Giter Club home page Giter Club logo

jquery-license's Introduction

jQuery โ€” New Wave JavaScript

Meetings are currently held on the matrix.org platform.

Meeting minutes can be found at meetings.jquery.org.

Contribution Guides

In the spirit of open source software development, jQuery always encourages community code contribution. To help you get started and before you jump into writing code, be sure to read these important contribution guidelines thoroughly:

  1. Getting Involved
  2. Core Style Guide
  3. Writing Code for jQuery Projects

References to issues/PRs

GitHub issues/PRs are usually referenced via gh-NUMBER, where NUMBER is the numerical ID of the issue/PR. You can find such an issue/PR under https://github.com/jquery/jquery/issues/NUMBER.

jQuery has used a different bug tracker - based on Trac - in the past, available under bugs.jquery.com. It is being kept in read only mode so that referring to past discussions is possible. When jQuery source references one of those issues, it uses the pattern trac-NUMBER, where NUMBER is the numerical ID of the issue. You can find such an issue under https://bugs.jquery.com/ticket/NUMBER.

Environments in which to use jQuery

  • Browser support
  • jQuery also supports Node, browser extensions, and other non-browser environments.

What you need to build your own jQuery

To build jQuery, you need to have the latest Node.js/npm and git 1.7 or later. Earlier versions might work, but are not supported.

For Windows, you have to download and install git and Node.js.

macOS users should install Homebrew. Once Homebrew is installed, run brew install git to install git, and brew install node to install Node.js.

Linux/BSD users should use their appropriate package managers to install git and Node.js, or build from source if you swing that way. Easy-peasy.

How to build your own jQuery

First, clone the jQuery git repo.

Then, enter the jquery directory, install dependencies, and run the build script:

cd jquery
npm install
npm run build

The built version of jQuery will be placed in the dist/ directory, along with a minified copy and associated map file.

Build all jQuery release files

To build all variants of jQuery, run the following command:

npm run build:all

This will create all of the variants that jQuery includes in a release, including jquery.js, jquery.slim.js, jquery.module.js, and jquery.slim.module.js along their associated minified files and sourcemaps.

jquery.module.js and jquery.slim.module.js are ECMAScript modules that export jQuery and $ as named exports are placed in the dist-module/ directory rather than the dist/ directory.

Building a Custom jQuery

The build script can be used to create a custom version of jQuery that includes only the modules you need.

Any module may be excluded except for core. When excluding selector, it is not removed but replaced with a small wrapper around native querySelectorAll (see below for more information).

Build Script Help

To see the full list of available options for the build script, run the following:

npm run build -- --help

Modules

To exclude a module, pass its path relative to the src folder (without the .js extension) to the --exclude option. When using the --include option, the default includes are dropped and a build is created with only those modules.

Some example modules that can be excluded or included are:

  • ajax: All AJAX functionality: $.ajax(), $.get(), $.post(), $.ajaxSetup(), .load(), transports, and ajax event shorthands such as .ajaxStart().

  • ajax/xhr: The XMLHTTPRequest AJAX transport only.

  • ajax/script: The <script> AJAX transport only; used to retrieve scripts.

  • ajax/jsonp: The JSONP AJAX transport only; depends on the ajax/script transport.

  • css: The .css() method. Also removes all modules depending on css (including effects, dimensions, and offset).

  • css/showHide: Non-animated .show(), .hide() and .toggle(); can be excluded if you use classes or explicit .css() calls to set the display property. Also removes the effects module.

  • deprecated: Methods documented as deprecated but not yet removed.

  • dimensions: The .width() and .height() methods, including inner- and outer- variations.

  • effects: The .animate() method and its shorthands such as .slideUp() or .hide("slow").

  • event: The .on() and .off() methods and all event functionality.

  • event/trigger: The .trigger() and .triggerHandler() methods.

  • offset: The .offset(), .position(), .offsetParent(), .scrollLeft(), and .scrollTop() methods.

  • wrap: The .wrap(), .wrapAll(), .wrapInner(), and .unwrap() methods.

  • core/ready: Exclude the ready module if you place your scripts at the end of the body. Any ready callbacks bound with jQuery() will simply be called immediately. However, jQuery(document).ready() will not be a function and .on("ready", ...) or similar will not be triggered.

  • deferred: Exclude jQuery.Deferred. This also excludes all modules that rely on Deferred, including ajax, effects, and queue, but replaces core/ready with core/ready-no-deferred.

  • exports/global: Exclude the attachment of global jQuery variables ($ and jQuery) to the window.

  • exports/amd: Exclude the AMD definition.

  • selector: The full jQuery selector engine. When this module is excluded, it is replaced with a rudimentary selector engine based on the browser's querySelectorAll method that does not support jQuery selector extensions or enhanced semantics. See the selector-native.js file for details.

Note: Excluding the full selector module will also exclude all jQuery selector extensions (such as effects/animatedSelector and css/hiddenVisibleSelectors).

AMD name

You can set the module name for jQuery's AMD definition. By default, it is set to "jquery", which plays nicely with plugins and third-party libraries, but there may be cases where you'd like to change this. Pass it to the --amd parameter:

npm run build -- --amd="custom-name"

Or, to define anonymously, leave the name blank.

npm run build -- --amd
File name and directory

The default name for the built jQuery file is jquery.js; it is placed under the dist/ directory. It's possible to change the file name using --filename and the directory using --dir. --dir is relative to the project root.

npm run build -- --slim --filename="jquery.slim.js" --dir="/tmp"

This would create a slim version of jQuery and place it under tmp/jquery.slim.js.

ECMAScript Module (ESM) mode

By default, jQuery generates a regular script JavaScript file. You can also generate an ECMAScript module exporting jQuery as the default export using the --esm parameter:

npm run build -- --filename=jquery.module.js --esm
Factory mode

By default, jQuery depends on a global window. For environments that don't have one, you can generate a factory build that exposes a function accepting window as a parameter that you can provide externally (see README of the published package for usage instructions). You can generate such a factory using the --factory parameter:

npm run build -- --filename=jquery.factory.js --factory

This option can be mixed with others like --esm or --slim:

npm run build -- --filename=jquery.factory.slim.module.js --factory --esm --slim --dir="/dist-module"

Custom Build Examples

Create a custom build using npm run build, listing the modules to be excluded. Excluding a top-level module also excludes its corresponding directory of modules.

Exclude all ajax functionality:

npm run build -- --exclude=ajax

Excluding css removes modules depending on CSS: effects, offset, dimensions.

npm run build -- --exclude=css

Exclude a bunch of modules (-e is an alias for --exclude):

npm run build -- -e ajax/jsonp -e css -e deprecated -e dimensions -e effects -e offset -e wrap

There is a special alias to generate a build with the same configuration as the official jQuery Slim build:

npm run build -- --filename=jquery.slim.js --slim

Or, to create the slim build as an esm module:

npm run build -- --filename=jquery.slim.module.js --slim --esm

Non-official custom builds are not regularly tested. Use them at your own risk.

Running the Unit Tests

Make sure you have the necessary dependencies:

npm install

Start npm start to auto-build jQuery as you work:

npm start

Run the unit tests with a local server that supports PHP. Ensure that you run the site from the root directory, not the "test" directory. No database is required. Pre-configured php local servers are available for Windows and Mac. Here are some options:

Essential Git

As the source code is handled by the Git version control system, it's useful to know some features used.

Cleaning

If you want to purge your working directory back to the status of upstream, the following commands can be used (remember everything you've worked on is gone after these):

git reset --hard upstream/main
git clean -fdx

Rebasing

For feature/topic branches, you should always use the --rebase flag to git pull, or if you are usually handling many temporary "to be in a github pull request" branches, run the following to automate this:

git config branch.autosetuprebase local

(see man git-config for more information)

Handling merge conflicts

If you're getting merge conflicts when merging, instead of editing the conflicted files manually, you can use the feature git mergetool. Even though the default tool xxdiff looks awful/old, it's rather useful.

The following are some commands that can be used there:

  • Ctrl + Alt + M - automerge as much as possible
  • b - jump to next merge conflict
  • s - change the order of the conflicted lines
  • u - undo a merge
  • left mouse button - mark a block to be the winner
  • middle mouse button - mark a line to be the winner
  • Ctrl + S - save
  • Ctrl + Q - quit

QUnit Reference

Test methods

expect( numAssertions );
stop();
start();

Note: QUnit's eventual addition of an argument to stop/start is ignored in this test suite so that start and stop can be passed as callbacks without worrying about their parameters.

Test assertions

ok( value, [message] );
equal( actual, expected, [message] );
notEqual( actual, expected, [message] );
deepEqual( actual, expected, [message] );
notDeepEqual( actual, expected, [message] );
strictEqual( actual, expected, [message] );
notStrictEqual( actual, expected, [message] );
throws( block, [expected], [message] );

Test Suite Convenience Methods Reference (See test/data/testinit.js)

Returns an array of elements with the given IDs

q( ... );

Example:

q("main", "foo", "bar");

=> [ div#main, span#foo, input#bar ]

Asserts that a selection matches the given IDs

t( testName, selector, [ "array", "of", "ids" ] );

Example:

t("Check for something", "//[a]", ["foo", "bar"]);

Fires a native DOM event without going through jQuery

fireNative( node, eventType )

Example:

fireNative( jQuery("#elem")[0], "click" );

Add random number to url to stop caching

url( "some/url" );

Example:

url("index.html");

=> "data/index.html?10538358428943"


url("mock.php?foo=bar");

=> "data/mock.php?foo=bar&10538358345554"

Run tests in an iframe

Some tests may require a document other than the standard test fixture, and these can be run in a separate iframe. The actual test code and assertions remain in jQuery's main test files; only the minimal test fixture markup and setup code should be placed in the iframe file.

testIframe( testName, fileName,
  function testCallback(
      assert, jQuery, window, document,
	  [ additional args ] ) {
	...
  } );

This loads a page, constructing a url with fileName "./data/" + fileName. The iframed page determines when the callback occurs in the test by including the "/test/data/iframeTest.js" script and calling startIframeTest( [ additional args ] ) when appropriate. Often this will be after either document ready or window.onload fires.

The testCallback receives the QUnit assert object created by testIframe for this test, followed by the global jQuery, window, and document from the iframe. If the iframe code passes any arguments to startIframeTest, they follow the document argument.

Questions?

If you have any questions, please feel free to ask on the Developing jQuery Core forum or in #jquery on libera.

jquery-license's People

Contributors

csnover avatar dmethvin avatar gnarf avatar jdalton avatar jzaefferer avatar kborchers avatar nzakas avatar scottgonzalez avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

jquery-license's Issues

./bin/audit-pr.js <repo> <pr> fails

Trying to manually check a PR again, it fails with this error:

Error auditing PR.
Error: Missing required option: owner.
    at new Audit (/Users/jzaefferer/dev/jquery/jquery-license/lib/pr.js:13:9)
    at exports.audit (/Users/jzaefferer/dev/jquery/jquery-license/lib/pr.js:254:11)
    at /Users/jzaefferer/dev/jquery/jquery-license/bin/audit-pr.js:18:10
    at $$$internal$$tryCatch (/Users/jzaefferer/dev/jquery/jquery-license/node_modules/es6-promise/dist/es6-promise.js:304:16)
    at $$$internal$$invokeCallback (/Users/jzaefferer/dev/jquery/jquery-license/node_modules/es6-promise/dist/es6-promise.js:316:17)
    at $$$internal$$publish (/Users/jzaefferer/dev/jquery/jquery-license/node_modules/es6-promise/dist/es6-promise.js:287:11)
    at $$asap$$flush (/Users/jzaefferer/dev/jquery/jquery-license/node_modules/es6-promise/dist/es6-promise.js:111:9)
    at nextTickCallbackWith0Args (node.js:433:9)
    at process._tickCallback (node.js:362:13)

Looks like the owner config wasn't integrated into this script.

Smarter labeling

Check the existing labels and only update if necessary. This cuts down on noise during synchronization.

Create a page to show the results for a PR

The results of PR audits should be stored in a viewable page. The page should be keyed off of the commit, not the PR, since that's where the data is actually stored. This page should then be set as the URL for the commit status.

One idea is to just store some data in a JSON file and have a page on contribute.jquery.org which reads the data and displays the data however we want so that we get the styling of jquery-wp-content.

Allow CLA check from commits sha

Today I fixed the commit author info on qunitjs/qunit#751 to make it pass on the CLA check. The commit wasn't originally mine and I figured out the only way to double check the CLA information was creating another PR.

I would be great to have anything to check the CLA through a commit SHA, even it's on a temp branch on the github project's repo.

from IRC, @scottgonzalez talked about a binary, which would also solve this problem:

$ jquery-audit-commit SHA

Fetch base branch for PRs

It's possible for the base commit to not exist in the PR branch if the base branch was updated before the PR was sent. This results in an invalid range since we're never fetching any non-PR branches after cloning. To fix this, we need to fetch the base branch in addition to the PR branch.

GitHub seems to be sending webhooks earlier than expected

We've had a few errors today where the PR branch doesn't exist when the webhook is received.

Error: Command failed: fatal: Couldn't find remote ref refs/pull/1459/head

I haven't seen this happen before today. Off the top of my head, I can think of two workarounds:

  • Introduce an arbitrary delay before processing the hook.
  • Implement delayed retries on our requests.

I'd prefer the delayed retries. We can wrap all requests in the same logic for some extra robustness.

Show time of attempted verification

From jquery/contribute.jquery.org#110 by gibson042:

For technical reasons, we don't automatically recheck pull requests when a contributor signs the CLA. But pages/CLA/status.md doesn't provide enough information on its own to detect post-failure resolution, because it lacks the invocation time.

We should add text to the effect of "The following authors didn't have appropriate CLA signatures as of [datetime]", and also instructions to reactivate the checker by pushing a new commit to each affected pull request (perhaps suggesting git commit --allow-empty -m '[CLA]: Recheck').

Better handling of errors in PR checks

Right now we can't really handle the error status because when we get an error back from a PR audit, we don't know what SHA to use for the commit status.

Recover from broken clones

If the process is killed in the middle of a clone, the repo will be left in a broken state. Since our repo detection is as simple as whether or not the directory exists, there's no recovery path from a broken repo. We can easily recover manually by just deleting the repo, but the script should be able to detect this and recover automatically.

Auto-recheck failed PRs

When we fetch updated CLA signatures, we can diff that against the previous list to find new signatures. If we then track which PRs failed because of which missing signatures, we can recheck PRs automatically.

Validate commit author

Checks might fail due to invalid email addresses or names on the commit itself. We currently don't validate that part, only the signautes themselves. So when the name on the commit is invalid, like this one, we don't provide any specific error messages hinting at the issue.

Failing to detect signed CLAs.

With the transition to the JS Foundation the jquerybot seems to be broken.
It no longer detects signed CLAs or at least isn't detected my signed CLA (I resigned after the transition and it shows up in the google spreadsheet).

I've disabled its web hook (temp disable) while the issues are being worked out.

More detail in GitHub API request errors

We had a PR this morning that didn't get any status set (not even pending). I was able to find this in syslog:

Mar  6 13:02:10 cla-01 jquery-license[12865]: Error auditing hook { repo: 'globalize',#012  pr: 410,#012  head: 'a24f26639a6e8f1e3b9685c480e7fa16a5692852',#012  error: 'Error: Not Found\n    at IncomingMessage.<anonymous> (/var/www/jquery-license/node_modules/github-request/lib/request.js:74:14)\n    at IncomingMessage.emit (events.js:117:20)\n    at _stream_readable.js:943:16\n    at process._tickCallback (node.js:419:13)' }

Without long stack traces, that doesn't provide much insight. All we can tell is that some request failed.

Set pending status immediately for PRs

When a webhook is received, we should immediately set the commit status to pending to indicate that the PR is not yet valid, even if there are no errors listed.

Email and Name checks should not be case sensitive

Email addresses still resolve no matter the case used and a person is still the same person if their name is in all caps or all lower case. This would also prevent false failures if someone accidentally capitalizes the first 2 letters in their name or something when typing.

Ability to recheck a PR

Once #1 is implemented, it should be possible to click a button to recheck the PR. This will be useful for the scenario where a contributor signs the CLA after the check has failed.

Set description on PR status

For Travis the status says "The Travis CI build passed" or "The Travis CI build failed". We should set that for the CLA as well.

Match all email addresses

We have multiple signatures with the same email address, but different names. Usually someone signs, gets told its wrong and signs again with the name matching the git author. The check then says its still wrong, since it only looks at the name of the first matching email address - we consider duplicate signatures as invalid:

// Check for duplicate signatures
if ( authorEmails.hasOwnProperty( email ) ) {
errors.push( email + " signed multiple times." );
return false;
} else {
authorEmails[ email ] = true;
}

I think the above is the cause of the CLA failure here: globalizejs/globalize#456

The simplest solution I can think of is to go through signatures in reverse order, that way newer signatures override older (invalid) signatures.

Provide more detail for noreply email addresses

We don't accept users.noreply.github.com email addresses because they're fairly meaningless (we can't use them to actually contact the author). Right now, the CLA status page just says "{email} is a private GitHub email address." We should provide more detail to explain why the email address isn't acceptable and what the author should do to fix the problem.

Move org authorization out of the config.json file

Right now the config.json file has to be edited in order to add each new organization. That's cumbersome. Let's either remove that completely or have a shared secret that can just be passed by the webhook.

Only validate email on pull requests

On ESLint, we are seeing a lot of issues with name mismatches and people having trouble configuring their git settings so it will work. At first glance, it seems like checking the author name is useless so long as the author email (a truly unique ID) is verified.

Can we remove the name check and only validate the email address?

Better error reporting for invalid signatures

Invalid CLA signatures are detected, but essentially thrown away. Now that audit results are shown, along with instructions on how to fix the problem, we should retain the information to provide better data on the details page.

jquery/jquery-ui#1451 shows an example of this. Though, interestingly, the CLA was first signed correctly, then signed incorrectly to match the git config. So in this case, the current instructions were correct.

bot doesn't work when added as a hook to an org

The bot worked for the lodash repo when the hook was on the lodash/lodash repo, but when I move the hook from the repo to the org it no longer works. The bot has access to the repo still (as an outside collaborator) and to all the repos in the org.

Notify users on CLA error

Right now we set the commit status and apply labels after auditing a PR, but neither of these will actually notify the contributor if there is a problem. We need to figure out how we want to notify contributors: via a PR comment or via an email.

Verify author name matches CLA signature

We validate the names on the CLA signatures and we verify that the author email matches a valid CLA signature email, but we don't verify that the author name matches the CLA signature. This results in incorrect author details being marked as valid, such as when users put their nickname in their git config.

CLA check missing.

With the transition to the JS Foundation site the CLA check functionality of the jquerybot is missing. It creates handy links like

If you've already signed our CLA, it's possible your git author information doesn't match your CLA signature (both your name and email have to match), for more information, check the status of your CLA check.

but instead of giving info on how to status of the check and how to resolve issues it just redirects to the CLA entry page.

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.