personal memory agent
Merge branch 'hopper-3std5nvh-stats-contract-test'
+217
+217
tests/test_stats_contract.py
+217
tests/test_stats_contract.py
···
1
+
# SPDX-License-Identifier: AGPL-3.0-only
2
+
# Copyright (c) 2026 sol pbc
3
+
4
+
import importlib
5
+
import json
6
+
from pathlib import Path
7
+
8
+
9
+
CONTRACT_FIELDS = [
10
+
("days", "stats.days"),
11
+
("totals", "stats.totals"),
12
+
("heatmap", "stats.heatmap"),
13
+
("totals.day_bytes", "totals.day_bytes"),
14
+
("totals.pending_segments", "totals.pending_segments"),
15
+
("totals.outputs_processed", "totals.outputs_processed"),
16
+
("totals.outputs_pending", "totals.outputs_pending"),
17
+
("days.*.day_bytes", "dayData.day_bytes"),
18
+
("totals.total_transcript_duration", "total_audio_duration"),
19
+
("totals.total_percept_duration", "total_screen_duration"),
20
+
("totals.transcript_sessions", "audio_sessions"),
21
+
("tokens.by_model", "token_totals_by_model"),
22
+
("tokens.by_day", "token_usage_by_day"),
23
+
("facets.counts_by_day", "facet_counts_by_day"),
24
+
("agents.counts_by_day", "agent_counts_by_day"),
25
+
("days.*.transcript_duration", "audio_duration"),
26
+
("days.*.percept_duration", "screen_duration"),
27
+
("tokens.by_day.*.*.input_tokens", "input_tokens"),
28
+
("tokens.by_day.*.*.output_tokens", "output_tokens"),
29
+
("tokens.by_day.*.*.reasoning_tokens", "reasoning_tokens"),
30
+
("tokens.by_model.*.total_tokens", "total_tokens"),
31
+
]
32
+
33
+
JS_PATH = (
34
+
Path(__file__).resolve().parent.parent
35
+
/ "apps"
36
+
/ "stats"
37
+
/ "static"
38
+
/ "dashboard.js"
39
+
)
40
+
41
+
42
+
def _resolve_path(data, path):
43
+
stack = [(data, path.split("."))]
44
+
while stack:
45
+
current, parts = stack.pop()
46
+
if not parts:
47
+
return True
48
+
if not isinstance(current, dict):
49
+
continue
50
+
part, rest = parts[0], parts[1:]
51
+
if part == "*":
52
+
stack.extend((value, rest) for value in current.values())
53
+
elif part in current:
54
+
stack.append((current[part], rest))
55
+
return False
56
+
57
+
58
+
def _build_journal(base_path):
59
+
journal = base_path
60
+
day = journal / "20240101"
61
+
seg1 = day / "default" / "123456_300"
62
+
seg2 = day / "default" / "134500_300"
63
+
seg1.mkdir(parents=True)
64
+
seg2.mkdir(parents=True)
65
+
(day / "agents").mkdir(parents=True)
66
+
67
+
audio_lines = [
68
+
{"raw": "raw.flac"},
69
+
{"start": "10:00:00", "text": "hello"},
70
+
{"start": "10:05:00", "text": "world"},
71
+
]
72
+
(seg1 / "audio.jsonl").write_text(
73
+
"\n".join(json.dumps(line) for line in audio_lines) + "\n"
74
+
)
75
+
76
+
screen_lines = [
77
+
{"raw": "screen.webm"},
78
+
{"frame_id": 1, "timestamp": 1000.0, "text": "frame one"},
79
+
{"frame_id": 2, "timestamp": 1060.0, "text": "frame two"},
80
+
]
81
+
(seg1 / "screen.jsonl").write_text(
82
+
"\n".join(json.dumps(line) for line in screen_lines) + "\n"
83
+
)
84
+
85
+
(seg2 / "audio.flac").write_bytes(b"fLaC")
86
+
(day / "agents" / "flow.md").write_text("")
87
+
88
+
events_dir = journal / "facets" / "work" / "events"
89
+
events_dir.mkdir(parents=True)
90
+
event = {
91
+
"type": "meeting",
92
+
"start": "09:00:00",
93
+
"end": "09:30:00",
94
+
"title": "standup",
95
+
"summary": "daily sync",
96
+
"work": True,
97
+
"participants": [],
98
+
"details": "",
99
+
"facet": "work",
100
+
"agent": "meetings",
101
+
"occurred": True,
102
+
"source": "20240101/agents/meetings.md",
103
+
}
104
+
(events_dir / "20240101.jsonl").write_text(json.dumps(event) + "\n")
105
+
106
+
tokens_dir = journal / "tokens"
107
+
tokens_dir.mkdir()
108
+
token = {
109
+
"timestamp": 1704067200.0,
110
+
"timestamp_str": "20240101_120000",
111
+
"model": "test-model",
112
+
"context": "test",
113
+
"usage": {
114
+
"input_tokens": 100,
115
+
"output_tokens": 50,
116
+
"cached_tokens": 10,
117
+
"reasoning_tokens": 5,
118
+
"total_tokens": 165,
119
+
},
120
+
}
121
+
(tokens_dir / "20240101.jsonl").write_text(json.dumps(token) + "\n")
122
+
123
+
return journal
124
+
125
+
126
+
def _scan_output(journal, stats_mod):
127
+
js = stats_mod.JournalStats()
128
+
js.scan(str(journal))
129
+
return js.to_dict()
130
+
131
+
132
+
def test_generated_stats_pass_schema(tmp_path, monkeypatch):
133
+
stats_mod = importlib.import_module("think.journal_stats")
134
+
schema_mod = importlib.import_module("think.stats_schema")
135
+
journal = _build_journal(tmp_path)
136
+
monkeypatch.setenv("_SOLSTONE_JOURNAL_OVERRIDE", str(journal))
137
+
138
+
output = _scan_output(journal, stats_mod)
139
+
140
+
errors = schema_mod.validate(output)
141
+
assert errors == [], f"Validation errors: {errors}"
142
+
assert output["schema_version"] == schema_mod.SCHEMA_VERSION
143
+
144
+
145
+
def test_contract_fields_exist_in_output(tmp_path, monkeypatch):
146
+
stats_mod = importlib.import_module("think.journal_stats")
147
+
journal = _build_journal(tmp_path)
148
+
monkeypatch.setenv("_SOLSTONE_JOURNAL_OVERRIDE", str(journal))
149
+
150
+
output = _scan_output(journal, stats_mod)
151
+
152
+
for python_path, _ in CONTRACT_FIELDS:
153
+
assert _resolve_path(output, python_path), f"{python_path} missing from stats output"
154
+
155
+
156
+
def test_contract_fields_referenced_in_js():
157
+
js_source = JS_PATH.read_text()
158
+
159
+
for _, js_ref in CONTRACT_FIELDS:
160
+
assert (
161
+
js_ref in js_source
162
+
), f"{js_ref} not found in dashboard.js — contract field may be stale"
163
+
164
+
165
+
def test_all_day_fields_have_nonzero_values(tmp_path, monkeypatch):
166
+
stats_mod = importlib.import_module("think.journal_stats")
167
+
schema_mod = importlib.import_module("think.stats_schema")
168
+
journal = _build_journal(tmp_path)
169
+
monkeypatch.setenv("_SOLSTONE_JOURNAL_OVERRIDE", str(journal))
170
+
171
+
output = _scan_output(journal, stats_mod)
172
+
day_entry = next(iter(output["days"].values()))
173
+
174
+
for field in schema_mod.DAY_FIELDS:
175
+
assert day_entry[field] > 0, f"{field} should be non-zero in fixture output"
176
+
177
+
178
+
def test_schema_rejects_missing_required_key():
179
+
schema_mod = importlib.import_module("think.stats_schema")
180
+
output = {
181
+
"schema_version": schema_mod.SCHEMA_VERSION,
182
+
"generated_at": "2026-04-10T00:00:00+00:00",
183
+
"day_count": 1,
184
+
"days": {},
185
+
"totals": {},
186
+
"heatmap": [],
187
+
"tokens": {},
188
+
"agents": {},
189
+
"facets": {},
190
+
}
191
+
del output["totals"]
192
+
193
+
errors = schema_mod.validate(output)
194
+
195
+
assert errors, "validate() should reject missing required keys"
196
+
assert any("totals" in error for error in errors)
197
+
198
+
199
+
def test_schema_rejects_wrong_version():
200
+
schema_mod = importlib.import_module("think.stats_schema")
201
+
202
+
errors = schema_mod.validate(
203
+
{
204
+
"schema_version": 99,
205
+
"generated_at": "2026-04-10T00:00:00+00:00",
206
+
"day_count": 0,
207
+
"days": {},
208
+
"totals": {},
209
+
"heatmap": [],
210
+
"tokens": {},
211
+
"agents": {},
212
+
"facets": {},
213
+
}
214
+
)
215
+
216
+
assert errors, "validate() should reject the wrong schema version"
217
+
assert any("schema_version" in error for error in errors)