rho-coding-agent 2.17.0

A fast Rust agent harness with a small footprint and opinionated defaults
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134

use std::sync::Arc;

use pretty_assertions::assert_eq;
use rho_providers::model::{ModelMetadata, ReasoningRequestSource};
use rho_sdk::{
    model::{
        context::estimate_context_tokens, ContentBlock, ContextUsage, ModelEvent, ModelIdentity,
        ModelResponse, ModelUsage,
    },
    provider::{ScriptedProvider, ScriptedTurn},
    ReasoningLevel, UserInput,
};
use tokio::sync::oneshot;

use crate::{
    app::interactive_runtime::{test_runtime, InteractiveRuntime},
    tui::{
        background_tasks::{SessionOutput, TaskId, TaskOutput},
        tests::test_app,
        App,
    },
};

/// The idle dispatch point for metadata, without a terminal: take what the
/// session accepts now and apply it.
async fn apply_finished_model_metadata(app: &mut App, agent: &mut InteractiveRuntime) {
    while let Some(output) = app.take_next_finished_task(agent) {
        let TaskOutput::Session(SessionOutput::ModelMetadata {
            reasoning_at_start,
            metadata,
        }) = output
        else {
            panic!("only a metadata task was registered");
        };
        app.apply_model_metadata(agent, reasoning_at_start, metadata)
            .await;
    }
}

// Covers: deferred, unchanged metadata discards a successful provider baseline,
// or a no-op guard prevents a genuine reasoning change from being installed.
// Owner: host metadata/session lifecycle. Inject completion in-process because
// PTY cannot control the catalog task without adding a production test hook;
// calibrated_context_auto_compact owns the visible next-prompt compact behavior.
#[tokio::test]
async fn deferred_metadata_preserves_context_unless_reasoning_changes() {
    for effective in [ReasoningLevel::Low, ReasoningLevel::High] {
        let mut app = test_app();
        let provider = ScriptedProvider::new(
            ModelIdentity::new("openai", "openai-responses", "gpt-5.5"),
            [ScriptedTurn::streaming(
                vec![ModelEvent::Usage(ModelUsage {
                    input_tokens: Some(100_000),
                    ..ModelUsage::default()
                })],
                ModelResponse::Assistant(vec![ContentBlock::Text("done".into())]),
            )],
        );
        let mut agent = test_runtime(Vec::new()).await;
        agent
            .replace_provider(Arc::new(provider), ReasoningLevel::Low, "api-key")
            .unwrap();
        agent.set_context_window(Some(131_072)).unwrap();

        let metadata = ModelMetadata {
            usable_context_window: Some(131_072),
            supported_reasoning_levels: Some(vec![effective]),
            reasoning_capabilities_known: true,
            reasoning_metadata_complete: true,
            ..ModelMetadata::default()
        };
        let (release, fetched) = oneshot::channel();
        let (finished, completion) = oneshot::channel();
        let reasoning_at_start = (
            ReasoningLevel::Low,
            ReasoningRequestSource::PersistedOrDefault,
        );
        app.tasks.spawn(
            TaskId::ModelMetadata,
            async move {
                let metadata = fetched.await.unwrap();
                // No await after this signal: the current-thread executor
                // finishes this task before the test can observe the result.
                finished.send(()).unwrap();
                Some(metadata)
            },
            move |metadata| {
                SessionOutput::ModelMetadata {
                    reasoning_at_start,
                    metadata,
                }
                .into()
            },
        );

        agent.start(UserInput::text("work"), None).await.unwrap();
        release.send(metadata.clone()).unwrap();
        completion.await.unwrap();
        assert!(app.tasks.has_finished());
        apply_finished_model_metadata(&mut app, &mut agent).await;
        assert!(app.tasks.has_pending());
        assert_eq!(app.model_metadata, None);

        agent.finish_run().await.unwrap();
        let calibrated = agent.take_context_usage().unwrap();
        assert!(calibrated.tokens.unwrap() >= 100_000);
        let history = agent.history();

        apply_finished_model_metadata(&mut app, &mut agent).await;

        let expected_context = if effective == ReasoningLevel::Low {
            calibrated
        } else {
            assert_eq!(
                app.info
                    .services
                    .config_repository
                    .load()
                    .unwrap()
                    .reasoning,
                effective
            );
            ContextUsage::estimated(
                estimate_context_tokens(&history, &[]),
                calibrated.context_window,
            )
        };
        assert!(!app.tasks.has_pending());
        assert_eq!(app.model_metadata, Some(metadata));
        assert_eq!(app.info.runtime.reasoning, effective);
        assert_eq!(agent.history(), history);
        assert_eq!(agent.take_context_usage(), Some(expected_context));
    }
}