PR Time Tracker

PR Time Tracker automates invoice generation processes and tracks core developer activity related to PRs and issues.

Contributing

If you want to contribute, please follow Holdex Developer Guidelines.

Possible issues docs

1. Cannot submit time: see docs

Installing application

1. Access the App public page
2. Install the App under your organization
3. Invite the @pr-time-tracker into your organization and make him an owner
4. To customize the included repositories, go to "Settings" -> "GitHub Apps"

Connect the installed organization with the Tracker flow

1. Add a record of your organization info in the Vercel env APP_INTEGRATIONS_LIST

  { "id": "org_id", "name": "org_slug", "nodeId": "oracle_project_id"}

Where:

• id is the identifier used for defining jobs (can be any string)
• name - the GitHub organization slug
• nodeId - Holdex Oracle project ID, if no project is defined, use explicit value 337c06eb for HX project column

2. Re-deploy the current production build so changes can take effect.

Scripts

• postinstall: Sets up Husky for Git hooks.
• pre-dev: Executes a script to pull environment variables before starting development.
• dev: Starts the development server on port 3000 after executing the pre-dev script.
• build: Builds the project.
• preview: Previews the build.
• format: Formats the code using Prettier with plugins for Svelte and Pug.
• lint: Lints the codebase using ESLint with support for JavaScript, TypeScript, Svelte, and CommonJS. check: Syncs Svelte kit and checks the TypeScript configuration.
• type-check: type checks the TypeScript files without emitting any output.
• check:watch: Watches for changes and syncs Svelte kit while checking TypeScript.
• proxy: Sets up ngrok to proxy requests to the development server with a specific domain.

Local Development Setup

Requirements

1. git https://git-scm.com/downloads
2. docker https://docs.docker.com/
3. nodejs https://nodejs.org/en/download/package-manager
4. pnpm https://pnpm.io/installation

Steps and Setups

1. Clone this repository

2. Make GitHub Access Token for installing some of the hosted dependencies:

   • Create a GitHub Classic Access Token with the necessary permissions https://github.com/settings/tokens.

     Github access token with `read:packages` access permission

     

3. Copy .npmrc.example to .npmrc in the root folder and replace the NPM_TOKEN with your generated GitHub access token.

4. Install the dependencies pnpm install

   You should have this if you setup your token properly with proper permission

   

5. Setup Telebit https://telebit.cloud/

   Following the telebit setup you will have these, take note of your own unique telebit address.

   

   

6. Setup your Supabase Project for Trigger.dev v2 based on this guide: https://v2docs.trigger.dev/documentation/guides/self-hosting/supabase (Optional)

   You can skip this step if you don't need to run your own supabase for Trigger.dev v2. We provide a shared supabase project for development purpose that you can use.

   But for some reason you have a problem when running with the shared supabase project, you can follow this guide to create your own supabase project.

   • You can follow this setup guide https://v2docs.trigger.dev/documentation/guides/self-hosting/supabase#create-a-supabase-db

   • Create your supabase project for trigger.dev

     When creating the project, take note of your Supabase DB uri and password.

     

     

   • Go to the /container folder

   • Duplicate the .env.example to .env

   • In the .env file change the DATABASE_URL and DIRECT_URL you created in the Supabase DB creation step (don't forget to adjust with your own password).

       # From this
       DATABASE_URL=postgresql://postgres.phtaofoaormwpxwrwtkw:[email protected]:6543/postgres?schema=triggerdotdev&pgbouncer=true
     
       DIRECT_URL=postgresql://postgres.phtaofoaormwpxwrwtkw:[email protected]:5432/postgres?schema=triggerdotdev
     
       # To this (EXAMPLE)
       DATABASE_URL=postgresql://postgres.arnehavoxpuewazwvpph:[email protected]:6543/postgres?schema=triggerdotdev&pgbouncer=true
     
       DIRECT_URL=postgresql://postgres.arnehavoxpuewazwvpph:[email protected]:5432/postgres?schema=triggerdotdev
     

7. Setup Local Trigger.dev v2

   • Start proxy server for localhost:3030 to Telebit pnpm trigger-dev:proxy

     

   • Go to the /container folder

   • Duplicate the .env.example to .env

   • In the .env file change the LOGIN_ORIGIN and API_ORIGIN with the forwarding URL from telebit above, e.g:

       # From this
       LOGIN_ORIGIN=https://alert-seemingly-moccasin.telebit.io
       APP_ORIGIN=https://alert-seemingly-moccasin.telebit.io
     
       # To this (EXAMPLE)
       LOGIN_ORIGIN=https://shaggy-termite-71.telebit.io
       APP_ORIGIN=https://shaggy-termite-71.telebit.io

   • Run docker compose up within the container. Make sure you don't have running redis that use port 0.0.0.0:6379. If you fail to run this see Possible issues on the bottom.

     You should see this successful docker compose

     

   • You should be able to access your local Trigger.dev using the forwarding URL from pnpm trigger-dev:proxy above e.g: https://shaggy-termite-71.telebit.io/

     You should see this successful docker compose

     

   • Setup your local account and triggerdotdev project

     Register your email here

     

     You'll get a login url from docker container

     

     Copy and paste to browser to access, and you'll be redirected to your project.

     

   • Go to the root path and download the environment variables using pnpm pre-dev and you will see a new file .env on root

   • Take note of the Trigger.dev API Keys in Trigger dev UI (Environment & API Keys) and change the TRIGGER_API_KEY and TRIGGER_API_URL in the .env file, and the endpointId in your package.json file.

       # .env
       # API Key from triggerdotdev UI
       TRIGGER_API_KEY="tr_dev_YIHNyqRKaKYt7lJ7hwF2"
     
       # URL from telebit forwarding proxy for accessing local triggerdotdev
       TRIGGER_API_URL="https://shaggy-termite-71.telebit.io"
     
       # slug for your created project in triggerdotdev (you can find this in the URL)
       TRIGGER_PROJECT_ID="pr-time-tracker-lmz0"

     // package.json
     {
       // ...
       "trigger.dev": {
         "endpointId": "pr-time-tracker-lmz0"
       }
     }

     

   • Run local dev of the project pnpm dev-only

   • Run ngrok proxy for the local dev project using pnpm proxy

     

   • In the Trigger dev UI (Environment & API Keys), set a development endpoint with the following URL: https://alert-seemingly-moccasin.ngrok-free.app/api/trigger

     

     

     

   • After the process is done, you should be able to see a new entry in Runs sidebar. Congrats! you have registered your local trigger.dev proxy url to github webhooks 🎉

     You should see this successful run

     

     

8. Database Connection:

   • Use MongoDB Compass to connect to the database using the MONGODB_URI environment variable.

Possible issues

1. Telebit can't run. You might find this issue when running pnpm trigger-dev:proxy.

    pnpm trigger-dev:proxy
   
   > [email protected] trigger-dev:proxy /home/ubuntu/Developer/pr-time-tracker
   > ~/telebit http 3030
   
   /home/ubuntu/Applications/telebit/bin/telebit-remote.js:643
     var verstrd = [ pkg.name + ' daemon v' + state.config.version ];
                                                           ^
   
   TypeError: Cannot read property 'version' of undefined
       at handleConfig (/home/ubuntu/Applications/telebit/bin/telebit-remote.js:643:57)
       at /home/ubuntu/Applications/telebit/bin/telebit-remote.js:359:22
       at /home/ubuntu/Applications/telebit/usr/share/install-launcher.js:241:26
       at ChildProcess.exithandler (child_process.js:280:7)
       at ChildProcess.emit (events.js:182:13)
       at maybeClose (internal/child_process.js:962:16)
       at Process.ChildProcess._handle.onexit (internal/child_process.js:251:5)
    ELIFECYCLE  Command failed with exit code 1.

   Try to reinstall telebit

   curl https://get.telebit.io/ | bash

   Then rerun pnpm trigger-dev:proxy

2. Trigger.dev local docker failed to run migration. You might face this issue when running docker compose up in /container folder.

   triggerdotdev  | 526 migrations found in prisma/migrations
   triggerdotdev  |
   triggerdotdev  | Error: P3009
   triggerdotdev  |
   triggerdotdev  | migrate found failed migrations in the target database, new migrations will not be applied. Read more about how to resolve migration issues in a production database: https://pris.ly/d/migrate-resolve
   triggerdotdev  | The `20230616104937_remove_api_identifier` migration started at 2024-11-12 08:37:52.191805 UTC failed
   triggerdotdev  |
   

   A fix mentioned in triggerdotdev/trigger.dev#1026

   Make new Supabase project and change the DATABASE_URL and DIRECT_URL accordingly

   Add this to the triggerdotdev container within docker-compose.yml

   # This is the important bit!
   command:
     - sh
     - -c
     - sed -i s/public/triggerdotdev/ ./packages/database/prisma/migrations/20240130165343_add_composite_index_to_job_run_for_job_id_and_created_at/migration.sql && ./scripts/entrypoint.sh

3. Redis can't find addrinfo. You might face this issue when running docker compose up in /container folder.

     triggerdotdev  | [ioredis] Unhandled error event: Error: getaddrinfo ENOTFOUND host.docker.internal
     triggerdotdev  |     at GetAddrInfoReqWrap.onlookup [as oncomplete] (node:dns:107:26)
     triggerdotdev  |     at GetAddrInfoReqWrap.callbackTrampoline (node:internal/async_hooks:130:17)

   You might need to change the REDIS_HOST="host.docker.internal" in the /container/.env to docker specific ip address e.g: REDIS_HOST=172.17.0.1

4. After setting a development endpoint, no new Runs entry made.

   Make sure the root `.env` `TRIGGER_API_KEY`, `TRIGGER_API_URL`, and `TRIGGER_PROJECT_ID` points to the correct values.

   If you have verified these settings but still don't see any new Runs entries, try creating a new project in your local trigger.dev and rerun the setup process using the new project data. You can also try to restart the whole docker container after changing the .env and package.json files.