[web] Back to school user extraction script (#28416)

* fix: the script was not exiting when no users were found

* feat: scaffold a new script

* fix: make the script actually runnable

* refactor: exit at the same place

* feat: filter out subscriptions and without subscriptions

* fix: filter out groupies

* feat: from IDs to emails

docs: fix wording

refactor: will not need this script anymore

fix: don't use concurrency

feat: add id files

fix: remove the data files

fix: this indeed works!

* fix: review comments

* fix: prettier complaints

GitOrigin-RevId: 839ca850ff0df6972afafbfc25d4b89c10284dc6

services/web/scripts/extract_day1_churn_users.js

@@ -232,7 +232,7 @@ async function runScript() {
 
   if (churnUsers.length === 0) {
     console.log('No day 1 churn users found for the specified period')
-    return
+    process.exit(0)
   }
 
   console.log(`Writing ${churnUsers.length} users to ${args.outputPath}...`)

services/web/scripts/get_emails_by_ids.mjs

@@ -0,0 +1,130 @@
+import { scriptRunner } from './lib/ScriptRunner.mjs'
+import fs from 'node:fs'
+import readline from 'node:readline'
+import minimist from 'minimist'
+import { db, ObjectId } from '../app/src/infrastructure/mongodb.js'
+
+/**
+ * This script extracts user emails given a list of newline separated IDs
+ *
+ * Usage:
+ *   - Locally:
+ *     - docker compose exec web bash
+ *     - node scripts/get_emails_by_ids.mjs
+ *   - On the server:
+ *     - rake run:pod[staging,web]
+ *     - node scripts/get_emails_by_ids.mjs
+ *     - exit
+ *     - kubectl cp web-standalone-prod-XXXXX:/tmp/emails.txt ~/emails.txt
+ */
+
+function usage() {
+  console.log(
+    `
+    User email extraction, outputs to /tmp/emails.txt
+
+    Usage:
+      node scripts/get_emails_by_ids.js [--inputPath=<path>] [--outputPath=<path>] [--batchSize=<number>]
+
+    Options:
+    --help                     Show this screen
+
+    --inputPath=<path>         Input file path (default: ids.txt)
+
+    --outputPath=<path>        Output file path (default: /tmp/emails.txt)
+
+    --batchSize=<number>       Number of emails to be fetched in one query
+
+    Description:
+    This script extracts user emails given a list of newline separated IDs
+    `
+  )
+}
+
+function parseArgs() {
+  const argv = minimist(process.argv.slice(2), {
+    string: ['inputPath', 'outputPath'],
+    bool: ['help'],
+    number: ['batchSize'],
+    default: {
+      help: false,
+      inputPath: 'ids.txt',
+      outputPath: '/tmp/emails.txt',
+      batchSize: 1000,
+    },
+  })
+
+  if (argv.help) {
+    usage()
+    process.exit(0)
+  }
+
+  return argv
+}
+
+async function processBatch(idBatch, writeStream) {
+  try {
+    const cursor = db.users
+      .find({
+        _id: { $in: idBatch },
+      })
+      .project({
+        _id: 0,
+        email: 1,
+      })
+
+    for await (const doc of cursor) {
+      if (doc.email) {
+        writeStream.write(doc.email + '\n')
+      }
+    }
+  } catch (err) {
+    console.error('Error processing batch:', err)
+  }
+}
+
+async function main(trackProgress) {
+  const args = parseArgs()
+
+  const readStream = fs.createReadStream(args.inputPath)
+  const writeStream = fs.createWriteStream(args.outputPath)
+  const rl = readline.createInterface({
+    input: readStream,
+    crlfDelay: Infinity,
+  })
+
+  let idBatch = []
+
+  for await (const line of rl) {
+    const id = line.trim()
+    if (id) {
+      try {
+        idBatch.push(new ObjectId(id))
+      } catch (e) {
+        console.warn(`Skipping invalid ObjectId: ${id}`)
+      }
+    }
+
+    if (idBatch.length >= args.batchSize) {
+      await processBatch(idBatch, writeStream)
+      idBatch = []
+    }
+  }
+
+  if (idBatch.length > 0) {
+    await processBatch(idBatch, writeStream)
+  }
+
+  writeStream.end()
+
+  console.log(`✅ Success! Found emails written to ${args.outputPath}`)
+  await trackProgress('Job finished')
+}
+
+try {
+  await scriptRunner(main)
+  process.exit(0)
+} catch (error) {
+  console.error(error)
+  process.exit(1)
+}