forked from CodebuffAI/codebuff
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathtop-freebuff-users.ts
More file actions
285 lines (248 loc) · 11.1 KB
/
Copy pathtop-freebuff-users.ts
File metadata and controls
285 lines (248 loc) · 11.1 KB
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
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
import { db } from '@codebuff/internal/db'
import * as schema from '@codebuff/internal/db/schema'
import { sql } from 'drizzle-orm'
interface UserStats {
userId: string
email: string | null
messageCount: number
totalCredits: number
totalCost: number
totalInputTokens: number
totalOutputTokens: number
totalCacheReadTokens: number
cacheHitRate: number
daysActive: number
avgMessagesPerDay: number
maxMessagesInDay: number
firstMessage: string
lastMessage: string
hourlyDistribution: Map<number, number>
}
async function topFreebuffUsers() {
const hoursBack = parseInt(process.argv[2] || '168') // default 1 week
const limit = parseInt(process.argv[3] || '50')
const agentId = process.argv[4] || 'base2-free' // configurable agent ID
const cutoff = new Date(Date.now() - hoursBack * 60 * 60 * 1000)
const excludeAgents = ['base2', 'base2-max']
console.log(`\n${'='.repeat(100)}`)
console.log(` TOP FREEBUFF USERS - DETAILED STATS (last ${hoursBack} hours)`)
console.log(` Agent: ${agentId}`)
console.log(` Since: ${cutoff.toISOString()}`)
console.log(` Excluding: ${excludeAgents.join(', ')}`)
console.log(`${'='.repeat(100)}\n`)
// Get all base2-free messages in the period (excluding users with base2/base2-max)
const results = await db
.select({
userId: schema.message.user_id,
email: schema.user.email,
messageCount: sql<number>`COUNT(*)`,
totalCredits: sql<number>`COALESCE(SUM(${schema.message.credits}), 0)`,
totalCost: sql<number>`COALESCE(SUM(${schema.message.cost}), 0)`,
totalInputTokens: sql<number>`COALESCE(SUM(${schema.message.input_tokens}), 0)`,
totalOutputTokens: sql<number>`COALESCE(SUM(${schema.message.output_tokens}), 0)`,
totalCacheReadTokens: sql<number>`COALESCE(SUM(${schema.message.cache_read_input_tokens}), 0)`,
firstMessage: sql<string>`MIN(${schema.message.finished_at})`,
lastMessage: sql<string>`MAX(${schema.message.finished_at})`,
})
.from(schema.message)
.leftJoin(schema.user, sql`${schema.message.user_id} = ${schema.user.id}`)
.where(
sql`${schema.message.finished_at} >= ${cutoff.toISOString()}
AND ${schema.message.agent_id} = ${agentId}
AND ${schema.message.user_id} NOT IN (
SELECT ${schema.message.user_id}
FROM ${schema.message}
WHERE ${schema.message.agent_id} IN (${sql.join(excludeAgents.map(a => sql`${a}`), sql`, `)})
AND ${schema.message.finished_at} >= ${cutoff.toISOString()}
)`,
)
.groupBy(schema.message.user_id, schema.user.email)
.orderBy(sql`COUNT(*) DESC`)
.limit(limit)
if (results.length === 0) {
console.log(`No ${agentId} messages found in this time range.`)
console.log('\nTip: Run with a different agent_id as the 4th argument, e.g.:')
console.log(' bun run scripts/top-freebuff-users.ts 168 50 claude-sonnet-4-20250514')
return
}
// Now run detailed queries since we have users
const userIds = results.map(r => r.userId).filter((id): id is string => !!id)
const dailyStats = await db
.select({
userId: schema.message.user_id,
date: sql<string>`DATE(${schema.message.finished_at})`,
count: sql<number>`COUNT(*)`,
})
.from(schema.message)
.where(
sql`${schema.message.finished_at} >= ${cutoff.toISOString()}
AND ${schema.message.agent_id} = ${agentId}
AND ${schema.message.user_id} IN (${sql.join(userIds.map(id => sql`${id}`), sql`, `)})`,
)
.groupBy(sql`DATE(${schema.message.finished_at})`, schema.message.user_id)
const hourlyStats = await db
.select({
userId: schema.message.user_id,
hour: sql<number>`EXTRACT(HOUR FROM ${schema.message.finished_at})`,
count: sql<number>`COUNT(*)`,
})
.from(schema.message)
.where(
sql`${schema.message.finished_at} >= ${cutoff.toISOString()}
AND ${schema.message.agent_id} = ${agentId}
AND ${schema.message.user_id} IN (${sql.join(userIds.map(id => sql`${id}`), sql`, `)})`,
)
.groupBy(sql`EXTRACT(HOUR FROM ${schema.message.finished_at})`, schema.message.user_id)
// Aggregate daily stats per user
const dailyByUser = new Map<string, { date: string; count: number }[]>()
for (const d of dailyStats) {
const uid = d.userId ?? ''
if (!dailyByUser.has(uid)) dailyByUser.set(uid, [])
dailyByUser.get(uid)!.push({ date: d.date ?? '', count: Number(d.count) })
}
// Aggregate hourly stats per user
const hourlyByUser = new Map<string, Map<number, number>>()
for (const h of hourlyStats) {
const hour = Number(h.hour)
const uid = h.userId ?? ''
if (!hourlyByUser.has(uid)) hourlyByUser.set(uid, new Map())
const hourMap = hourlyByUser.get(uid)!
hourMap.set(hour, (hourMap.get(hour) || 0) + Number(h.count))
}
// Build user stats objects
const userStats: UserStats[] = results.map(r => {
const uid = r.userId ?? ''
const daysData = dailyByUser.get(uid) || []
const hourMap = hourlyByUser.get(uid) || new Map()
const daysActive = daysData.length
const maxMessagesInDay = daysData.reduce((max, d) => Math.max(max, d.count), 0)
const avgMessagesPerDay = daysData.length > 0
? Math.round(daysData.reduce((sum, d) => sum + d.count, 0) / daysData.length)
: 0
const totalTokens = Number(r.totalInputTokens) + Number(r.totalOutputTokens)
const cacheReadTokens = Number(r.totalCacheReadTokens)
const cacheHitRate = totalTokens > 0 ? (cacheReadTokens / totalTokens) * 100 : 0
return {
userId: r.userId ?? 'unknown',
email: r.email,
messageCount: Number(r.messageCount),
totalCredits: Number(r.totalCredits),
totalCost: Number(r.totalCost),
totalInputTokens: Number(r.totalInputTokens),
totalOutputTokens: Number(r.totalOutputTokens),
totalCacheReadTokens: cacheReadTokens,
cacheHitRate: Math.round(cacheHitRate * 10) / 10,
daysActive,
avgMessagesPerDay,
maxMessagesInDay,
firstMessage: r.firstMessage ?? '',
lastMessage: r.lastMessage ?? '',
hourlyDistribution: hourMap,
}
})
// Print summary table
console.log(`${'#'.padStart(3)} ${'Email'.padEnd(35)} ${'Msgs'.padStart(7)} ${'Days'.padStart(5)} ${'Avg/Day'.padStart(8)} ${'Max/Day'.padStart(8)} ${'InTok'.padStart(9)} ${'OutTok'.padStart(9)} ${'Cache%'.padStart(7)} ${'Credits'.padStart(9)}`)
console.log(`${'='.repeat(105)}`)
let totalMessages = 0
let totalCredits = 0
let totalCost = 0
let totalInputTokens = 0
let totalOutputTokens = 0
for (let i = 0; i < userStats.length; i++) {
const u = userStats[i]
totalMessages += u.messageCount
totalCredits += u.totalCredits
totalCost += u.totalCost
totalInputTokens += u.totalInputTokens
totalOutputTokens += u.totalOutputTokens
const emailDisplay = (u.email ?? u.userId.slice(0, 8) + '...')
.slice(0, 33)
console.log(
`${String(i + 1).padStart(3)} ${emailDisplay.padEnd(35)} ${u.messageCount.toLocaleString().padStart(7)} ${u.daysActive.toString().padStart(5)} ${u.avgMessagesPerDay.toString().padStart(8)} ${u.maxMessagesInDay.toString().padStart(8)} ${u.totalInputTokens.toLocaleString().padStart(9)} ${u.totalOutputTokens.toLocaleString().padStart(9)} ${(u.cacheHitRate + '%').padStart(7)} ${u.totalCredits.toLocaleString().padStart(9)}`,
)
}
console.log(`${'='.repeat(105)}`)
console.log(
`\nTotal: ${userStats.length} users, ${totalMessages.toLocaleString()} messages, ${totalCredits.toLocaleString()} credits, $${totalCost.toFixed(2)}`,
)
console.log(`Tokens: ${totalInputTokens.toLocaleString()} in / ${totalOutputTokens.toLocaleString()} out\n`)
// Time distribution analysis - top 10 users by message count
console.log(`${'='.repeat(100)}`)
console.log(` TIME DISTRIBUTION ANALYSIS (Top 10 users)`)
console.log(`${'='.repeat(100)}\n`)
const top10 = userStats.slice(0, 10)
// Aggregate hourly distribution across top users
const overallHourly = new Map<number, number>()
for (const u of top10) {
for (const [hour, count] of u.hourlyDistribution) {
overallHourly.set(hour, (overallHourly.get(hour) || 0) + count)
}
}
// Sort by hour and display
const sortedHours = [...overallHourly.entries()].sort((a, b) => a[0] - b[0])
const maxHourCount = Math.max(...sortedHours.map(([_, c]) => c))
console.log('Hourly activity distribution (all top 10 users combined):')
console.log('')
for (const [hour, count] of sortedHours) {
const bar = '='.repeat(Math.round((count / maxHourCount) * 40))
const hourStr = hour.toString().padStart(2, '0') + ':00'
console.log(` ${hourStr} ${count.toString().padStart(5)} ${bar}`)
}
// Day of week analysis
const dayOfWeekStats = await db
.select({
dayOfWeek: sql<number>`EXTRACT(DOW FROM ${schema.message.finished_at})`,
count: sql<number>`COUNT(*)`,
})
.from(schema.message)
.where(
sql`${schema.message.finished_at} >= ${cutoff.toISOString()}
AND ${schema.message.agent_id} = ${agentId}
AND ${schema.message.user_id} IN (${sql.join(userIds.map(id => sql`${id}`), sql`, `)})`,
)
.groupBy(sql`EXTRACT(DOW FROM ${schema.message.finished_at})`)
const dayNames = ['Sun', 'Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat']
console.log('\nDay of week distribution:')
const sortedDays = dayOfWeekStats.sort((a, b) => Number(a.dayOfWeek) - Number(b.dayOfWeek))
const maxDayCount = Math.max(...sortedDays.map(d => Number(d.count)))
for (const d of sortedDays) {
const dayName = dayNames[Number(d.dayOfWeek)]
const count = Number(d.count)
const bar = '='.repeat(Math.round((count / maxDayCount) * 30))
console.log(` ${dayName} ${count.toString().padStart(5)} ${bar}`)
}
// Active days histogram
console.log('\nDays active histogram:')
const daysActiveCounts = new Map<number, number>()
for (const u of userStats) {
daysActiveCounts.set(u.daysActive, (daysActiveCounts.get(u.daysActive) || 0) + 1)
}
const sortedDaysActive = [...daysActiveCounts.entries()].sort((a, b) => a[0] - b[0])
const maxActiveUsers = Math.max(...sortedDaysActive.map(([_, c]) => c))
for (const [days, count] of sortedDaysActive) {
const bar = '='.repeat(Math.round((count / maxActiveUsers) * 40))
console.log(` ${days.toString().padStart(2)} days ${count.toString().padStart(3)} users ${bar}`)
}
// Session stats - users with highest avg messages per active day
console.log('\nTop 10 users by avg messages per active day:')
console.log(`${'Email'.padEnd(40)} ${'Days Active'.padStart(12)} ${'Avg/Day'.padStart(10)} ${'Max/Day'.padStart(10)}`)
console.log(`${'='.repeat(75)}`)
const byAvgPerDay = [...userStats]
.filter(u => u.daysActive > 0)
.sort((a, b) => b.avgMessagesPerDay - a.avgMessagesPerDay)
.slice(0, 10)
for (const u of byAvgPerDay) {
const emailDisplay = (u.email ?? u.userId.slice(0, 8) + '...')
.slice(0, 38)
console.log(
`${emailDisplay.padEnd(40)} ${u.daysActive.toString().padStart(12)} ${u.avgMessagesPerDay.toString().padStart(10)} ${u.maxMessagesInDay.toString().padStart(10)}`,
)
}
console.log('\n')
}
topFreebuffUsers()
.then(() => process.exit(0))
.catch((err) => {
console.error(err)
process.exit(1)
})