Merge pull request #1819 from 18F/dependabot/npm_and_yarn/puppeteer-2…

…3.6.0

Bump puppeteer from 19.11.1 to 23.6.0

docker-compose.yml

@@ -1,6 +1,6 @@
 version: "3.0"
 services:
-  app:
+  application:
     build: .
     volumes:
       - ./tock:/tock
@@ -28,7 +28,7 @@ services:
       - POSTGRES_PASSWORD=tock_password
 
   gulp:
-    image: node:16.15.0
+    image: node:23.0.0
     volumes:
       - .:/tock
       - /tock/node_modules
@@ -40,11 +40,11 @@ services:
       context: .
       dockerfile: Dockerfile-tests
     links:
-      - app
+      - application
     volumes:
       - .:/home/circleci/project
       - /home/circleci/project/node_modules
     environment: 
       IS_DOCKER: "true"
-      TOCK_URL: "http://app:8000"
+      TOCK_URL: "http://application:8000"
     command: /bin/true

docs/dependency-management.md

@@ -14,10 +14,10 @@ With pipenv installed locally, you can update development and production depende
 pipenv update --dev
 ```
 
-Within the `app` docker container, the approach is slightly different due to the volumes and working_dir set in `docker-compose.yml`.
+Within the `application` docker container, the approach is slightly different due to the volumes and working_dir set in `docker-compose.yml`.
 
 ```sh
-docker-compose run app bash
+docker-compose run application bash
 $ pipenv update --dev
 $ cp ../Pipfile.lock ./Pipfile.lock
 # Exit docker container

docs/deployment-process.md

@@ -130,6 +130,15 @@ cloud.gov UAA application for its users.
 Tock uses the cloud.gov service account service to provide deployer accounts for
 staging and production environments.
 
+### New Relic configuration
+
+Basic New Relic configuration is done in [newrelic.ini](../newrelic.ini), with additional settings
+specified via environment variables in each deployment environment's manifest file.
+
+As described in [Environment variables](#cloud-foundry-environment-variables), you will need
+to supply the `NEW_RELIC_LICENSE_KEY` as part of each deployment's
+[user-provided service](#user-provided-service-ups).
+
 ### Code review
 
 Submissions to the Tock codebase are made via GitHub, and are only accepted into the main
@@ -154,16 +163,47 @@ and checks on security flaws of Tock's dependencies.
 Tock uses CircleCI to continuously integrate code, deliver the code to staging
 servers, and deploy the latest release to production servers.
 
-Occasionally CircleCI builds will fail with an error like: `FileNotFoundError: [Errno 2] No such file or directory: '/home/circleci/project/.venv/bin/python'.` In this case, it is necessary to modify the `CACHE_VERSION` in the Environment Variables section in the CircleCI Tock Project Settings. (The exact value does not matter, just that the value is changed: this will force new cache dependencies to be built.)
+#### Job output
 
-### New Relic environment variables
+For each job in a CircleCI workflow, you can view the output of each step in the CircleCI:
 
-Basic New Relic configuration is done in [newrelic.ini](../newrelic.ini), with
-additional settings specified in each deployment environment's manifest file.
+- Navigate to the CircleCI project dashboard for this organization.
+- Click "Projects" in the left nav, then click "tock".
+- On the project page, click the "Workflow" link for the workflow run you're interested in. For a PR build, the workflow is named "build_pull_requests".
+- Click on the job (for example, "build") that you want to view.
+- On the job page, expand the accordion for each step to view its output.
+
+#### Enable verbose logging
+
+To troubleshoot issues within a CircleCI workflow, it may be helpful to configure the project to log more verbose output.
+
+##### Jest and Puppeteer test configuration
+
+In the [jest-puppeteer.config.js](../jest-puppeteer.config.js) file's `module.exports` -> `launch` section:
+
+- Add `'--enable-logging', '--v=1'` to the `args` array
+- To log Chrome driver messages to console, add `dumpio: true`
+
+Within `*.test.js` test files, use `console.log()` calls to output debugging information.
+
+##### CircleCI jobs
+
+In [`.circleci/config.yml`](../.circleci/config.yml), add additional steps to output information about the Docker container environment. For example:
+
+```yml
+# Add this to jobs -> <job name> -> steps
+- run:
+    name: Report Python, Node, and Chrome versions
+    command: |
+      python --version
+      node --version
+      google-chrome --version
+```
+
+#### Update the CircleCI project cache
+
+Occasionally CircleCI builds will fail with an error like: `FileNotFoundError: [Errno 2] No such file or directory: '/home/circleci/project/.venv/bin/python'.` In this case, it is necessary to modify the `CACHE_VERSION` in the Environment Variables section in the CircleCI Tock Project Settings. (The exact value does not matter, just that the value is changed: this will force new cache dependencies to be built.)
 
-As described in [Environment variables](#cloud-foundry-environment-variables), you will need
-to supply the `NEW_RELIC_LICENSE_KEY` as part of each deployment's
-[user-provided service](#user-provided-service-ups).
 
 ### Staging server
 

docs/local-development.md

@@ -14,9 +14,9 @@
 
   ```shell
   docker-compose build
-  docker-compose run app python manage.py migrate
-  docker-compose run app python manage.py loaddata test_data/data-update-deduped.json
-  docker-compose run app \
+  docker-compose run application python manage.py migrate
+  docker-compose run application python manage.py loaddata test_data/data-update-deduped.json
+  docker-compose run application \
     python manage.py \
         createsuperuser \
         --username admin.user \
@@ -48,20 +48,20 @@ security-focused linter.
 
 To run these tools locally, run:
 ```sh
-docker-compose run app bandit -r . -x manage.py,docker_entrypoint.py
-docker-compose run app flake8
+docker-compose run application bandit -r . -x manage.py,docker_entrypoint.py
+docker-compose run application flake8
 ```
 
-### Accessing the app container
+### Accessing the application container
 
 You'll likely want to run `manage.py` to do other things at some point.
 To do this, it's probably easiest to run:
 
 ```
-docker-compose run app bash
+docker-compose run application bash
 ```
 
-This will run an interactive bash session inside the main app container.
+This will run an interactive bash session inside the main application container.
 In this container, the `/tock` directory is mapped to the `tock`
 directory of the repository on your host; you can run `manage.py` from there.
 
@@ -92,7 +92,7 @@ The easiest, most reliable way to test locally is from within the docker contain
 which lets you access `manage.py`:
 
 ```
-docker-compose run app bash
+docker-compose run application bash
 python manage.py test
 ```
 

jest-puppeteer.config.js

@@ -1,5 +1,5 @@
 // These arguments are what jest-puppeteer specifies when running in CI
-// (see https://github.com/smooth-code/jest-puppeteer/blob/master/packages/jest-environment-puppeteer/src/readConfig.js#L14-L24)
+// (see https://github.com/argos-ci/jest-puppeteer/blob/main/packages/jest-environment-puppeteer/src/config.ts#L55-L63)
 const DOCKER_LAUNCH_ARGS = [
   "--no-sandbox",
   "--disable-setuid-sandbox",
@@ -16,7 +16,7 @@ module.exports = {
 
     // When running outside Docker, set `headless: false` to show the browser while tests run.
     devtools: false,
-    headless: true,
+    headless: 'shell', // see: https://github.com/puppeteer/puppeteer/issues/10091#issuecomment-2273275094
   },
   browserContext: "incognito",
 };

jest.config.js

@@ -6,4 +6,5 @@ module.exports = {
   notify: true,
   preset: 'jest-puppeteer',
   testMatch: ['**/*.test.js'],
+  testTimeout: 30000,
 };