Implement rudimentary version of dynamic Slack to GitHub ID username mapping. Part of #12

.env.example

@@ -9,4 +9,7 @@ WEBHOOK_SECRET=development
 WEBHOOK_PROXY_URL=https://smee.io/ZyQCjZTDPT3pd4SD
 
 # The "Bot User OAuth Access Token" of your Slack App
-SLACK_BOT_TOKEN=
+SLACK_BOT_TOKEN=
+
+# A "User Legacy Token" of your Slack App, used to access the "users.profile.get" API 
+SLACK_USER_TOKEN=

index.js

@@ -1,11 +1,21 @@
-module.exports = (robot) => {
+var MemCache = require('mem-cache')
+var SlackGitHubCacheBuilder = require('./lib/retrieve-slack-github-users')
+
+module.exports = async (robot) => {
   console.log('Yay, the app was loaded!')
 
+  var slackGitHubCache = new MemCache({ timeoutDisabled: true })
+  var slackCachePromise = SlackGitHubCacheBuilder.build(robot, slackGitHubCache)
+
   require('./scripts/assign-new-pr-to-review')(robot)
   require('./scripts/assign-approved-pr-to-test')(robot)
   require('./scripts/assign-to-bounty-awaiting-for-approval')(robot)
   require('./scripts/greet-new-contributor')(robot)
 
+  await slackCachePromise
+
+  // Add scripts which require using the Slack/GitHub cache after this comment
+
   // For more information on building apps:
   // https://probot.github.io/docs/
 

lib/retrieve-slack-github-users.js

@@ -0,0 +1,124 @@
+// Description:
+//   Configuration-related functionality
+//
+// Dependencies:
+//   "mem-cache": "0.0.5"
+//
+// Author:
+//   PombeirP
+
+const { WebClient } = require('@slack/client')
+const token = process.env.SLACK_USER_TOKEN || ''
+
+module.exports.build = async (robot, cache) => {
+  const web = new WebClient(token)
+  await populateCache(robot, web, cache)
+}
+
+module.exports.getGitHubIdFromSlackUsername = (slackUsername, cache) => {
+  return cache.get(getSlackCacheKeyName(slackUsername))
+}
+
+module.exports.getSlackUsernameFromGitHubId = (gitHubId, cache) => {
+  return cache.get(getGitHubCacheKeyName(gitHubId))
+}
+
+async function populateCache (robot, web, cache) {
+  robot.log.info('Populating Slack username cache...')
+
+  try {
+    const usersList = await web.users.list() // TODO: This call should be paginated to avoid hitting limits (memory, API): https://api.slack.com/docs/pagination#cursors
+    const activeUsersList = usersList.members.filter(u => !u.deleted && !u.is_bot && u.id !== 'USLACKBOT')
+
+    let gitHubFieldId = null
+    let usersMissingGitHubInfo = []
+    let usersContainingGitHubInfo = []
+    let rateLimitWait = 10000
+    let profileFetchPreviousBatchCount = 3
+    let profileFetchBatchCount = 0
+    for (let i = 0; i < activeUsersList.length;) {
+      const user = activeUsersList[i]
+
+      try {
+        ++profileFetchBatchCount
+        const { profile } = await web.users.profile.get({ user: user.id, include_labels: !gitHubFieldId })
+        const username = profile.display_name_normalized || profile.real_name_normalized
+
+        if (!gitHubFieldId) {
+          // Find the field ID for the field with the 'Github ID' label
+          gitHubFieldId = findGitHubLabelId(profile)
+        }
+
+        if (!gitHubFieldId) {
+          robot.log.warn(`No GitHub ID field found in profile (@${username})!`)
+          ++i
+          continue
+        }
+
+        if (profile.fields && profile.fields[gitHubFieldId]) {
+          const gitHubUsername = profile.fields[gitHubFieldId].value
+          robot.log.debug(`@${username} -> ${gitHubUsername}`)
+
+          cache.set(getSlackCacheKeyName(username), gitHubUsername)
+          cache.set(getGitHubCacheKeyName(gitHubUsername), username)
+          usersContainingGitHubInfo = usersContainingGitHubInfo.concat(username)
+        } else {
+          robot.log.warn(`@${username} (${user.id}) has no GitHub ID set`)
+          usersMissingGitHubInfo = usersMissingGitHubInfo.concat(username)
+        }
+
+        ++i
+        await sleep(1500)
+      } catch (e) {
+        if (e.name === 'Error' && e.message === 'ratelimited') {
+          robot.log.trace(`Rate-limited, waiting ${rateLimitWait / 1000}s...`)
+          await sleep(rateLimitWait)
+          if (profileFetchBatchCount < profileFetchPreviousBatchCount) {
+            // If we managed to fetch fewer profiles than the last time we got rate-limited, then try increasing the wait period
+            rateLimitWait += 5000
+          }
+          profileFetchPreviousBatchCount = profileFetchBatchCount
+          profileFetchBatchCount = 0
+          continue
+        }
+
+        throw e
+      }
+    }
+    robot.log.info(`Populated Slack username cache with ${usersContainingGitHubInfo.length} users: ${usersContainingGitHubInfo.map(s => '@' + s).join(', ')}`)
+    if (usersMissingGitHubInfo) {
+      robot.log.warn(`The following ${usersMissingGitHubInfo.length} Slack users have no GitHub info in their profiles: ${usersMissingGitHubInfo.map(s => '@' + s).join(', ')}`)
+    }
+  } catch (e) {
+    robot.log.error(`Error while populating Slack username cache: ${e}`)
+  }
+}
+
+function findGitHubLabelId (profile) {
+  if (profile.fields) {
+    for (const fieldId in profile.fields) {
+      const field = profile.fields[fieldId]
+      if (field.label === 'Github ID') {
+        return fieldId
+      }
+    }
+  }
+
+  return null
+}
+
+function getSlackCacheKeyName (slackUsername) {
+  return `Slack-${slackUsername}`
+}
+
+function getGitHubCacheKeyName (gitHubUsername) {
+  return `GitHub-${gitHubUsername}`
+}
+
+function timeout (ms) {
+  return new Promise(resolve => setTimeout(resolve, ms))
+}
+
+async function sleep (timeoutMs) {
+  await timeout(timeoutMs)
+}

package-lock.json

@@ -4542,6 +4542,11 @@
         "mimic-fn": "1.1.0"
       }
     },
+    "mem-cache": {
+      "version": "0.0.5",
+      "resolved": "https://registry.npmjs.org/mem-cache/-/mem-cache-0.0.5.tgz",
+      "integrity": "sha1-EUCWcbMQEXPY3PQgcK5nv1Dzd/8="
+    },
     "merge": {
       "version": "1.2.0",
       "resolved": "https://registry.npmjs.org/merge/-/merge-1.2.0.tgz",

package.json

@@ -11,6 +11,7 @@
   },
   "dependencies": {
     "eslint": "^4.16.0",
+    "mem-cache": "0.0.5",
     "probot": "^5.0.0",
     "probot-config": "^0.1.0",
     "probot-gpg-status": "^0.5.4",