oxjs/source/Ox/js/String.js

573 lines
17 KiB
JavaScript

'use strict';
Ox.basename = function(str) {
/*
fixme: deprecate
>>> Ox.basename("foo/bar/foo.bar")
"foo.bar"
>>> Ox.basename("foo.bar")
"foo.bar"
*/
return str.replace(/^.*[\/\\]/g, '');
};
/*@
Ox.char <f> Alias for String.fromCharCode
@*/
// fixme: add some mapping? like Ox.char(9, 13) or Ox.char([9, 13])?
Ox.char = String.fromCharCode;
/*@
Ox.clean <f> Remove leading, trailing and double whitespace from a string
> Ox.clean("foo bar")
"foo bar"
> Ox.clean(" foo bar ")
"foo bar"
> Ox.clean(" foo \n bar ")
"foo\nbar"
> Ox.clean(" \nfoo\n\nbar\n ")
"foo\nbar"
> Ox.clean(" foo\tbar ")
"foo bar"
@*/
Ox.clean = function(str) {
return Ox.map(str.split('\n'), function(str) {
return Ox.trim(str.replace(/\s+/g, ' ')) || null;
}).join('\n');
};
/*@
Ox.endsWith <f> Checks if a string ends with a given substring
If the substring is a string literal (and not a variable),
<code>/sub$/.test(str)</code> or <code>!!/sub$/.exec(str)</code>
is shorter than <code>Ox.ends(str, sub)</code>.
> Ox.endsWith('foobar', 'bar')
true
@*/
Ox.ends = Ox.endsWith = function(str, sub) {
// fixme: rename to ends
return str.substr(str.length - sub.length) == sub;
};
/*@
Ox.highlight <f> Highlight matches in a string
> Ox.highlight('foobar', 'foo', 'match')
'<span class="match">foo</span>bar'
@*/
// fixme: with regexp, special chars have to be escaped
Ox.highlight = function(txt, str, classname) {
return str && str.length ? txt.replace(
new RegExp('(' + str + ')', 'ig'),
'<span class="' + classname + '">$1</span>'
) : txt;
};
/*@
Ox.highlightHTML <f> Highlight matches in an HTML string
> Ox.highlightHTML('<b>foo</b>bar', 'foobar', 'h')
'<b><span class="match">foo</span></b><span class="h">bar</span>'
> Ox.highlightHTML('<a href="/foo">foo</a>bar', 'foobar', 'h')
'<a href="/foo"><span class="h">foo</span></a><span class="h">bar</span>'
> Ox.highlightHTML('foo<br>bar', 'foobar', 'h')
'foo<br>bar'
> Ox.highlightHTML('AT&amp;T', 'AT&T', 'h')
'<span class="h">AT&amp;T</span>'
> Ox.highlightHTML('AT&amp;T', 'amp', 'h')
'AT&amp;T'
> Ox.highlightHTML('a &lt;b&gt; c', '<b>', 'h')
'a <span class="h">&lt;b&gt;</span> c'
> Ox.highlightHTML('a <br> c', 'b', 'h')
'a <br> c'
@*/
Ox.highlightHTML = function(html, str, classname, tags) {
var count = 0,
isEntity = false,
isTag = false,
position,
positions = [];
//fixme: default tags should be same as in parseHTML
tags = Ox.merge(tags || [], [
// inline formatting
'b', 'code', 'i', 's', 'sub', 'sup', 'u',
// block formatting
'blockquote', 'h1', 'h2', 'h3', 'p', 'pre',
// lists
'li', 'ol', 'ul',
// tables
'table', 'tbody', 'td', 'tfoot', 'th', 'thead', 'tr',
// other
'a', 'br', 'img',
]);
str = Ox.encodeHTML(str).toLowerCase();
Ox.forEach(html.toLowerCase(), function(chr, i) {
// check for entity or tag start
if (!isEntity && chr == '&') {
isEntity = true;
} else if (!isTag && chr == '<') {
Ox.forEach(tags, function(tag) {
if (html.substr(i + 1).match(new RegExp('^/?' + tag + '\\W'))) {
isTag = true;
return false;
}
});
}
// if outside entity or tag
if (!isEntity && !isTag) {
// if character matches
if (chr == str[count]) {
if (count == 0) {
position = i;
}
count++;
if (count == str.length) {
// make sure matches are last to first
positions.unshift([position, i + 1]);
}
} else {
count = 0;
}
}
// check for entity or tag end
if (isEntity && chr == ';') {
isEntity = false;
} else if (isTag && chr == '>') {
isTag = false;
}
});
positions.forEach(function(position) {
var match = '<span class="' + classname + '">'
+ html.substr(position[0], position[1] - position[0])
.replace(/(<.*?>)/g, '</span>$1<span class="' + classname + '">')
+ '</span>';
html = html.substr(0, position[0]) + match + html.substr(position[1]);
});
return html;
}
/*@
Ox.isValidEmail <f> Tests if a string is a valid e-mail address
(str) -> <b> True if the string is a valid e-mail address
str <s> Any string
> Ox.isValidEmail("foo@bar.com")
true
> Ox.isValidEmail("foo.bar@foobar.co.uk")
true
> Ox.isValidEmail("foo@bar")
false
> Ox.isValidEmail("foo@bar..com")
false
@*/
Ox.isValidEmail = function(str) {
return !!/^[0-9A-Z\.\+\-_]+@(?:[0-9A-Z\-]+\.)+[A-Z]{2,6}$/i.test(str);
};
/*@
Ox.pad <f> Pad a string to a given length
> Ox.pad(1, 2)
"01"
> Ox.pad("abc", -6, ".")
"abc..."
> Ox.pad("foobar", -3, ".")
"foo"
> Ox.pad("abc", -6, "123456")
"abc123"
> Ox.pad("abc", 6, "123456")
"456abc"
@*/
Ox.pad = function(str, len, pad) {
// fixme: slighly obscure signature
// fixme: weird for negative numbers
var pos = len / (len = Math.abs(len));
str = str.toString().substr(0, len);
pad = Ox.repeat(pad || '0', len - str.length);
str = pos == 1 ? pad + str : str + pad;
str = pos == 1 ?
str.substr(str.length - len, str.length) :
str.substr(0, len);
return str;
};
/*@
Ox.parsePath <f> Returns the components of a path
(str) -> <o> Path
extension <s> File extension
filename <s> Filename
pathname <s> Pathname
> Ox.parsePath('/foo/bar/foo.bar')
{extension: 'bar', filename: 'foo.bar', pathname: '/foo/bar/'}
> Ox.parsePath('foo/')
{extension: '', filename: '', pathname: 'foo/'}
> Ox.parsePath('foo')
{extension: '', filename: 'foo', pathname: ''}
> Ox.parsePath('.foo')
{extension: '', filename: '.foo', pathname: ''}
@*/
Ox.parsePath = function(str) {
var matches = /^(.+\/)?(.+?(\..+)?)?$/.exec(str);
return {
pathname: matches[1] || '',
filename: matches[2] || '',
extension: matches[3] ? matches[3].substr(1) : ''
};
}
/*@
Ox.parseSRT <f> Parses an srt subtitle file
(str) -> <o> Parsed subtitles
in <n> In point (sec)
out <n> Out point (sec)
text <s> Text
str <s> Contents of an srt subtitle file
> Ox.parseSRT('1\n01:02:00,000 --> 01:02:03,400\nHello World')
[{'in': 3720, out: 3723.4, text: 'Hello World'}]
@*/
Ox.parseSRT = function(str, fps) {
return str.replace(/\r\n/g, '\n').replace(/\n+$/, '').split('\n\n')
.map(function(block) {
var lines = block.split('\n'), points;
lines.shift();
points = lines.shift().split(' --> ').map(function(point) {
return point.replace(',', ':').split(':')
.reduce(function(prev, curr, i) {
return prev + parseInt(curr, 10) *
[3600, 60, 1, 0.001][i];
}, 0);
});
if (fps) {
points = points.map(function(point) {
return Math.round(point * fps) / fps;
});
}
return {
'in': points[0],
out: points[1],
text: lines.join('\n')
};
});
};
Ox.parseUserAgent = function(str) {
var names = {
chromeframe: 'Internet Explorer (Chrome Frame)',
'iPhone OS': 'iOS',
Fennec: 'Mobile Firefox',
Mobile: 'Mobile Safari',
MSIE: 'Internet Explorer',
},
regexps = {
browser: [
/(chromeframe)\/(\d+)/,
/(Chrome)\/(\d+)/,
/(Fennec)\/(\d+)/,
/(Firefox)\/(\d+)/,
/(MSIE)\/(\d+)/,
/(Opera)\/.+Version\/(\d+)/,
/Version\/(\d+).+(Mobile)\/.+Safari/,
/Version\/(\d+).+(Safari)/
],
system: [
/(iPhone OS) (\d+)/,
/ (Linux) /,
/(Mac OS X) (10.\d)/,
/(Windows) (NT \d\.\d)/
]
},
userAgent = {
browser: {name: '', version: ''},
system: {name: '', version: ''}
},
versions = {
'10.3': 'Panther',
'10.4': 'Tiger',
'10.5': 'Leopard',
'10.6': 'Snow Leopard',
'10.7': 'Lion',
'NT 5.0': '2000',
'NT 5.1': 'XP',
'NT 5.2': '2003',
'NT 6.0': 'Vista',
'NT 6.1': '7'
}
Ox.forEach(regexps, function(regexps, key) {
regexps.forEach(function(regexp) {
var matches = str.match(regexp), name, swap, version;
if (matches) {
matches[2] = matches[2] || '';
swap = matches[1].match(/^\d+$/);
name = matches[swap ? 2 : 1];
version = matches[swap ? 1 : 2].replace('_', '.');
userAgent[key] = {
name: names[name] || name,
version: versions[version] || version
};
return false;
}
});
});
return userAgent;
};
/*@
Ox.repeat <f> Repeat a value multiple times
Works for arrays, numbers and strings
> Ox.repeat(1, 3)
"111"
> Ox.repeat("foo", 3)
"foofoofoo"
> Ox.repeat([1, 2], 3)
[1, 2, 1, 2, 1, 2]
> Ox.repeat([{k: "v"}], 3)
[{k: "v"}, {k: "v"}, {k: "v"}]
@*/
Ox.repeat = function(val, num) {
var ret;
if (Ox.isArray(val)) {
ret = [];
num >= 1 && Ox.loop(num, function() {
ret = Ox.merge(ret, val);
});
} else {
ret = num >= 1 ? new Array(num + 1).join(val.toString()) : '';
}
return ret;
};
/*@
Ox.reverse <f> Reverses a string
> Ox.reverse('foobar')
'raboof'
@*/
Ox.reverse = function(str) {
return str.toString().split('').reverse().join('');
};
/*@
Ox.startsWith <f> Checks if a string starts with a given substring
If the substring is a string literal (and not a variable),
<code>/^sub/.test(str)</code> or <code>!!/^sub/.exec(str)</code>
is shorter than <code>Ox.starts(str, sub)</code>.
> Ox.startsWith('foobar', 'foo')
true
@*/
Ox.starts = Ox.startsWith = function(str, sub) {
// fixme: rename to starts
return str.substr(0, sub.length) == sub;
};
/*@
Ox.stripTags <f> Strips HTML tags from a string
> Ox.stripTags('f<span>o</span>o')
'foo'
@*/
Ox.stripTags = function(str) {
return str.replace(/<.*?>/g, '');
};
/*@
Ox.toCamelCase <f> Takes a string with '-', '/' or '_', returns a camelCase string
> Ox.toCamelCase('foo-bar-baz')
'fooBarBaz'
> Ox.toCamelCase('foo/bar/baz')
'fooBarBaz'
> Ox.toCamelCase('foo_bar_baz')
'fooBarBaz'
@*/
Ox.toCamelCase = function(str) {
return str.replace(/[\-\/_][a-z]/g, function(str) {
return str[1].toUpperCase();
});
};
/*@
Ox.toDashes <f> Takes a camelCase string, returns a string with dashes
> Ox.toDashes('fooBarBaz')
'foo-bar-baz'
@*/
Ox.toDashes = function(str) {
return str.replace(/[A-Z]/g, function(str) {
return '-' + str.toLowerCase();
});
};
/*@
Ox.toSlashes <f> Takes a camelCase string, returns a string with slashes
> Ox.toSlashes('fooBarBaz')
'foo/bar/baz'
@*/
Ox.toSlashes = function(str) {
/*
*/
return str.replace(/[A-Z]/g, function(str) {
return '/' + str.toLowerCase();
});
};
/*@
Ox.toTitleCase <f> Returns a string with capitalized words
> Ox.toTitleCase('foo')
'Foo'
> Ox.toTitleCase('Apple releases iPhone, IBM stock plummets')
'Apple Releases iPhone, IBM Stock Plummets'
@*/
Ox.toTitleCase = function(str) {
return Ox.map(str.split(' '), function(val) {
var sub = val.substr(1),
low = sub.toLowerCase();
if (sub == low) {
val = val.substr(0, 1).toUpperCase() + low;
}
return val;
}).join(' ');
};
/*@
Ox.toUnderscores <f> Takes a camelCase string, returns string with underscores
> Ox.toUnderscores('fooBarBaz')
'foo_bar_baz'
@*/
Ox.toUnderscores = function(str) {
return str.replace(/[A-Z]/g, function(str) {
return '_' + str.toLowerCase();
});
};
Ox.trim = function(str) { // is in jQuery, and in JavaScript itself
/*
Ox.trim(" foo ")
"foo"
*/
return str.replace(/^\s+|\s+$/g, '');
};
/*@
Ox.truncate <f> Truncate a string to a given length
(string, length) <s> Truncated string
(string, length, position) -> <s> Truncated string
(string, length, placeholder) -> <s> Truncated string
(string, length, position, placeholder) -> <s> Truncated string
> Ox.truncate('anticonstitutionellement', 16)
'anticonstitut...'
> Ox.truncate('anticonstitutionellement', 16, '...', 'left')
'...utionellement'
> Ox.truncate('anticonstitutionellement', 16, '>')
'anticonstitutio>'
> Ox.truncate('anticonstitutionellement', 16, '...', 'center')
'anticon...lement'
@*/
Ox.truncate = function(str, len, pad, pos) {
var pad = pad || '...',
pos = pos || 'right',
strlen = str.length,
padlen = pad.length,
left, right;
if (strlen > len) {
if (pos == 'left') {
str = pad + str.substr(padlen + strlen - len);
} else if (pos == 'center') {
left = Math.ceil((len - padlen) / 2);
right = Math.floor((len - padlen) / 2);
str = str.substr(0, left) + pad + str.substr(-right);
} else if (pos == 'right') {
str = str.substr(0, len - padlen) + pad;
}
}
return str;
};
/*@
Ox.words <f> Splits a string into words, removing punctuation
(string) -> <[s]> Array of words
string <s> Any string
> Ox.words('Let\'s "split" array-likes into key/value pairs--okay?')
["let's", "split", "array-likes", "into", "key", "value", "pairs", "okay"]
@*/
Ox.words = function(str) {
var arr = str.toLowerCase().split(/\b/),
chr = "-'",
len = arr.length,
startsWithWord = /\w/.test(arr[0]);
arr.forEach(function(v, i) {
// find single occurrences of "-" or "'" that are not at the beginning
// or end of the string, and join the surrounding words with them
if (
i > 0 && i < len - 1
&& v.length == 1 && chr.indexOf(v) > -1
) {
arr[i + 1] = arr[i - 1] + arr[i] + arr[i + 1];
arr[i - 1] = arr[i] = '';
}
});
// remove elements that have been emptied above
arr = arr.filter(function(v) {
return v.length;
});
// return words, not spaces or punctuation
return arr.filter(function(v, i) {
return i % 2 == !startsWithWord;
});
}
/*@
Ox.wordwrap <f> Wrap a string at word boundaries
> Ox.wordwrap("Anticonstitutionellement, Paris s'eveille", 25, '<br/>')
"Anticonstitutionellement, <br/>Paris s'eveille"
> Ox.wordwrap("Anticonstitutionellement, Paris s'eveille", 16, '<br/>')
"Anticonstitution<br/>ellement, Paris <br/>s'eveille"
> Ox.wordwrap('These are short words', 16, '<br/>', true)
'These are <br/>short words'
@*/
Ox.wordwrap = function(str, len, sep, bal, spa) {
// fixme: bad API, sep/bal/spa should be in options object
var str = str === null ? '' : str.toString(),
len = len || 80,
sep = sep || '<br/>',
bal = bal || false,
spa = Ox.isUndefined(spa) ? true : spa,
words = str.split(' '),
lines;
if (bal) {
// balance lines: test if same number of lines
// can be achieved with a shorter line length
lines = Ox.wordwrap(str, len, sep, false).split(sep);
if (lines.length > 1) {
// test shorter line, unless
// that means cutting a word
var max = Ox.max(Ox.map(words, function(word) {
return word.length;
}));
while (len > max) {
len--;
if (Ox.wordwrap(str, len, sep, false).split(sep).length > lines.length) {
len++;
break;
}
}
}
}
lines = [''];
Ox.forEach(words, function(word) {
if ((lines[lines.length - 1] + word + ' ').length <= len + 1) {
// word fits in current line
lines[lines.length - 1] += word + ' ';
} else {
if (word.length <= len) {
// word fits in next line
lines.push(word + ' ');
} else {
// word is longer than line
var chr = len - lines[lines.length - 1].length;
lines[lines.length - 1] += word.substr(0, chr);
for (var pos = chr; pos < word.length; pos += len) {
lines.push(word.substr(pos, len));
}
lines[lines.length - 1] += ' ';
}
}
});
if (!spa) {
lines = Ox.map(lines, function(line) {
return Ox.trim(line);
});
}
return Ox.trim(lines.join(sep));
};