csift 0.9.3

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
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235

//! search exchange headers: stable id-prefix tokens, collisions, parent tokens, the match banner.

use crate::harness::*;

#[test]
fn search_header_tokens_are_stable_across_invocations() {
    // The header token derives from the transcript id (its leading chars), never from
    // enumeration order - two identical invocations emit byte-identical output, so a token
    // pasted from an earlier run still names the same transcript.
    let h = populated_home();
    let a = h.run(&["search", "carry"]);
    let b = h.run(&["search", "carry"]);
    assert!(
        a.success && b.success,
        "stderr: {} / {}",
        a.stderr,
        b.stderr
    );
    assert_eq!(
        a.stdout, b.stdout,
        "byte-identical output across identical invocations"
    );
}

#[test]
fn search_header_token_collision_lengthens_the_group_only() {
    // Two DISTINCT ids sharing their first 8 chars lengthen TOGETHER to their first 12 raw
    // chars (for a uuid that spans the first dash - still a valid `@` target); the
    // non-colliding third id stays at 8. The bare collided 8-prefix never appears as a token.
    let h = Home::new();
    let _ = header_collision_scenario(&h);
    let out = h.run(&["search", "SEEDWORD"]);
    assert!(out.success, "stderr: {}", out.stderr);
    assert!(
        out.stdout.contains("aaaabbbb-111\u{b7}t"),
        "colliding id 1 lengthens to 12: {}",
        out.stdout
    );
    assert!(
        out.stdout.contains("aaaabbbb-222\u{b7}t"),
        "colliding id 2 lengthens to 12: {}",
        out.stdout
    );
    assert!(
        out.stdout.contains("ccccdddd\u{b7}t"),
        "non-collider stays at 8: {}",
        out.stdout
    );
    assert!(
        !out.stdout.contains("aaaabbbb\u{b7}t"),
        "the collided bare 8-prefix must not appear as a token: {}",
        out.stdout
    );
}

#[test]
fn search_lengthened_uuid_token_resolves_and_short_prefix_fails_loud() {
    // A collision-lengthened header token (12 chars, spanning the uuid's first dash) is a
    // valid `@` target; the ambiguous bare 8-prefix fails loud naming the candidates.
    let h = Home::new();
    let (c1, c2, _) = header_collision_scenario(&h);
    let one = h.run(&["search", "SEEDWORD", "@aaaabbbb-111"]);
    assert!(one.success, "stderr: {}", one.stderr);
    assert!(one.stdout.contains("COLLIDEONE"), "got: {}", one.stdout);
    assert!(
        !one.stdout.contains("COLLIDETWO"),
        "the sibling session must be out of scope: {}",
        one.stdout
    );
    let ambi = h.run(&["search", "SEEDWORD", "@aaaabbbb"]);
    assert!(!ambi.success, "an ambiguous prefix must error");
    assert!(
        ambi.stderr.contains("AMBIGUOUS") && ambi.stderr.contains(c1) && ambi.stderr.contains(c2),
        "the error names both candidates: {}",
        ambi.stderr
    );
}

#[test]
fn search_subagent_header_carries_parent_token_on_every_exchange() {
    // EVERY subagent exchange header carries `(parent <first-8-of-owning-uuid>)` - a
    // tail-truncated read must still resolve ownership; top-level headers carry no parent.
    let h = Home::new();
    subagents_only_scenario(&h);
    let out = h.run(&["search", "", at(SESS).as_str()]);
    assert!(out.success, "stderr: {}", out.stderr);
    assert!(
        out.stdout
            .contains(&format!("sub111\u{b7}t0 (parent {})", &SESS[..8])),
        "subagent header carries the parent token: {}",
        out.stdout
    );
    for line in out.stdout.lines() {
        if line.starts_with(&format!("{}\u{b7}t", &SESS[..8])) {
            assert!(
                !line.contains("(parent"),
                "a top-level header must not carry a parent: {line}"
            );
        }
    }
}

#[test]
fn search_and_show_resolve_an_agent_id_prefix_token() {
    // An 8-char prefix of a subagent's bare-hex id - the exact header token `search`
    // emits - resolves as an `@` target on every target-taking surface.
    let h = Home::new();
    let (_, _, gamma) = agent_prefix_scenario(&h);
    let out = h.run(&["search", "AGENTGAMMA", &format!("@{}", &gamma[..8])]);
    assert!(out.success, "stderr: {}", out.stderr);
    assert!(out.stdout.contains("AGENTGAMMA"), "got: {}", out.stdout);
    let s = h.run(&["show", &format!("@{}", &gamma[..8]), "--line", "1"]);
    assert!(s.success, "show by agent prefix: {}", s.stderr);
    assert!(s.stdout.contains("AGENTGAMMA"), "got: {}", s.stdout);
}

#[test]
fn search_agent_twelve_hex_token_falls_back_to_unique_prefix() {
    // Two agents sharing their first 8 hex lengthen their header tokens to 12; a 12-hex
    // token routes as an exact agent id, misses, and falls back to a UNIQUE literal-prefix
    // match. The ambiguous 8-hex form fails loud naming both ids.
    let h = Home::new();
    let (alpha, beta, _) = agent_prefix_scenario(&h);
    let out = h.run(&["search", "seed"]);
    assert!(out.success, "stderr: {}", out.stderr);
    assert!(
        out.stdout.contains(&format!("{}\u{b7}t", &alpha[..12]))
            && out.stdout.contains(&format!("{}\u{b7}t", &beta[..12])),
        "colliding agent tokens lengthen to 12: {}",
        out.stdout
    );
    let one = h.run(&["search", "AGENT", &format!("@{}", &alpha[..12])]);
    assert!(one.success, "stderr: {}", one.stderr);
    assert!(one.stdout.contains("AGENTALPHA"), "got: {}", one.stdout);
    assert!(
        !one.stdout.contains("AGENTBETA"),
        "the sibling agent must be out of scope: {}",
        one.stdout
    );
    let ambi = h.run(&["search", "AGENT", &format!("@{}", &alpha[..8])]);
    assert!(!ambi.success, "an ambiguous agent prefix must error");
    assert!(
        ambi.stderr.contains("AMBIGUOUS")
            && ambi.stderr.contains(alpha)
            && ambi.stderr.contains(beta),
        "the error names both agent ids: {}",
        ambi.stderr
    );
}

#[test]
fn search_match_banner_at_head_mirrors_footer_and_json() {
    // The head banner carries the TRUE totals + direction before the first exchange; the
    // footer repeats the same numbers; the JSON summary's post-cap `matched` +
    // `dropped_by_cap` reconcile to the banner total.
    let h = Home::new();
    let _ = header_collision_scenario(&h);
    let out = h.run(&["search", "SEEDWORD"]);
    assert!(out.success, "stderr: {}", out.stderr);
    // No subagents in scope → the scope banner is suppressed, so the banner is line 1.
    assert_eq!(
        out.stdout.lines().next(),
        Some("matches  3 exchanges · 3 sessions · oldest first"),
        "head banner: {}",
        out.stdout
    );
    assert!(
        out.stdout
            .contains("matched 3 exchanges · 3 sessions · label=all"),
        "footer repeats the totals: {}",
        out.stdout
    );
    // Clean-corpus duals for the footer's `> 0` note gates: no malformed note, no
    // sidecar note, no zero-count drop note in the plain text mode.
    assert!(
        !out.stdout.contains("malformed") && !out.stdout.contains("sidecar"),
        "no zero-count footer notes on a clean run: {}",
        out.stdout
    );
    let js = h.run(&["search", "SEEDWORD", "--format", "json"]);
    assert!(js.success, "stderr: {}", js.stderr);
    assert!(
        !js.stdout.contains("matches  "),
        "no banner in JSON mode: {}",
        js.stdout
    );
    let last: serde_json::Value =
        serde_json::from_str(js.stdout.lines().next_back().unwrap()).unwrap();
    assert_eq!(last["matched"], serde_json::json!(3));
    assert_eq!(last["sessions"], serde_json::json!(3));

    // Capped: the banner keeps the TRUE total and discloses the window; JSON reconciles.
    let cap = h.run(&["search", "SEEDWORD", "--max-count", "2"]);
    assert!(cap.success, "stderr: {}", cap.stderr);
    assert_eq!(
        cap.stdout.lines().next(),
        Some("matches  3 exchanges · 3 sessions · oldest first · showing earliest 2"),
        "capped head banner: {}",
        cap.stdout
    );
    assert!(
        cap.stdout.contains("matched 3 exchanges · 3 sessions")
            && cap.stdout.contains("1 later dropped by --max-count"),
        "capped footer: {}",
        cap.stdout
    );
    let capjs = h.run(&["search", "SEEDWORD", "--max-count", "2", "--format", "json"]);
    let last: serde_json::Value =
        serde_json::from_str(capjs.stdout.lines().next_back().unwrap()).unwrap();
    assert_eq!(
        last["matched"].as_u64().unwrap() + last["dropped_by_cap"].as_u64().unwrap(),
        3,
        "JSON post-cap matched + dropped reconcile to the banner total"
    );

    // Single-purpose modes carry no banner.
    let c = h.run(&["search", "SEEDWORD", "-c"]);
    assert!(
        !c.stdout.contains("matches  "),
        "no banner under -c: {}",
        c.stdout
    );
    let l = h.run(&["search", "SEEDWORD", "-l"]);
    assert!(
        !l.stdout.contains("matches  "),
        "no banner under -l: {}",
        l.stdout
    );
    let cb = h.run(&["search", "SEEDWORD", "--count-by", "label"]);
    assert!(
        !cb.stdout.contains("matches  "),
        "no banner under --count-by: {}",
        cb.stdout
    );
}