pub all_supported: Option<bool>,
    pub not_supported: Option<Vec<String>>,
    pub compressed: Option<String>,
    pub kept: Option<Vec<usize>>,
    pub units: Option<Vec<CompressionUnit>>,
    pub unit: Option<CompressionUnitKind>,
    pub original_chars: Option<usize>,
    pub compressed_chars: Option<usize>,
    pub ratio: Option<f64>,
    pub data: Option<IndexMap<String, Value>>,
    pub fields: Option<IndexMap<String, ExtractField>>,
    pub extra: Map<String, Value>,
}

Native answer. Fields used by other question types are absent. Unknown fields are retained in extra.

Fields

kind: QuestionType

Question type.

probabilities: IndexMap<String, f64>

Native probabilities (for multi: independent P(yes) values).

confidence: Option<f64>

Native max probability for choice, noul, score and act. Multi uses set_confidence.

choice: Option<String>

Chosen key for choice.

noul: Option<f64>

P(yes) for noul.

score: Option<f64>

Weighted zero-based score.

legend: Option<IndexMap<String, String>>

Displayed option texts for score, indexed by key.

selected: Option<Vec<String>>

Selected keys for multi.

threshold: Option<f64>

Multi selection threshold.

set_confidence: Option<f64>

Probability of precisely the selected set for multi.

action: Option<String>

Chosen action for act.

answer: Option<String>

Most probable outcome for act.

expected_costs: Option<IndexMap<String, f64>>

Expected costs by action for act.

calibration: Option<String>

Calibration status for act.

refused: Option<String>

Refused automatic action for act.

max_error: Option<String>

Message when the requested certified threshold is unavailable.

evidence: Option<Vec<EvidenceSpan>>

Evidence offsets count Unicode characters, not bytes or UTF-16 code units.

objective: Option<MultiObjective>

Multi selection objective; absent for legacy threshold responses.

expected_score: Option<f64>

Expected score of the selected set.

constraints_applied: Option<Value>

Constraint diagnostics preserved from the server.

tool: Option<String>

Selected single tool, null on certified deferral.

tools: Option<Vec<String>>

Selected tools in multi mode.

call: Option<bool>

Whether the result selects an actual tool.

multi: Option<bool>

Whether this is a multi-tool result.

ranking: Option<Vec<ToolRanking>>

Ranked tool candidates.

deferred: Option<bool>

Whether the automatic decision was deferred.

experimental: Option<bool>

Experimental task marker.

label: Option<String>

Single classification label.

top_label: Option<String>

Top classification label.

labels: Option<Vec<String>>

Selected classification labels.

scores: Option<IndexMap<String, f64>>

Classification probabilities.

automated: Option<Automation>

Task-specific certified automation; null values mean unavailable.

claims: Option<IndexMap<String, ClaimResult>>

Per-claim verification results.

all_supported: Option<bool>

Whether every claim is supported.

not_supported: Option<Vec<String>>

Flagged claim keys.

compressed: Option<String>

Source-preserving compressed text.

kept: Option<Vec<usize>>

Kept source unit indices.

units: Option<Vec<CompressionUnit>>

Source unit spans in Unicode code points.

unit: Option<CompressionUnitKind>

Compression segmentation.

original_chars: Option<usize>

Source Unicode code point count.

compressed_chars: Option<usize>

Compressed Unicode code point count.

ratio: Option<f64>

Compressed/source character ratio.

data: Option<IndexMap<String, Value>>

Nested extracted data; missing values are null.

fields: Option<IndexMap<String, ExtractField>>

Extraction field provenance and probabilities.

extra: Map<String, Value>

Future native fields.

Trait Implementations

impl Clone for Answer

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 Answer

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for Answer

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.