1
0
mirror of https://github.com/twitter/twemoji.git synced 2024-12-27 03:16:23 +00:00
twemoji/1/twemoji.amd.js
2016-03-02 19:43:00 +00:00

569 lines
21 KiB
JavaScript

define(function () {
/*jslint indent: 2, browser: true, bitwise: true, plusplus: true */
var twemoji = (function (
/*! Copyright Twitter Inc. and other contributors. Licensed under MIT *//*
https://github.com/twitter/twemoji/blob/gh-pages/LICENSE
*/
// WARNING: this file is generated automatically via
// `node twemoji-generator.js`
// please update its `createTwemoji` function
// at the bottom of the same file instead.
) {
'use strict';
/*jshint maxparams:4 */
var
// the exported module object
twemoji = {
/////////////////////////
// properties //
/////////////////////////
// default assets url, by default will be Twitter Inc. CDN
base: 'https://twemoji.maxcdn.com/',
// default assets file extensions, by default '.png'
ext: '.png',
// default assets/folder size, by default "72x72"
// available via Twitter CDN: 72
size: '72x72',
// default class name, by default 'emoji'
className: 'emoji',
// basic utilities / helpers to convert code points
// to JavaScript surrogates and vice versa
convert: {
/**
* Given an HEX codepoint, returns UTF16 surrogate pairs.
*
* @param string generic codepoint, i.e. '1F4A9'
* @return string codepoint transformed into utf16 surrogates pair,
* i.e. \uD83D\uDCA9
*
* @example
* twemoji.convert.fromCodePoint('1f1e8');
* // "\ud83c\udde8"
*
* '1f1e8-1f1f3'.split('-').map(twemoji.convert.fromCodePoint).join('')
* // "\ud83c\udde8\ud83c\uddf3"
*/
fromCodePoint: fromCodePoint,
/**
* Given UTF16 surrogate pairs, returns the equivalent HEX codepoint.
*
* @param string generic utf16 surrogates pair, i.e. \uD83D\uDCA9
* @param string optional separator for double code points, default='-'
* @return string utf16 transformed into codepoint, i.e. '1F4A9'
*
* @example
* twemoji.convert.toCodePoint('\ud83c\udde8\ud83c\uddf3');
* // "1f1e8-1f1f3"
*
* twemoji.convert.toCodePoint('\ud83c\udde8\ud83c\uddf3', '~');
* // "1f1e8~1f1f3"
*/
toCodePoint: toCodePoint
},
/////////////////////////
// methods //
/////////////////////////
/**
* User first: used to remove missing images
* preserving the original text intent when
* a fallback for network problems is desired.
* Automatically added to Image nodes via DOM
* It could be recycled for string operations via:
* $('img.emoji').on('error', twemoji.onerror)
*/
onerror: function onerror() {
if (this.parentNode) {
this.parentNode.replaceChild(createText(this.alt), this);
}
},
/**
* Main method/logic to generate either <img> tags or HTMLImage nodes.
* "emojify" a generic text or DOM Element.
*
* @overloads
*
* String replacement for `innerHTML` or server side operations
* twemoji.parse(string);
* twemoji.parse(string, Function);
* twemoji.parse(string, Object);
*
* HTMLElement tree parsing for safer operations over existing DOM
* twemoji.parse(HTMLElement);
* twemoji.parse(HTMLElement, Function);
* twemoji.parse(HTMLElement, Object);
*
* @param string|HTMLElement the source to parse and enrich with emoji.
*
* string replace emoji matches with <img> tags.
* Mainly used to inject emoji via `innerHTML`
* It does **not** parse the string or validate it,
* it simply replaces found emoji with a tag.
* NOTE: be sure this won't affect security.
*
* HTMLElement walk through the DOM tree and find emoji
* that are inside **text node only** (nodeType === 3)
* Mainly used to put emoji in already generated DOM
* without compromising surrounding nodes and
* **avoiding** the usage of `innerHTML`.
* NOTE: Using DOM elements instead of strings should
* improve security without compromising too much
* performance compared with a less safe `innerHTML`.
*
* @param Function|Object [optional]
* either the callback that will be invoked or an object
* with all properties to use per each found emoji.
*
* Function if specified, this will be invoked per each emoji
* that has been found through the RegExp except
* those follwed by the invariant \uFE0E ("as text").
* Once invoked, parameters will be:
*
* iconId:string the lower case HEX code point
* i.e. "1f4a9"
*
* options:Object all info for this parsing operation
*
* variant:char the optional \uFE0F ("as image")
* variant, in case this info
* is anyhow meaningful.
* By default this is ignored.
*
* If such callback will return a falsy value instead
* of a valid `src` to use for the image, nothing will
* actually change for that specific emoji.
*
*
* Object if specified, an object containing the following properties
*
* callback Function the callback to invoke per each found emoji.
* base string the base url, by default twemoji.base
* ext string the image extension, by default twemoji.ext
* size string the assets size, by default twemoji.size
*
* @example
*
* twemoji.parse("I \u2764\uFE0F emoji!");
* // I <img class="emoji" draggable="false" alt="❤️" src="/assets/2764.gif"> emoji!
*
*
* twemoji.parse("I \u2764\uFE0F emoji!", function(iconId, options) {
* return '/assets/' + iconId + '.gif';
* });
* // I <img class="emoji" draggable="false" alt="❤️" src="/assets/2764.gif"> emoji!
*
*
* twemoji.parse("I \u2764\uFE0F emoji!", {
* size: 72,
* callback: function(iconId, options) {
* return '/assets/' + options.size + '/' + iconId + options.ext;
* }
* });
* // I <img class="emoji" draggable="false" alt="❤️" src="/assets/72x72/2764.png"> emoji!
*
*/
parse: parse,
/**
* Given a string, invokes the callback argument
* per each emoji found in such string.
* This is the most raw version used by
* the .parse(string) method itself.
*
* @param string generic string to parse
* @param Function a generic callback that will be
* invoked to replace the content.
* This calback wil receive standard
* String.prototype.replace(str, callback)
* arguments such:
* callback(
* rawText, // the emoji match
* );
*
* and others commonly received via replace.
*/
replace: replace,
/**
* Simplify string tests against emoji.
*
* @param string some text that might contain emoji
* @return boolean true if any emoji was found, false otherwise.
*
* @example
*
* if (twemoji.test(someContent)) {
* console.log("emoji All The Things!");
* }
*/
test: test
},
// used to escape HTML special chars in attributes
escaper = {
'&': '&amp;',
'<': '&lt;',
'>': '&gt;',
"'": '&#39;',
'"': '&quot;'
},
// RegExp based on emoji's official Unicode standards
// http://www.unicode.org/Public/UNIDATA/EmojiSources.txt
re = /(?:[\u0039\u0038\u0037\u0036\u0035\u0034\u0033\u0032\u0031\u0030\u0023])\ufe0f?\u20e3|\ud83c\udde8\ud83c\uddf3|\ud83c\udde9\ud83c\uddea|\ud83c\uddea\ud83c\uddf8|\ud83c\uddeb\ud83c\uddf7|\ud83c\uddec\ud83c\udde7|\ud83c\uddee\ud83c\uddf9|\ud83c\uddef\ud83c\uddf5|\ud83c\uddf0\ud83c\uddf7|\ud83c\uddf7\ud83c\uddfa|\ud83c\uddfa\ud83c\uddf8|\ud83d[\udc00-\udc3e\udc40\udc42-\udcf7\udcf9-\udcfc\udd00-\udd3d\udd50-\udd67\uddfb-\ude42\ude45-\ude4f\ude80-\udec5]|\ud83c[\udccf\udd8e\udd91-\udd9a\udde6-\uddff\ude01\ude32-\ude36\ude38-\ude3a\ude50\ude51\udf00-\udf20\udf30-\udf35\udf37-\udf7c\udf80-\udf93\udfa0-\udfc4\udfc6-\udfca\udfe0-\udff0]|[\ue50a\u27bf\u27b0\u2797\u2796\u2795\u2755\u2754\u2753\u274e\u274c\u2728\u270b\u270a\u2705\u26ce\u23f3\u23f0\u23ec\u23eb\u23ea\u23e9]|(?:\ud83c[\udc04\udd70\udd71\udd7e\udd7f\ude02\ude1a\ude2f\ude37]|[\u3299\u3297\u303d\u3030\u2b55\u2b50\u2b1c\u2b1b\u2b07\u2b06\u2b05\u2935\u2934\u27a1\u2764\u2757\u2747\u2744\u2734\u2733\u2716\u2714\u2712\u270f\u270c\u2709\u2708\u2702\u26fd\u26fa\u26f5\u26f3\u26f2\u26ea\u26d4\u26c5\u26c4\u26be\u26bd\u26ab\u26aa\u26a1\u26a0\u2693\u267f\u267b\u2668\u2666\u2665\u2663\u2660\u2653\u2652\u2651\u2650\u264f\u264e\u264d\u264c\u264b\u264a\u2649\u2648\u263a\u261d\u2615\u2614\u2611\u260e\u2601\u2600\u25fe\u25fd\u25fc\u25fb\u25c0\u25b6\u25ab\u25aa\u24c2\u231b\u231a\u21aa\u21a9\u2199\u2198\u2197\u2196\u2195\u2194\u2139\u2122\u2049\u203c\u00ae\u00a9])(?:\ufe0f|(?!\ufe0e))/g,
// avoid runtime RegExp creation for not so smart,
// not JIT based, and old browsers / engines
UFE0Fg = /\uFE0F/g,
// avoid using a string literal like '\u200D' here because minifiers expand it inline
U200D = String.fromCharCode(0x200D),
// used to find HTML special chars in attributes
rescaper = /[&<>'"]/g,
// nodes with type 1 which should **not** be parsed (including lower case svg)
shouldntBeParsed = /IFRAME|NOFRAMES|NOSCRIPT|SCRIPT|SELECT|STYLE|TEXTAREA|[a-z]/,
// just a private shortcut
fromCharCode = String.fromCharCode;
return twemoji;
/////////////////////////
// private functions //
// declaration //
/////////////////////////
/**
* Shortcut to create text nodes
* @param string text used to create DOM text node
* @return Node a DOM node with that text
*/
function createText(text) {
return document.createTextNode(text);
}
/**
* Utility function to escape html attribute text
* @param string text use in HTML attribute
* @return string text encoded to use in HTML attribute
*/
function escapeHTML(s) {
return s.replace(rescaper, replacer);
}
/**
* Default callback used to generate emoji src
* based on Twitter CDN
* @param string the emoji codepoint string
* @param string the default size to use, i.e. "36x36"
* @return string the image source to use
*/
function defaultImageSrcGenerator(icon, options) {
return ''.concat(options.base, options.size, '/', icon, options.ext);
}
/**
* Given a generic DOM nodeType 1, walk through all children
* and store every nodeType 3 (#text) found in the tree.
* @param Element a DOM Element with probably some text in it
* @param Array the list of previously discovered text nodes
* @return Array same list with new discovered nodes, if any
*/
function grabAllTextNodes(node, allText) {
var
childNodes = node.childNodes,
length = childNodes.length,
subnode,
nodeType;
while (length--) {
subnode = childNodes[length];
nodeType = subnode.nodeType;
// parse emoji only in text nodes
if (nodeType === 3) {
// collect them to process emoji later
allText.push(subnode);
}
// ignore all nodes that are not type 1 or that
// should not be parsed as script, style, and others
else if (nodeType === 1 && !shouldntBeParsed.test(subnode.nodeName)) {
grabAllTextNodes(subnode, allText);
}
}
return allText;
}
/**
* Used to both remove the possible variant
* and to convert utf16 into code points.
* If there is a zero-width-joiner (U+200D), leave the variants in.
* @param string the raw text of the emoji match
*/
function grabTheRightIcon(rawText) {
// if variant is present as \uFE0F
return toCodePoint(rawText.indexOf(U200D) < 0 ?
rawText.replace(UFE0Fg, '') :
rawText
);
}
/**
* DOM version of the same logic / parser:
* emojify all found sub-text nodes placing images node instead.
* @param Element generic DOM node with some text in some child node
* @param Object options containing info about how to parse
*
* .callback Function the callback to invoke per each found emoji.
* .base string the base url, by default twemoji.base
* .ext string the image extension, by default twemoji.ext
* .size string the assets size, by default twemoji.size
*
* @return Element same generic node with emoji in place, if any.
*/
function parseNode(node, options) {
var
allText = grabAllTextNodes(node, []),
length = allText.length,
attrib,
attrname,
modified,
fragment,
subnode,
text,
match,
i,
index,
img,
rawText,
iconId,
src;
while (length--) {
modified = false;
fragment = document.createDocumentFragment();
subnode = allText[length];
text = subnode.nodeValue;
i = 0;
while ((match = re.exec(text))) {
index = match.index;
if (index !== i) {
fragment.appendChild(
createText(text.slice(i, index))
);
}
rawText = match[0];
iconId = grabTheRightIcon(rawText);
i = index + rawText.length;
src = options.callback(iconId, options);
if (src) {
img = new Image();
img.onerror = options.onerror;
img.setAttribute('draggable', 'false');
attrib = options.attributes(rawText, iconId);
for (attrname in attrib) {
if (
attrib.hasOwnProperty(attrname) &&
// don't allow any handlers to be set + don't allow overrides
attrname.indexOf('on') !== 0 &&
!img.hasAttribute(attrname)
) {
img.setAttribute(attrname, attrib[attrname]);
}
}
img.className = options.className;
img.alt = rawText;
img.src = src;
modified = true;
fragment.appendChild(img);
}
if (!img) fragment.appendChild(createText(rawText));
img = null;
}
// is there actually anything to replace in here ?
if (modified) {
// any text left to be added ?
if (i < text.length) {
fragment.appendChild(
createText(text.slice(i))
);
}
// replace the text node only, leave intact
// anything else surrounding such text
subnode.parentNode.replaceChild(fragment, subnode);
}
}
return node;
}
/**
* String/HTML version of the same logic / parser:
* emojify a generic text placing images tags instead of surrogates pair.
* @param string generic string with possibly some emoji in it
* @param Object options containing info about how to parse
*
* .callback Function the callback to invoke per each found emoji.
* .base string the base url, by default twemoji.base
* .ext string the image extension, by default twemoji.ext
* .size string the assets size, by default twemoji.size
*
* @return the string with <img tags> replacing all found and parsed emoji
*/
function parseString(str, options) {
return replace(str, function (rawText) {
var
ret = rawText,
iconId = grabTheRightIcon(rawText),
src = options.callback(iconId, options),
attrib,
attrname;
if (src) {
// recycle the match string replacing the emoji
// with its image counter part
ret = '<img '.concat(
'class="', options.className, '" ',
'draggable="false" ',
// needs to preserve user original intent
// when variants should be copied and pasted too
'alt="',
rawText,
'"',
' src="',
src,
'"'
);
attrib = options.attributes(rawText, iconId);
for (attrname in attrib) {
if (
attrib.hasOwnProperty(attrname) &&
// don't allow any handlers to be set + don't allow overrides
attrname.indexOf('on') !== 0 &&
ret.indexOf(' ' + attrname + '=') === -1
) {
ret = ret.concat(' ', attrname, '="', escapeHTML(attrib[attrname]), '"');
}
}
ret = ret.concat('>');
}
return ret;
});
}
/**
* Function used to actually replace HTML special chars
* @param string HTML special char
* @return string encoded HTML special char
*/
function replacer(m) {
return escaper[m];
}
/**
* Default options.attribute callback
* @return null
*/
function returnNull() {
return null;
}
/**
* Given a generic value, creates its squared counterpart if it's a number.
* As example, number 36 will return '36x36'.
* @param any a generic value.
* @return any a string representing asset size, i.e. "36x36"
* only in case the value was a number.
* Returns initial value otherwise.
*/
function toSizeSquaredAsset(value) {
return typeof value === 'number' ?
value + 'x' + value :
value;
}
/////////////////////////
// exported functions //
// declaration //
/////////////////////////
function fromCodePoint(codepoint) {
var code = typeof codepoint === 'string' ?
parseInt(codepoint, 16) : codepoint;
if (code < 0x10000) {
return fromCharCode(code);
}
code -= 0x10000;
return fromCharCode(
0xD800 + (code >> 10),
0xDC00 + (code & 0x3FF)
);
}
function parse(what, how) {
if (!how || typeof how === 'function') {
how = {callback: how};
}
// if first argument is string, inject html <img> tags
// otherwise use the DOM tree and parse text nodes only
return (typeof what === 'string' ? parseString : parseNode)(what, {
callback: how.callback || defaultImageSrcGenerator,
attributes: typeof how.attributes === 'function' ? how.attributes : returnNull,
base: typeof how.base === 'string' ? how.base : twemoji.base,
ext: how.ext || twemoji.ext,
size: how.folder || toSizeSquaredAsset(how.size || twemoji.size),
className: how.className || twemoji.className,
onerror: how.onerror || twemoji.onerror
});
}
function replace(text, callback) {
return String(text).replace(re, callback);
}
function test(text) {
// IE6 needs a reset before too
re.lastIndex = 0;
var result = re.test(text);
re.lastIndex = 0;
return result;
}
function toCodePoint(unicodeSurrogates, sep) {
var
r = [],
c = 0,
p = 0,
i = 0;
while (i < unicodeSurrogates.length) {
c = unicodeSurrogates.charCodeAt(i++);
if (p) {
r.push((0x10000 + ((p - 0xD800) << 10) + (c - 0xDC00)).toString(16));
p = 0;
} else if (0xD800 <= c && c <= 0xDBFF) {
p = c;
} else {
r.push(c.toString(16));
}
}
return r.join(sep || '-');
}
}());
return twemoji;
});