We'd love for you to contribute to our source code and to make Webex Javascript SDK even better than it is today! If you would like to contribute to this repository by adding features, enhancements or bug fixes, you must follow our process:
- Let core members know about your proposal by posting a message in the contributor's Webex space
- A core member will review your proposal and if necessary may suggest to have a meeting to better understand your approach
- You are welcomed you join our weekly review meeting (Thursdays, 11:30a-12:30p PST) to propose your contribution as well
- If your proposal is approved you should start coding at this point
- We recommend opening a draft PR to receive feedback before finalizing your solution
- When opening a draft PR, specify with PR comments where in the code you would like to get feedback
- Before opening a PR ensure all PR guidelines are followed
- Let core members know about your PR by posting a message in the contributor's Webex space
- Core members will review the pull request and provide feedback when necessary
- If a PR is too large, you may be asked to break it down into multiple smaller-scoped PRs
- Once the PR is approved by a core member, it will be merged
- Celebrate! Your code is released 🎈🎉🍻
- Contributing
Please reach out to our developer support team for any issues you may be experiencings with the SDK.
Before you can build the Cisco Webex JS SDK, you will need the following dependencies:
- Node.js (LTS)
- We recommend using nvm (or nvs on Windows (not WSL)) to easily switch between Node.js versions
- Run
nvm useto set your node version to the one this package expects- If the required node version is not installed,
nvm/nvswill tell you the command needed to install it
- If the required node version is not installed,
- Install the latest npm to enable security audits using
npm install -g npm@latest
- Git
- node-gyp
- This is used during the dependency install process and is used to compile some native add-on modules
- Install with
npm install -g node-gyp- On Windows (not WSL), follow Option 1 on node-gyp's repo, as this will automatically install python 2 and neccessary build dependencies
- Python 2.7
- This is also used during the dependency install process
- Attempting to update dependencies with a Python 3.x environment will fail
We suggest using Windows Subsystem for Linux (WSL) for the best experience on Windows. We've seen multiple tooling/package related errors when trying to build on a regular Windows environment. The build errors are a result of third-party node_modules that we do not maintain. These issues are not seen when using WSL.
Ensure that you have followed all the steps outlined in Build Dependencies.
Fork the webex-js-sdk repository and git clone your fork:
git clone https://github.com/your-username/webex-js-sdk.gitInstall tooling dependencies with:
nvm use
npm installBuild the SDK:
npm run buildIf at any point your out-of-the-box builds or failing or if you are tests are failing with complaints of an invalid node version, the following commands will reset and rebuild everything:
nvm use; npm ciYou will need to create a file called .env that defines, at a minimum:
WEBEX_CLIENT_IDWEBEX_CLIENT_SECRETWEBEX_REDIRECT_URIWEBEX_SCOPE
You can get these values by registering a new integration on Cisco Webex for Developers.
The JS SDK allows you to customize your experience via configuration and environment variables. In general, external developers will not need to set any of the URL related environment variables.
| Environment Variable | Details | Default |
|---|---|---|
| ATLAS_SERVICE_URL | Used to populate the atlasServiceUrl pre discovery config | https://atlas-a.wbx2.com/admin/api/v1 |
| CLIENT_LOGS_SERVICE_URL | Used to populate the clientLogsServiceUrl pre discovery config | https://client-logs-a.wbx2.com/api/v1 |
| CONVERSATION_SERVICE | Used for validating an auth token | https://conv-a.wbx2.com/conversation/api/v1 |
| ENABLE_MERCURY_LOGGING | When set, will log all mercury messages | undefined |
| ENABLE_VERBOSE_NETWORK_LOGGING | Utilized to enable interceptor logging | undefined |
| ENCRYPTION_SERVICE_URL | Used for plugin-board tests | https://encryption-a.wbx2.com |
| HYDRA_SERVICE_URL | Stores the public hydra api url for managing Webex resources. | https://api.ciscospark.com/v1/ |
| IDBROKER_BASE_URL | Used throughout the SDK as the endpoint for authorization | https://idbroker.webex.com |
| IDENTITY_BASE_URL | Used to communicate with the identity api | https://identity.webex.com |
| MERCURY_FORCE_CLOSE_DELAY | Milliseconds to wait for a before declaring the socket dead | 2000 |
| MERCURY_PING_INTERVAL | Milliseconds between pings sent up the socket | 15000 |
| MERCURY_PONG_TIMEOUT | Milliseconds to wait for a pong before declaring the connection dead | 14000 |
| METRICS_SERVICE_URL | Used to populate the metricsServiceUrl pre discovery config | https://metrics-a.wbx2.com/metrics/api/v1 |
| U2C_SERVICE_URL | Stores the service catalog collecting url, typically the U2C service. | https://u2c.wbx2.com/u2c/api/v1 |
| WEBEX_ACCESS_TOKEN | Used to provide access token when using "webex/env" | undefined |
| WEBEX_AUTHORIZE_URL | Populdates the Authorization URL which prompts for user's password. | https://idbroker.webex.com/idb/oauth2/v1/authorize |
| WEBEX_AUTHORIZATION_STRING | This is the authorization url for the integration from Cisco Webex for Developers | undefined |
| WEBEX_CLIENT_ID | The Webex client ID used to authorize | undefined |
| WEBEX_CLIENT_SECRET | The Webex client secret used to authorize | undefined |
| WEBEX_CONVERSATION_CLUSTER_SERVICE | Service identifier used to lookup conversation servers in hostmap | identityLookup |
| WEBEX_CONVERSATION_DEFAULT_CLUSTER | Cluster used to convert from "us" cluster to actual cluster | urn:TEAM:us-east-2_a:identityLookup |
| WEBEX_LOG_LEVEL | Maximum log level that should be printed to the console. | log |
| WEBEX_REDIRECT_URI | The URI to redirect to after authorization | undefined |
| WEBEX_SCOPE | The Webex scope the users will authorize with | undefined |
| WDM_SERVICE_URL | The WDM service url before the catalog is downloaded | https://wdm-a.wbx2.com/wdm/api/v1 |
| WHISTLER_API_SERVICE_URL | The url to the whistler test service | https://whistler-prod.allnint.ciscospark.com/api/v1 |
npm test is the entrypoint to our test runner, but its not practical to use without parameters; the full suite would take over two hours to run and cross talk would probably cause tests to break each other.
Get the full test-runner docs via
npm test -- --help.
A local development flow might look like
- Edit source code in
MYPACKAGE. - Use
npm run buildto build all packages . - Use
npm test -- --packages @webex/MYPACKAGE --nodeto run the tests for just that package only in nodejs (Usually, we don't need to test both in node and the browser during development). - Repeats steps 1-3 until the tests pass.
npm run build is a bit tedious when making lots of changes, so instead, we can use npm run distsrc to point each package's main entry at the raw src and let babel compile on the fly.
- At the start of development, run
npm run distsrconce. - Edit source code in
MYPACKAGE. - Use
npm test -- --packages @webex/MYPACKAGE --nodeto run the tests for just that package only in nodejs. - Optionally, add environment variables to mimize logging and show any test specific logging, ie:
- WEBEX_LOG_LEVEL - set this to "log" to minimize the default verbose output
- DEBUG - if your test source includes the debug package set this to the appropriate string to enable debug output For exampe if you want to run only the plugin-messages test, and see the package specific logging, your command line would be:
WEBEX_LOG_LEVEL=log DEBUG=messages npm test -- --packages @webex/plugin-messages --node - Repeat steps 2-3 until the tests pass.
If you use VS Code, we've created a configuration to utilize the built-in debugger
- Set breakpoints within the package you're working on
- Select the
Test packageconfiguration - Enter the package you'd like to test (i.e.
MYPACKAGE)- The configuration already prepends
@webex/for you unlike the cli command, so justplugin-teamsis fine
- The configuration already prepends
- Add any optional flags (i.e.
--node)- If you don't want to add any flags, just add a space (current workaround)
- Run
npm run srcdistto restore the package.jsons to avoid committing those changes.
You can use the --unit, --integration, --automation, and --documentation switches to control what types of tests you run and --node and --browser to control which environments your tests run in.
The --packages flags will allow you to test multiple packages in one command instead of separate commands for each package --packages @webex/plugin-meetings @webex/plugin-rooms @webex/plugin-teams. Packages are still tested synchronously to allow for proper output to the terminal.
--browser --karma-debug will run the browser tests with {singleRun: false}, thus allowing automatic rerunning everytime you save a file (though, karma does eventually get confused and you need to interrupt and restart the command).
You can use the --browsers (not to be confused with the --browser tag) allows you to specify which browsers you want to run locally. This is restricted to what browsers are installed and available to you on your OS.
The default browsers that launch are Headless version of Firefox and Chrome, so --browsers Chrome Edge will only launch a normal version of Chrome along with Edge. If you add defaults to the browsers flag, it will also launch ChromeHeadless and FirefoxHeadless along with other browsers you've specified. All browsers include flags to enable WebRTC features and permissions.
To run tests on SauceLabs locally, you'll need to add a inline environment variable, SAUCE=true. Like mentioned above you can specify which browsers you'd like to test on with the --browers flag, but with SauceLabs service available to you, you can also specify which OS you'd like to test on. With the --os flag you have the option on testing on Mac and Windows. You can filter down the browsers that get launched by using the --browsers flag, so if you use --os Windows --browsers Edge IE it will launch only Edge and IE. Specifying just --browsers with SAUCE=true will launch that browsers in all available OSs, so --browsers Firefox will launch Firefox in Mac and Windows.
The default SauceLabs configuration "
SAUCE=true npm run test" is the latest versions ofChromeandFirefoxon bothMacandWindows, along withEdgeandIE 11on Windows, andSafarion Mac
--os Macwill launchChrome,Firefox, andSafari
--os Windowswill launchChrome,Firefox,Edge, andIE 11
--os LinuxWILL NEED--browsers Firefoxas SauceLabs only supportsFirefox 45for Linux. This is why it's also not included by default and requires two flags
See more scripts at SCRIPTS.md to learn how to run tests and more.
git clone git@github.com:webex/webex-js-sdk.git
cd webex-js-sdk
npm install
npm run build
npm run samples:serveNOTE: This installs all of the SDK's tooling dependencies, so you'll need
libgcryptand (possibly)graphicsmagick.
Mac
- You can install these with
brew install graphicsmagick libgcrypt.Ubuntu
- You can install these with
sudo apt-get install graphicsmagick libgcrypt-devWindows
- You can install
graphicsmagickusing either scoop or Chocolatey
- scoop:
scoop install graphicsmagick- chocolotey:
choco install graphicsmagick- Also globally install
win-node-envto resolveNODE_ENVwindows related command issues
Head to https://localhost:8000/ to use the samples
NOTE: The samples use a self-signed certificate generated by Webpack Dev Server. Some browsers such as Google Chrome (or any other Chromium-based browser) may not allow access by default for security reasons. You may need to configure browser settings to allow these at your own risk by enabling the
chrome://flags/#allow-insecure-localhostflag.
The samples tests are run by https://webdriver.io which spins up two browser instances and has them communicate between each other.
These tests are run with npm run samples:test.
We have found that due to the h.264 codec downloading in Firefox, the best way to run these test is on SauceLabs.
You can run them on SauceLabs with SAUCE=true npm run samples:test.
To run a specific sample test instead of the full suite, append the --spec flag to the samples:test command and the path to the specific test
npm run samples:test -- --spec packages/node_modules/samples/browser-call-with-screenshareIf an error occurs when running the above command that appears to be related to a missing Selenium driver, the following command should install the needed external dependencies:
./node_modules/.bin/selenium-standalone installIf you wish to run the samples tests locally, we suggest changing from the Chrome-to-Firefox multiremote setup to Chrome-to-Chrome.
You can do so by modifying the wdio.conf.js file.
Simply change the browserFirefox's capabilities object to the same as browserChrome (the Chrome instance).
When you run, you should see two instances of Chrome open.
We follow the (Conventional Commits)[https://www.conventionalcommits.org/] specification when writing commits and are run/linted through conventional changelog to generate the changelog. Please adhere to the following guidelines when formatting your commit messages.
Each commit message consists of a header, a body and a footer. The header has a special format that includes a type, a scope and a subject:
<type>(<scope>): <subject>
<BLANK LINE>
<body>
<BLANK LINE>
<footer>
The header is mandatory and the scope of the header is optional.
Any line of the commit message cannot be longer 100 characters! This allows the message to be easier to read on GitHub as well as in various git tools.
If the commit reverts a previous commit, it should begin with revert:, followed by the header of the reverted commit. In the body it should say: This reverts commit <hash>., where the hash is the SHA of the commit being reverted.
Examples can be found on the (Conventional Commits website)[https://www.conventionalcommits.org/en/v1.0.0/#examples]
The following types will cause a version bump:
- fix: Patches a bug in the code and directly corresponds to the PATCH
- perf: A code change that improves performance and corresponds to the PATCH
- feat: Describes a new feature and corresponds to the MINOR
- BREAKING CHANGE: a commit that has a footer
BREAKING CHANGE:, or appends a!after the type/scope, introduces a breaking API change (correlating with MAJOR in semantic versioning).
Appending a
!and/or aBREAKING CHANGE:footer to ANY TYPE will denote a BREAKING CHANGE and will cause a MAJOR bump
The following types will not cause a version bump:
- build: Changes that affect the build system or external dependencies
- ci: Changes to our CI configuration files and scripts
- docs: Documentation only changes
- refactor: A code change that neither fixes a bug, adds a feature, nor changes affecting the public API and corresponds to the PATCH
- style: Changes that do not affect the meaning of the code (white-space, formatting, missing semi-colons, etc)
- test: Adding missing tests or correcting existing tests
The scope should indicate what is being changed. Generally, these should match package names. For example, http-core, common, webex, etc. Other than package names, tooling tends to be the most common.
The subject contains succinct description of the change:
- use the imperative, present tense: "change" not "changed" nor "changes"
- don't capitalize first letter
- no dot (.) at the end
Just as in the subject the imperative, present tense: "change" not "changed" nor "changes". The body should include the motivation for the change and contrast this with previous behavior.
The footer should contain any information about Breaking changes and is also the place to reference GitHub issues that this commit closes.
Breaking Changes should start with the word BREAKING CHANGE: with a space or two newlines. The rest of the commit message is then used for this.
These are commit messages that will have an impact on how the build pipeline behaves. They are not to be used without prior approval.
All of these commit messages should include an explanation for why you're using them. You'll need to commit with -n or --no-verify to bypass the commit message linter.
For example
git commit -m "docs(webex-core): [skip npm] - docs change" --no-verify
This will run through the all the Github Checks, but will skip any version bumping, tagging, and subsequent publishing to npm after a pull request is merged.
This will skip the CircleCI pipeline entirely.
Prior to developing a new feature, be sure to search the Pull Requests for your idea to ensure you're not creating a duplicate change. Then, create a development branch in your forked repository for your idea and start coding!
When you're ready to submit your change, first check that new commits haven't been made in the upstream's master branch. If there are new commits, rebase your development branch to ensure a fast-forward merge when your Pull Request is approved:
# Fetch upstream master and update your local master branch
git fetch upstream
git checkout master
git merge upstream/master
# Rebase your development branch
git checkout feature
git rebase masterFinally, open a new Pull Request with your changes. Be sure to mention the issues this request addresses in the body of the request. Once your request is opened, a developer will review, comment, and, when approved, merge your changes!
Before you open that new pull request, make sure to have completed the following checklist:
- Code follows the style guidelines of this project
- I have performed a self-review of my own code
- I have commented my code, particularly in hard-to-understand areas
- I have made corresponding changes to the documentation
- My changes generate no new warnings
- I have added tests that prove my fix is effective or that my feature works
- New and existing unit tests pass locally with my changes
- Any dependent changes have been merged and published in downstream modules
To compile the documentation locally, make sure you have Bundler or Jekyll installed then run the following:
cd docs
bundle installcd docs
npm run build:docs
bundle exec jekyll serve --config=_config.yml,_config.local.yml