A user account is required in order to edit this wiki, but we've had to disable public user registrations due to spam.
To request an account, ask an autoconfirmed user on Chat (such as one of these permanent autoconfirmed members).
Web ECMAScript: Difference between revisions
No edit summary |
|||
(20 intermediate revisions by 7 users not shown) | |||
Line 1: | Line 1: | ||
This page is for documenting the differences between | {{obsolete|spec=[http://javascript.spec.whatwg.org/ JavaScript, aka. Web ECMAScript]}} | ||
This page is for documenting the differences between ES5 specification and the requirements for ECMAScript implementations in web browsers. | |||
'''It is now maintained as specification: [http://javascript.spec.whatwg.org/ JavaScript, aka. Web ECMAScript]''' | |||
== Identifiers == | == Identifiers == | ||
Line 97: | Line 101: | ||
When called with fewer than 2 arguments Date.UTC must return NaN. | When called with fewer than 2 arguments Date.UTC must return NaN. | ||
=== toString === | |||
TODO | |||
== Global scope == | == Global scope == | ||
ES5 claims the global scope "this" is the same as the global object, which is not always true in HTML5. | ES5 claims the global scope "this" is the same as the global object, which is not always true in HTML5. | ||
== var statements == | |||
The erratum in https://bugs.ecmascript.org/show_bug.cgi?id=78#c0 needs to be followed so that var statements at the top level of scripts can shadow any properties from the global object's prototype chain. | |||
== Eval == | == Eval == | ||
Use of eval not called "eval". | Use of eval not called "eval". Should work but implementations differ on the scope (Spidermonkey first tries global and then local if the object as not found globally, JScript uses local scope, others use global scope), may not have strong compat requirements | ||
eval.apply(this, code) should work but scope again varies when "this" is not the global object | eval.apply(this, code) should work but scope again varies when "this" is not the global object | ||
=== Eval and Global Scopes === | |||
See: http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2011-March/030798.html | |||
== Date Parsing == | == Date Parsing == | ||
Line 119: | Line 133: | ||
In JScript, the following are ignored if they are at the end of the script (i.e. just followed by whitespace lines): either (1) a line that starts with just whitespace and comments and consists of "-->" followed by anything except new lines and finally followed by another "-->", or (2) one or two lines that start with just whitespace and comments and consist of just "-->". --> at the end of the last line of the script causes the line to be ignored unless the "-->" occurs within a comment. The last non-whitespace line of the script is ignored if it ends with "-->" and doesn't contain "//" or "<!--" (but not in eval). | In JScript, the following are ignored if they are at the end of the script (i.e. just followed by whitespace lines): either (1) a line that starts with just whitespace and comments and consists of "-->" followed by anything except new lines and finally followed by another "-->", or (2) one or two lines that start with just whitespace and comments and consist of just "-->". --> at the end of the last line of the script causes the line to be ignored unless the "-->" occurs within a comment. The last non-whitespace line of the script is ignored if it ends with "-->" and doesn't contain "//" or "<!--" (but not in eval). | ||
In SpiderMonkey, "-->" on any line that starts with just whitespace and comments is treated as a line comment (same as //). (Also, ;version=1.6, 1.7 or 1.8 or ;e4x=1 enables E4X <!-- --> comments.) | In SpiderMonkey, "-->" on any line that starts with just whitespace and comments is treated as a line comment (same as //). (Also, ;version=1.6, 1.7 or 1.8 or ;e4x=1 enables E4X <!-- --> comments.) | ||
In Carakan | In Carakan, "-->" on any line that starts with just whitespace (but not comments) is treated as a line comment (same as //). | ||
Chakra (IE) doesn’t support these syntax extensions. | |||
== Property Enumeration == | == Property Enumeration == | ||
Enumeration of objects is in insertion order (but host property order compared to user defined property order does not seem to be significant) | Enumeration of objects is in insertion order (but host property order compared to user defined property order does not seem to be significant). Property order seems to survive the use of "delete" (i.e. removing then readding a property doesn't change its position), at least in Chrome and Firefox. | ||
== Object Properties == | == Object Properties == | ||
All objects have a mutable __proto__ property that is a reference to the prototype of the object. Note that ES5 defines that objects with extensible:false must not have their prototype mutated. In the case that setting an objects __proto__ would cause a prototype chain to become cyclic, the setter must fail and throw Error(). | All objects have a mutable __proto__ property that is a reference to the prototype of the object. Note that ES5 defines that objects with extensible:false must not have their prototype mutated. In the case that setting an objects __proto__ would cause a prototype chain to become cyclic, the setter must fail and throw Error(). | ||
Note: __proto__ will be fully specced in ES6. | |||
== Getters and Setters == | == Getters and Setters == | ||
Line 146: | Line 160: | ||
TODO http://www.w3.org/mid/[email protected] | TODO http://www.w3.org/mid/[email protected] | ||
== Also see == | |||
* http://kangax.github.com/es5-compat-table/non-standard/ | |||
* https://bugs.ecmascript.org/buglist.cgi?product=ECMA-262&component=technical%20content&resolution=--- – open bugs on the ES spec | |||
[[Category:Spec_coordination]] |
Latest revision as of 16:54, 26 November 2012
This document is obsolete.
For the current specification, see: JavaScript, aka. Web ECMAScript
This page is for documenting the differences between ES5 specification and the requirements for ECMAScript implementations in web browsers.
It is now maintained as specification: JavaScript, aka. Web ECMAScript
Identifiers
(this is very rough)
Identifiers containing escape sequences are not equivalent to fully unescaped identifiers in the case that, after fully unescaping identifier, it is a ReservedWord. In particular it is possible to create Identifiers that unescape to a reserved word so long as at least one character is fully escaped. Subsequent use of such identifiers must also have at least one character escaped (otherwise the reserved word will be used instead) but it need not be the same character(s) as that originally used to create the identifier.
15.5.4 Properties of the String Prototype Object
Several extra methods are found on String.prototype for wrapping text in HTML elements (these are all generic; the this value need not be a String object):
Algorithm ToHTMLTag(tag_name, content, attribute_name, attribute_value):
- if attribute_name is undefined return "<" + tag_name + ">" + content + "</" + tag_name + ">"
- otherwise return "<" + tag_name + " " + attribute_name + "=\"" + attribute_value + "\">" + content + "</" + tag_name + ">"
String.prototype.anchor(name)
- Let content be ToString(this)
- Let attribute_value be ToString(name)
- Return ToHTMLTag("a", content, "name", attribute_value)
String.prototype.big()
- Let content be ToString(this)
- Return ToHTMLTag("big", content)
String.prototype.blink()
- Let content be ToString(this)
- Return ToHTMLTag("blink", content)
String.prototype.bold()
- Let content be ToString(this)
- Return ToHTMLTag("b", content)
String.prototype.fixed()
- Let content be ToString(this)
- Return ToHTMLTag("tt", content)
String.prototype.fontcolor(color)
- Let content be ToString(this)
- Let attribute_value be ToString(color)
- Return ToHTMLTag("font", content, "color", attribute_value)
String.prototype.fontsize(size)
- Let content be ToString(this)
- Let attribute_value be ToString(size)
- Return ToHTMLTag("font", content, "size", attribute_value)
String.prototype.italics()
- Let content be ToString(this)
- Return ToHTMLTag("i", content)
String.prototype.link(href)
- Let content be ToString(this)
- Let attribute_value be ToString(href)
- Return ToHTMLTag("a", content, "href", attribute_value)
String.prototype.small()
- Let content be ToString(this)
- Return ToHTMLTag("small", content)
String.prototype.strike()
- Let content be ToString(this)
- Return ToHTMLTag("strike", content)
String.prototype.sub()
- Let content be ToString(this)
- Return ToHTMLTag("sub", content)
String.prototype.sup()
- Let content be ToString(this)
- Return ToHTMLTag("sup", content)
RegExp
After a regexp is executed the RegExp constructor object has properties $1...$9 which are assigned the values of the first 9 match groups from the previous regexp. (more detail here)
DecimalEscapes in CharacterRanges all behave like \0 rather than throwing syntax errors i.e. /[\1-Z]/ will match any character with a codepoint between 0 and 90.
RegExp.prototype.compile changes the regexp in place. In Carakan/Nitro/V8 the method returns undefined; in SpiderMonkey it returns the regexp object.
Date
15.9.4.3 Date.UTC
When called with fewer than 2 arguments Date.UTC must return NaN.
toString
TODO
Global scope
ES5 claims the global scope "this" is the same as the global object, which is not always true in HTML5.
var statements
The erratum in https://bugs.ecmascript.org/show_bug.cgi?id=78#c0 needs to be followed so that var statements at the top level of scripts can shadow any properties from the global object's prototype chain.
Eval
Use of eval not called "eval". Should work but implementations differ on the scope (Spidermonkey first tries global and then local if the object as not found globally, JScript uses local scope, others use global scope), may not have strong compat requirements
eval.apply(this, code) should work but scope again varies when "this" is not the global object
Eval and Global Scopes
See: http://lists.whatwg.org/pipermail/whatwg-whatwg.org/2011-March/030798.html
Date Parsing
TODO:This is a mess
HTML comments
<!-- is a line comment (same as //)
In JScript, the following are ignored if they are at the end of the script (i.e. just followed by whitespace lines): either (1) a line that starts with just whitespace and comments and consists of "-->" followed by anything except new lines and finally followed by another "-->", or (2) one or two lines that start with just whitespace and comments and consist of just "-->". --> at the end of the last line of the script causes the line to be ignored unless the "-->" occurs within a comment. The last non-whitespace line of the script is ignored if it ends with "-->" and doesn't contain "//" or "<!--" (but not in eval).
In SpiderMonkey, "-->" on any line that starts with just whitespace and comments is treated as a line comment (same as //). (Also, ;version=1.6, 1.7 or 1.8 or ;e4x=1 enables E4X <!-- --> comments.)
In Carakan, "-->" on any line that starts with just whitespace (but not comments) is treated as a line comment (same as //).
Chakra (IE) doesn’t support these syntax extensions.
Property Enumeration
Enumeration of objects is in insertion order (but host property order compared to user defined property order does not seem to be significant). Property order seems to survive the use of "delete" (i.e. removing then readding a property doesn't change its position), at least in Chrome and Firefox.
Object Properties
All objects have a mutable __proto__ property that is a reference to the prototype of the object. Note that ES5 defines that objects with extensible:false must not have their prototype mutated. In the case that setting an objects __proto__ would cause a prototype chain to become cyclic, the setter must fail and throw Error().
Note: __proto__ will be fully specced in ES6.
Getters and Setters
* __lookupGetter__ * __lookupSetter__ * __defineGetter__ * __defineSetter__
foo.arguments
TODO http://www.w3.org/mid/[email protected]