systemprompt_evaluation/models/
run.rs

1//! Evaluation run model and lifecycle states.
2//!
3//! Copyright (c) systemprompt.io — Business Source License 1.1.
4//! See <https://systemprompt.io> for licensing details.
5
6use chrono::{DateTime, Utc};
7use serde::{Deserialize, Serialize};
8use systemprompt_identifiers::{EvalRubricId, EvalRunId, UserId};
9
10#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
11#[serde(rename_all = "lowercase")]
12pub enum EvalRunKind {
13    Judge,
14    Replay,
15    Pairwise,
16}
17
18impl EvalRunKind {
19    #[must_use]
20    pub const fn as_str(self) -> &'static str {
21        match self {
22            Self::Judge => "judge",
23            Self::Replay => "replay",
24            Self::Pairwise => "pairwise",
25        }
26    }
27}
28
29impl std::str::FromStr for EvalRunKind {
30    type Err = String;
31
32    fn from_str(s: &str) -> Result<Self, Self::Err> {
33        match s {
34            "judge" => Ok(Self::Judge),
35            "replay" => Ok(Self::Replay),
36            "pairwise" => Ok(Self::Pairwise),
37            other => Err(format!("unknown eval run kind: {other}")),
38        }
39    }
40}
41
42#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
43#[serde(rename_all = "lowercase")]
44pub enum EvalRunStatus {
45    Running,
46    Completed,
47    Failed,
48}
49
50impl EvalRunStatus {
51    #[must_use]
52    pub const fn as_str(self) -> &'static str {
53        match self {
54            Self::Running => "running",
55            Self::Completed => "completed",
56            Self::Failed => "failed",
57        }
58    }
59}
60
61#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
62#[serde(rename_all = "lowercase")]
63pub enum TriggerSource {
64    Scheduled,
65    Cli,
66    Manual,
67}
68
69impl TriggerSource {
70    #[must_use]
71    pub const fn as_str(self) -> &'static str {
72        match self {
73            Self::Scheduled => "scheduled",
74            Self::Cli => "cli",
75            Self::Manual => "manual",
76        }
77    }
78}
79
80#[derive(Debug, Clone)]
81pub struct EvalRun {
82    pub id: EvalRunId,
83    pub kind: EvalRunKind,
84    pub status: EvalRunStatus,
85    pub judge_provider: String,
86    pub judge_model: String,
87    pub sample_size: i32,
88    pub scored_count: i32,
89    pub failed_count: i32,
90    pub cost_microdollars: i64,
91    pub created_by: UserId,
92    pub created_at: DateTime<Utc>,
93    pub completed_at: Option<DateTime<Utc>>,
94    pub error_message: Option<String>,
95    pub rubric_id: Option<EvalRubricId>,
96    pub trigger_source: TriggerSource,
97}
98
99#[derive(Debug, Clone)]
100pub struct NewRunParams {
101    pub kind: EvalRunKind,
102    pub judge_provider: String,
103    pub judge_model: String,
104    pub sample_size: i32,
105    pub created_by: UserId,
106    pub rubric_id: Option<EvalRubricId>,
107    pub trigger_source: TriggerSource,
108}
systemprompt_evaluation/models/run.rs

systemprompt_evaluation/models/
run.rs
