csift 0.12.1

ripgrep for Claude Code session transcripts: fast regex list/search over ~/.claude/projects/**/*.jsonl
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
135
136

//! Token usage: Claude Code repeats the identical usage object on every per-block
//! record of one API message, so the sums dedupe by message.id per transcript.

use crate::harness::*;

const DEDUP_ENC: &str = "-Users-dev-example-project";
const DEDUP_SESS: &str = "3f2e1d0c-9b8a-4765-b432-10fedcba9876";

#[test]
fn stats_usage_counts_one_api_message_once() {
    // Two records share one message.id with the IDENTICAL usage echo (the normal CC
    // per-block shape); a third record carries its own id.
    let h = Home::new();
    h.write(
        &format!("{DEDUP_ENC}/{DEDUP_SESS}.jsonl"),
        concat!(
            r#"{"type":"assistant","uuid":"a1","timestamp":"2026-06-07T05:00:01.000Z","message":{"role":"assistant","id":"msg_p1","model":"claude-test-1","usage":{"input_tokens":3,"output_tokens":40,"cache_read_input_tokens":100,"cache_creation_input_tokens":7},"content":[{"type":"thinking","thinking":"placeholder reasoning"}]}}"#, "\n",
            r#"{"type":"assistant","uuid":"a2","timestamp":"2026-06-07T05:00:02.000Z","message":{"role":"assistant","id":"msg_p1","model":"claude-test-1","usage":{"input_tokens":3,"output_tokens":40,"cache_read_input_tokens":100,"cache_creation_input_tokens":7},"content":[{"type":"text","text":"charted"}]}}"#, "\n",
            r#"{"type":"assistant","uuid":"a3","timestamp":"2026-06-07T05:00:03.000Z","message":{"role":"assistant","id":"msg_p2","model":"claude-test-1","usage":{"input_tokens":1,"output_tokens":2,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"done"}]}}"#, "\n",
        ),
    );
    let out = h.run(&["stats", &at(DEDUP_SESS), "--format", "json"]);
    assert!(out.success, "stderr: {}", out.stderr);
    let row: serde_json::Value = out
        .stdout
        .lines()
        .filter_map(|l| serde_json::from_str(l).ok())
        .find(|v: &serde_json::Value| v["kind"] == "session")
        .expect("session row");
    let t = &row["tokens"]["claude-test-1"];
    assert_eq!(t["output"], 42, "40 once + 2, never 82: {}", out.stdout);
    assert_eq!(t["input"], 4, "{}", out.stdout);
    assert_eq!(t["cache_read"], 100, "{}", out.stdout);
    assert_eq!(t["cache_creation"], 7, "{}", out.stdout);
}

#[test]
fn stats_usage_zeroed_replay_never_wins() {
    // The compaction-replay shape: the same message.id written again with ZEROED
    // usage. The per-field MAX keeps the real numbers regardless of order.
    let h = Home::new();
    h.write(
        &format!("{DEDUP_ENC}/{DEDUP_SESS}.jsonl"),
        concat!(
            r#"{"type":"assistant","uuid":"z1","timestamp":"2026-06-07T05:00:01.000Z","message":{"role":"assistant","id":"msg_r1","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":0,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"replayed copy"}]}}"#, "\n",
            r#"{"type":"assistant","uuid":"z2","timestamp":"2026-06-07T05:00:02.000Z","message":{"role":"assistant","id":"msg_r1","model":"claude-test-1","usage":{"input_tokens":2,"output_tokens":708,"cache_read_input_tokens":474834,"cache_creation_input_tokens":3997},"content":[{"type":"text","text":"original"}]}}"#, "\n",
        ),
    );
    let out = h.run(&["stats", &at(DEDUP_SESS), "--format", "json"]);
    let row: serde_json::Value = out
        .stdout
        .lines()
        .filter_map(|l| serde_json::from_str(l).ok())
        .find(|v: &serde_json::Value| v["kind"] == "session")
        .expect("session row");
    let t = &row["tokens"]["claude-test-1"];
    assert_eq!(
        t["output"], 708,
        "the zeroed replay never wins: {}",
        out.stdout
    );
    assert_eq!(t["cache_read"], 474834, "{}", out.stdout);
}

#[test]
fn stats_usage_dedupe_is_per_file_and_total_sums_files() {
    // The spawn message is copied into a child transcript with DIFFERENT usage: each
    // file keeps its own value (a genuine per-transcript fact); TOTAL sums both.
    let h = Home::new();
    h.write(
        &format!("{DEDUP_ENC}/{DEDUP_SESS}.jsonl"),
        concat!(
            r#"{"type":"assistant","uuid":"m1","timestamp":"2026-06-07T05:00:01.000Z","message":{"role":"assistant","id":"msg_x1","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":6945,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"spawning"}]}}"#, "\n",
        ),
    );
    h.write(
        &format!("{DEDUP_ENC}/{DEDUP_SESS}/subagents/agent-1234567890abcdef.jsonl"),
        concat!(
            r#"{"type":"assistant","uuid":"c1","timestamp":"2026-06-07T05:00:02.000Z","message":{"role":"assistant","id":"msg_x1","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":10,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"child copy"}]}}"#, "\n",
        ),
    );
    let out = h.run(&["stats", &at(DEDUP_SESS), "--format", "json"]);
    let outputs: Vec<u64> = out
        .stdout
        .lines()
        .filter_map(|l| serde_json::from_str::<serde_json::Value>(l).ok())
        .filter(|v| v["kind"] == "session")
        .filter_map(|v| v["tokens"]["claude-test-1"]["output"].as_u64())
        .collect();
    let mut sorted = outputs.clone();
    sorted.sort_unstable();
    assert_eq!(
        sorted,
        vec![10, 6945],
        "per-file values kept: {}",
        out.stdout
    );
    let summary: serde_json::Value = out
        .stdout
        .lines()
        .filter_map(|l| serde_json::from_str(l).ok())
        .find(|v: &serde_json::Value| v["kind"] == "summary")
        .expect("summary");
    assert_eq!(
        summary["tokens"]["claude-test-1"]["output"], 6955,
        "TOTAL sums files, never dedupes across them: {}",
        out.stdout
    );
}

#[test]
fn stats_usage_idless_records_count_individually() {
    // A record with no message.id keeps the old per-record semantics beside an
    // id-bearing pair (every pre-id transcript takes this path).
    let h = Home::new();
    h.write(
        &format!("{DEDUP_ENC}/{DEDUP_SESS}.jsonl"),
        concat!(
            r#"{"type":"assistant","uuid":"n1","timestamp":"2026-06-07T05:00:01.000Z","message":{"role":"assistant","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":5,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"old format"}]}}"#, "\n",
            r#"{"type":"assistant","uuid":"n2","timestamp":"2026-06-07T05:00:02.000Z","message":{"role":"assistant","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":5,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"old format two"}]}}"#, "\n",
            r#"{"type":"assistant","uuid":"n3","timestamp":"2026-06-07T05:00:03.000Z","message":{"role":"assistant","id":"msg_q1","model":"claude-test-1","usage":{"input_tokens":0,"output_tokens":30,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"content":[{"type":"text","text":"new format"}]}}"#, "\n",
        ),
    );
    let out = h.run(&["stats", &at(DEDUP_SESS), "--format", "json"]);
    let row: serde_json::Value = out
        .stdout
        .lines()
        .filter_map(|l| serde_json::from_str(l).ok())
        .find(|v: &serde_json::Value| v["kind"] == "session")
        .expect("session row");
    assert_eq!(
        row["tokens"]["claude-test-1"]["output"], 40,
        "5 + 5 individually + 30 once: {}",
        out.stdout
    );
}