angularjs-blow_your_mind/beautify-html.js

738 lines
33 KiB
JavaScript
Raw Normal View History

2013-07-15 01:49:03 +00:00
/*jshint curly:true, eqeqeq:true, laxbreak:true, noempty:false */
/*
The MIT License (MIT)
Copyright (c) 2007-2013 Einar Lielmanis and contributors.
Permission is hereby granted, free of charge, to any person
obtaining a copy of this software and associated documentation files
(the "Software"), to deal in the Software without restriction,
including without limitation the rights to use, copy, modify, merge,
publish, distribute, sublicense, and/or sell copies of the Software,
and to permit persons to whom the Software is furnished to do so,
subject to the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Style HTML
---------------
Written by Nochum Sossonko, (nsossonko@hotmail.com)
Based on code initially developed by: Einar Lielmanis, <elfz@laacz.lv>
http://jsbeautifier.org/
Usage:
style_html(html_source);
style_html(html_source, options);
The options are:
indent_size (default 4) indentation size,
indent_char (default space) character to indent with,
wrap_line_length (default 250) - maximum amount of characters per line (0 = disable)
brace_style (default "collapse") - "collapse" | "expand" | "end-expand"
put braces on the same line as control statements (default), or put braces on own line (Allman / ANSI style), or just put end braces on own line.
unformatted (defaults to inline tags) - list of tags, that shouldn't be reformatted
indent_scripts (default normal) - "keep"|"separate"|"normal"
preserve_newlines (default true) - whether existing line breaks before elements should be preserved
Only works before elements, not inside tags or for text.
max_preserve_newlines (default unlimited) - maximum number of line breaks to be preserved in one chunk
e.g.
style_html(html_source, {
'indent_size': 2,
'indent_char': ' ',
'wrap_line_length': 78,
'brace_style': 'expand',
'unformatted': ['a', 'sub', 'sup', 'b', 'i', 'u'],
'preserve_newlines': true,
'max_preserve_newlines': 5
});
*/
(function() {
function trim(s) {
return s.replace(/^\s+|\s+$/g, '');
}
function ltrim(s) {
return s.replace(/^\s+/g, '');
}
function style_html(html_source, options, js_beautify, css_beautify) {
//Wrapper function to invoke all the necessary constructors and deal with the output.
var multi_parser,
indent_size,
indent_character,
wrap_line_length,
brace_style,
unformatted,
preserve_newlines,
max_preserve_newlines;
options = options || {};
// backwards compatibility to 1.3.4
if (options.wrap_line_length === undefined && options.max_char !== undefined) {
options.wrap_line_length = options.max_char;
}
indent_size = parseInt(options.indent_size || 4, 10);
indent_character = options.indent_char || ' ';
brace_style = options.brace_style || 'collapse';
wrap_line_length = options.wrap_line_length === 0 ? 32786 : parseInt(options.wrap_line_length || 250, 10);
unformatted = options.unformatted || ['a', 'span', 'bdo', 'em', 'strong', 'dfn', 'code', 'samp', 'kbd', 'var', 'cite', 'abbr', 'acronym', 'q', 'sub', 'sup', 'tt', 'i', 'b', 'big', 'small', 'u', 's', 'strike', 'font', 'ins', 'del', 'pre', 'address', 'dt', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6'];
preserve_newlines = options.preserve_newlines || true;
max_preserve_newlines = preserve_newlines ? parseInt(options.max_preserve_newlines || 32786, 10) : 0;
function Parser() {
this.pos = 0; //Parser position
this.token = '';
this.current_mode = 'CONTENT'; //reflects the current Parser mode: TAG/CONTENT
this.tags = { //An object to hold tags, their position, and their parent-tags, initiated with default values
parent: 'parent1',
parentcount: 1,
parent1: ''
};
this.tag_type = '';
this.token_text = this.last_token = this.last_text = this.token_type = '';
this.newlines = 0;
this.indent_content = false;
this.Utils = { //Uilities made available to the various functions
whitespace: "\n\r\t ".split(''),
single_token: 'br,input,link,meta,!doctype,basefont,base,area,hr,wbr,param,img,isindex,?xml,embed,?php,?,?='.split(','), //all the single tags for HTML
extra_liners: 'head,body,/html'.split(','), //for tags that need a line of whitespace before them
in_array: function(what, arr) {
for (var i = 0; i < arr.length; i++) {
if (what === arr[i]) {
return true;
}
}
return false;
}
};
this.traverse_whitespace = function() {
var input_char = '';
input_char = this.input.charAt(this.pos);
if (this.Utils.in_array(input_char, this.Utils.whitespace)) {
this.newlines = 0;
while (this.Utils.in_array(input_char, this.Utils.whitespace)) {
if (preserve_newlines && input_char === '\n' && this.newlines <= max_preserve_newlines) {
this.newlines += 1;
}
this.pos++;
input_char = this.input.charAt(this.pos);
}
return true;
}
return false;
};
this.get_content = function() { //function to capture regular content between tags
var input_char = '',
content = [],
space = false; //if a space is needed
while (this.input.charAt(this.pos) !== '<') {
if (this.pos >= this.input.length) {
return content.length ? content.join('') : ['', 'TK_EOF'];
}
if (this.traverse_whitespace()) {
if (content.length) {
space = true;
}
continue; //don't want to insert unnecessary space
}
input_char = this.input.charAt(this.pos);
this.pos++;
if (space) {
if (this.line_char_count >= this.wrap_line_length) { //insert a line when the wrap_line_length is reached
this.print_newline(false, content);
this.print_indentation(content);
} else {
this.line_char_count++;
content.push(' ');
}
space = false;
}
this.line_char_count++;
content.push(input_char); //letter at-a-time (or string) inserted to an array
}
return content.length ? content.join('') : '';
};
this.get_contents_to = function(name) { //get the full content of a script or style to pass to js_beautify
if (this.pos === this.input.length) {
return ['', 'TK_EOF'];
}
var input_char = '';
var content = '';
var reg_match = new RegExp('</' + name + '\\s*>', 'igm');
reg_match.lastIndex = this.pos;
var reg_array = reg_match.exec(this.input);
var end_script = reg_array ? reg_array.index : this.input.length; //absolute end of script
if (this.pos < end_script) { //get everything in between the script tags
content = this.input.substring(this.pos, end_script);
this.pos = end_script;
}
return content;
};
this.record_tag = function(tag) { //function to record a tag and its parent in this.tags Object
if (this.tags[tag + 'count']) { //check for the existence of this tag type
this.tags[tag + 'count']++;
this.tags[tag + this.tags[tag + 'count']] = this.indent_level; //and record the present indent level
} else { //otherwise initialize this tag type
this.tags[tag + 'count'] = 1;
this.tags[tag + this.tags[tag + 'count']] = this.indent_level; //and record the present indent level
}
this.tags[tag + this.tags[tag + 'count'] + 'parent'] = this.tags.parent; //set the parent (i.e. in the case of a div this.tags.div1parent)
this.tags.parent = tag + this.tags[tag + 'count']; //and make this the current parent (i.e. in the case of a div 'div1')
};
this.retrieve_tag = function(tag) { //function to retrieve the opening tag to the corresponding closer
if (this.tags[tag + 'count']) { //if the openener is not in the Object we ignore it
var temp_parent = this.tags.parent; //check to see if it's a closable tag.
while (temp_parent) { //till we reach '' (the initial value);
if (tag + this.tags[tag + 'count'] === temp_parent) { //if this is it use it
break;
}
temp_parent = this.tags[temp_parent + 'parent']; //otherwise keep on climbing up the DOM Tree
}
if (temp_parent) { //if we caught something
this.indent_level = this.tags[tag + this.tags[tag + 'count']]; //set the indent_level accordingly
this.tags.parent = this.tags[temp_parent + 'parent']; //and set the current parent
}
delete this.tags[tag + this.tags[tag + 'count'] + 'parent']; //delete the closed tags parent reference...
delete this.tags[tag + this.tags[tag + 'count']]; //...and the tag itself
if (this.tags[tag + 'count'] === 1) {
delete this.tags[tag + 'count'];
} else {
this.tags[tag + 'count']--;
}
}
};
this.get_tag = function(peek) { //function to get a full tag and parse its type
var input_char = '',
content = [],
comment = '',
space = false,
tag_start, tag_end,
orig_pos = this.pos,
orig_line_char_count = this.line_char_count;
peek = peek !== undefined ? peek : false;
do {
if (this.pos >= this.input.length) {
if (peek) {
this.pos = orig_pos;
this.line_char_count = orig_line_char_count;
}
return content.length ? content.join('') : ['', 'TK_EOF'];
}
input_char = this.input.charAt(this.pos);
this.pos++;
if (this.Utils.in_array(input_char, this.Utils.whitespace)) { //don't want to insert unnecessary space
space = true;
continue;
}
if (input_char === "'" || input_char === '"') {
input_char += this.get_unformatted(input_char);
space = true;
}
if (input_char === '=') { //no space before =
space = false;
}
if (content.length && content[content.length - 1] !== '=' && input_char !== '>' && space) {
//no space after = or before >
if (this.line_char_count >= this.wrap_line_length) {
this.print_newline(false, content);
this.print_indentation(content);
} else {
content.push(' ');
this.line_char_count++;
}
space = false;
}
if (input_char === '<' && !tag_start) {
tag_start = this.pos - 1;
}
this.line_char_count++;
content.push(input_char); //inserts character at-a-time (or string)
if (content[1] && content[1] === '!') { //if we're in a comment, do something special
// We treat all comments as literals, even more than preformatted tags
// we just look for the appropriate close tag
content = [this.get_comment(tag_start)];
break;
}
} while (input_char !== '>');
var tag_complete = content.join('');
var tag_index;
if (tag_complete.indexOf(' ') !== -1) { //if there's whitespace, thats where the tag name ends
tag_index = tag_complete.indexOf(' ');
} else { //otherwise go with the tag ending
tag_index = tag_complete.indexOf('>');
}
var tag_check = tag_complete.substring(1, tag_index).toLowerCase();
if (tag_complete.charAt(tag_complete.length - 2) === '/' ||
this.Utils.in_array(tag_check, this.Utils.single_token)) { //if this tag name is a single tag type (either in the list or has a closing /)
if (!peek) {
this.tag_type = 'SINGLE';
}
} else if (tag_check === 'script') { //for later script handling
if (!peek) {
this.record_tag(tag_check);
this.tag_type = 'SCRIPT';
}
} else if (tag_check === 'style') { //for future style handling (for now it justs uses get_content)
if (!peek) {
this.record_tag(tag_check);
this.tag_type = 'STYLE';
}
} else if (this.is_unformatted(tag_check, unformatted)) { // do not reformat the "unformatted" tags
comment = this.get_unformatted('</' + tag_check + '>', tag_complete); //...delegate to get_unformatted function
content.push(comment);
// Preserve collapsed whitespace either before or after this tag.
if (tag_start > 0 && this.Utils.in_array(this.input.charAt(tag_start - 1), this.Utils.whitespace)) {
content.splice(0, 0, this.input.charAt(tag_start - 1));
}
tag_end = this.pos - 1;
if (this.Utils.in_array(this.input.charAt(tag_end + 1), this.Utils.whitespace)) {
content.push(this.input.charAt(tag_end + 1));
}
this.tag_type = 'SINGLE';
} else if (tag_check.charAt(0) === '!') { //peek for <! comment
// for comments content is already correct.
if (!peek) {
this.tag_type = 'SINGLE';
this.traverse_whitespace();
}
} else if (!peek) {
if (tag_check.charAt(0) === '/') { //this tag is a double tag so check for tag-ending
this.retrieve_tag(tag_check.substring(1)); //remove it and all ancestors
this.tag_type = 'END';
this.traverse_whitespace();
} else { //otherwise it's a start-tag
this.record_tag(tag_check); //push it on the tag stack
if (tag_check.toLowerCase() !== 'html') {
this.indent_content = true;
}
this.tag_type = 'START';
// Allow preserving of newlines after a start tag
this.traverse_whitespace();
}
if (this.Utils.in_array(tag_check, this.Utils.extra_liners)) { //check if this double needs an extra line
this.print_newline(false, this.output);
if (this.output.length && this.output[this.output.length - 2] !== '\n') {
this.print_newline(true, this.output);
}
}
}
if (peek) {
this.pos = orig_pos;
this.line_char_count = orig_line_char_count;
}
return content.join(''); //returns fully formatted tag
};
this.get_comment = function(start_pos) { //function to return comment content in its entirety
// this is will have very poor perf, but will work for now.
var comment = '',
delimiter = '>',
matched = false;
this.pos = start_pos;
input_char = this.input.charAt(this.pos);
this.pos++;
while (this.pos <= this.input.length) {
comment += input_char;
// only need to check for the delimiter if the last chars match
if (comment[comment.length - 1] === delimiter[delimiter.length - 1] &&
comment.indexOf(delimiter) !== -1) {
break;
}
// only need to search for custom delimiter for the first few characters
if (!matched && comment.length < 10) {
if (comment.indexOf('<![if') === 0) { //peek for <![if conditional comment
delimiter = '<![endif]>';
matched = true;
} else if (comment.indexOf('<![cdata[') === 0) { //if it's a <[cdata[ comment...
delimiter = ']]>';
matched = true;
} else if (comment.indexOf('<![') === 0) { // some other ![ comment? ...
delimiter = ']>';
matched = true;
} else if (comment.indexOf('<!--') === 0) { // <!-- comment ...
delimiter = '-->';
matched = true;
}
}
input_char = this.input.charAt(this.pos);
this.pos++;
}
return comment;
};
this.get_unformatted = function(delimiter, orig_tag) { //function to return unformatted content in its entirety
if (orig_tag && orig_tag.toLowerCase().indexOf(delimiter) !== -1) {
return '';
}
var input_char = '';
var content = '';
var space = true;
do {
if (this.pos >= this.input.length) {
return content;
}
input_char = this.input.charAt(this.pos);
this.pos++;
if (this.Utils.in_array(input_char, this.Utils.whitespace)) {
if (!space) {
this.line_char_count--;
continue;
}
if (input_char === '\n' || input_char === '\r') {
content += '\n';
/* Don't change tab indention for unformatted blocks. If using code for html editing, this will greatly affect <pre> tags if they are specified in the 'unformatted array'
for (var i=0; i<this.indent_level; i++) {
content += this.indent_string;
}
space = false; //...and make sure other indentation is erased
*/
this.line_char_count = 0;
continue;
}
}
content += input_char;
this.line_char_count++;
space = true;
} while (content.toLowerCase().indexOf(delimiter) === -1);
return content;
};
this.get_token = function() { //initial handler for token-retrieval
var token;
if (this.last_token === 'TK_TAG_SCRIPT' || this.last_token === 'TK_TAG_STYLE') { //check if we need to format javascript
var type = this.last_token.substr(7);
token = this.get_contents_to(type);
if (typeof token !== 'string') {
return token;
}
return [token, 'TK_' + type];
}
if (this.current_mode === 'CONTENT') {
token = this.get_content();
if (typeof token !== 'string') {
return token;
} else {
return [token, 'TK_CONTENT'];
}
}
if (this.current_mode === 'TAG') {
token = this.get_tag();
if (typeof token !== 'string') {
return token;
} else {
var tag_name_type = 'TK_TAG_' + this.tag_type;
return [token, tag_name_type];
}
}
};
this.get_full_indent = function(level) {
level = this.indent_level + level || 0;
if (level < 1) {
return '';
}
return Array(level + 1).join(this.indent_string);
};
this.is_unformatted = function(tag_check, unformatted) {
//is this an HTML5 block-level link?
if (!this.Utils.in_array(tag_check, unformatted)) {
return false;
}
if (tag_check.toLowerCase() !== 'a' || !this.Utils.in_array('a', unformatted)) {
return true;
}
//at this point we have an tag; is its first child something we want to remain
//unformatted?
var next_tag = this.get_tag(true /* peek. */ );
// tets next_tag to see if it is just html tag (no external content)
var tag = (next_tag || "").match(/^\s*<\s*\/?([a-z]*)\s*[^>]*>\s*$/);
// if next_tag comes back but is not an isolated tag, then
// let's treat the 'a' tag as having content
// and respect the unformatted option
if (!tag || this.Utils.in_array(tag, unformatted)) {
return true;
} else {
return false;
}
};
this.printer = function(js_source, indent_character, indent_size, wrap_line_length, brace_style) { //handles input/output and some other printing functions
this.input = js_source || ''; //gets the input for the Parser
this.output = [];
this.indent_character = indent_character;
this.indent_string = '';
this.indent_size = indent_size;
this.brace_style = brace_style;
this.indent_level = 0;
this.wrap_line_length = wrap_line_length;
this.line_char_count = 0; //count to see if wrap_line_length was exceeded
for (var i = 0; i < this.indent_size; i++) {
this.indent_string += this.indent_character;
}
this.print_newline = function(force, arr) {
this.line_char_count = 0;
if (!arr || !arr.length) {
return;
}
if (force || (arr[arr.length - 1] !== '\n')) { //we might want the extra line
arr.push('\n');
}
};
this.print_indentation = function(arr) {
for (var i = 0; i < this.indent_level; i++) {
arr.push(this.indent_string);
this.line_char_count += this.indent_string.length;
}
};
this.print_token = function(text) {
if (text || text !== '') {
if (this.output.length && this.output[this.output.length - 1] === '\n') {
this.print_indentation(this.output);
text = ltrim(text);
}
}
this.print_token_raw(text);
};
this.print_token_raw = function(text) {
if (text && text !== '') {
if (text.length > 1 && text[text.length - 1] === '\n') {
// unformatted tags can grab newlines as their last character
this.output.push(text.slice(0, -1));
this.print_newline(false, this.output);
} else {
this.output.push(text);
}
}
for (var n = 0; n < this.newlines; n++) {
this.print_newline(n > 0, this.output);
}
this.newlines = 0;
};
this.indent = function() {
this.indent_level++;
};
this.unindent = function() {
if (this.indent_level > 0) {
this.indent_level--;
}
};
};
return this;
}
/*_____________________--------------------_____________________*/
multi_parser = new Parser(); //wrapping functions Parser
multi_parser.printer(html_source, indent_character, indent_size, wrap_line_length, brace_style); //initialize starting values
while (true) {
var t = multi_parser.get_token();
multi_parser.token_text = t[0];
multi_parser.token_type = t[1];
if (multi_parser.token_type === 'TK_EOF') {
break;
}
switch (multi_parser.token_type) {
case 'TK_TAG_START':
multi_parser.print_newline(false, multi_parser.output);
multi_parser.print_token(multi_parser.token_text);
if (multi_parser.indent_content) {
multi_parser.indent();
multi_parser.indent_content = false;
}
multi_parser.current_mode = 'CONTENT';
break;
case 'TK_TAG_STYLE':
case 'TK_TAG_SCRIPT':
multi_parser.print_newline(false, multi_parser.output);
multi_parser.print_token(multi_parser.token_text);
multi_parser.current_mode = 'CONTENT';
break;
case 'TK_TAG_END':
//Print new line only if the tag has no content and has child
if (multi_parser.last_token === 'TK_CONTENT' && multi_parser.last_text === '') {
var tag_name = multi_parser.token_text.match(/\w+/)[0];
var tag_extracted_from_last_output = multi_parser.output[multi_parser.output.length - 1].match(/<\s*(\w+)/);
if (tag_extracted_from_last_output === null ||
tag_extracted_from_last_output[1] !== tag_name) {
multi_parser.print_newline(false, multi_parser.output);
}
}
multi_parser.print_token(multi_parser.token_text);
multi_parser.current_mode = 'CONTENT';
break;
case 'TK_TAG_SINGLE':
// Don't add a newline before elements that should remain unformatted.
var tag_check = multi_parser.token_text.match(/^\s*<([a-z]+)/i);
if (!tag_check || !multi_parser.Utils.in_array(tag_check[1], unformatted)) {
multi_parser.print_newline(false, multi_parser.output);
}
multi_parser.print_token(multi_parser.token_text);
multi_parser.current_mode = 'CONTENT';
break;
case 'TK_CONTENT':
multi_parser.print_token(multi_parser.token_text);
multi_parser.current_mode = 'TAG';
break;
case 'TK_STYLE':
case 'TK_SCRIPT':
if (multi_parser.token_text !== '') {
multi_parser.print_newline(false, multi_parser.output);
var text = multi_parser.token_text,
_beautifier,
script_indent_level = 1;
if (multi_parser.token_type === 'TK_SCRIPT') {
_beautifier = typeof js_beautify === 'function' && js_beautify;
} else if (multi_parser.token_type === 'TK_STYLE') {
_beautifier = typeof css_beautify === 'function' && css_beautify;
}
if (options.indent_scripts === "keep") {
script_indent_level = 0;
} else if (options.indent_scripts === "separate") {
script_indent_level = -multi_parser.indent_level;
}
var indentation = multi_parser.get_full_indent(script_indent_level);
if (_beautifier) {
// call the Beautifier if avaliable
text = _beautifier(text.replace(/^\s*/, indentation), options);
} else {
// simply indent the string otherwise
var white = text.match(/^\s*/)[0];
var _level = white.match(/[^\n\r]*$/)[0].split(multi_parser.indent_string).length - 1;
var reindent = multi_parser.get_full_indent(script_indent_level - _level);
text = text.replace(/^\s*/, indentation)
.replace(/\r\n|\r|\n/g, '\n' + reindent)
.replace(/\s+$/, '');
}
if (text) {
multi_parser.print_token_raw(indentation + trim(text));
multi_parser.print_newline(false, multi_parser.output);
}
}
multi_parser.current_mode = 'TAG';
break;
}
multi_parser.last_token = multi_parser.token_type;
multi_parser.last_text = multi_parser.token_text;
}
return multi_parser.output.join('');
}
if (typeof define === "function") {
// Add support for require.js
define(["./beautify.js", "./beautify-css.js"], function(js_beautify, css_beautify) {
return {
html_beautify: function(html_source, options) {
return style_html(html_source, options, js_beautify, css_beautify);
}
};
});
} else if (typeof exports !== "undefined") {
// Add support for CommonJS. Just put this file somewhere on your require.paths
// and you will be able to `var html_beautify = require("beautify").html_beautify`.
var js_beautify = require('./beautify.js').js_beautify;
var css_beautify = require('./beautify-css.js').css_beautify;
exports.html_beautify = function(html_source, options) {
return style_html(html_source, options, js_beautify, css_beautify);
};
} else if (typeof window !== "undefined") {
// If we're running a web page and don't have either of the above, add our one global
window.html_beautify = function(html_source, options) {
return style_html(html_source, options, window.js_beautify, window.css_beautify);
};
} else if (typeof global !== "undefined") {
// If we don't even have window, try global.
global.html_beautify = function(html_source, options) {
return style_html(html_source, options, global.js_beautify, global.css_beautify);
};
}
}());