mirror of
https://github.com/gryf/pentadactyl-pm.git
synced 2026-02-05 20:05:45 +01:00
6a25312c7d
* Standard module format. All modules are explicitly declared as modules, they're created via a constructor and instantiated automatically. They're dependency aware. They stringify properly. * Classes are declared the same way (rather like Structs already were). They also stringify properly. Plus, each instance has a rather nifty closure member that closes all of its methods around 'this', so you can pass them to map, forEach, setTimeout, etc. Modules are themselves classes, with a special metaclass, as it were. * Doug Crockford is dead, metaphorically speaking. Closure-based classes just don't fit into any of the common JavaScript frameworks, and they're inefficient and confusing. Now, all class and module members are accessed explicitly via 'this', which makes it very clear that they're class members and not (e.g.) local variables, without anything nasty like Hungarian notation. * Strictly one module per file. Classes that belong to a module live in the same file. * For the moment, there are quite a few utility functions sitting in base.c, because my class implementation used them, and I haven't had the time or inclination to sort them out. I plan to reconcile them with the current mess that is the util namespace. * Changed bracing style.
A terse introduction to E4X
Public Domain
The inline XML literals in this code are part of E4X, a standard
XML processing interface for ECMAScript. In addition to syntax
for XML literals, E4X provides a new kind of native object,
"xml", and a syntax, similar to XPath, for accessing and
modifying the tree. Here is a brief synopsis of the kind of
usage you'll see herein:
> let xml =
<foo bar="baz" baz="qux">
<bar>
<baz id="1"/>
</bar>
<baz id="2"/>
</foo>;
// Select all bar elements of the root foo element
> xml.bar
<bar><baz id="1"/></bar>
// Select all baz elements anywhere beneath the root
> xml..baz
<baz id="1"/>
<baz id="2"/>
// Select all of the immediate children of the root
> xml.*
<bar><baz id="1"/></bar>
<baz id="2"/>
// Select the bar attribute of the root node
> xml.@bar
baz
// Select all id attributes in the tree
> xml..@id
1
2
// Select all attributes of the root node
> xml.@*
baz
quz
// Add a quux elemend beneath the first baz
> xml..baz[0] += <quux/>
<baz id="1"/>
<quux/>
> xml
<foo bar="baz" baz="qux">
<bar>
<baz id="1"/>
<quux/>
</bar>
<baz id="2"/>
</foo>
// and beneath the second
> xml.baz[1] = <quux id="1"/>
> xml
<foo bar="baz" baz="qux">
<bar>
<baz id="1"/>
<quux/>
</bar>
<baz id="2"/>
<quux id="1"/>
</foo>
// Replace bar's subtree with a foo element
> xml.bar.* = <foo id="1"/>
> xml
<foo bar="baz" baz="qux">
<bar>
<foo id="1"/>
</bar>
<baz id="2"/>
<quux id="1"/>
</foo>
// Add a bar below bar
> xml.bar.* += <bar id="1"/>
<foo id="1"/>
<bar id="1"/>
> xml
<foo bar="baz" baz="qux">
<bar>
<foo id="1"/>
<bar id="1"/>
</bar>
<baz id="2"/>
<quux id="1"/>
</foo>
// Adding a quux attribute to the root
> xml.@quux = "foo"
foo
> xml
<foo bar="baz" baz="qux" quux="foo">
<bar>
<foo id="1"/>
<bar id="1"/>
</bar>
<baz id="2"/>
<quux id="1"/>
</foo>
> xml.bar.@id = "0"
> xml..foo[0] = "Foo"
Foo
> xml..bar[1] = "Bar"
Bar
> xml
js> xml
<foo bar="baz" baz="qux" quux="foo" id="0">
<bar id="0">
<foo id="1">Foo</foo>
<bar id="1">Bar</bar>
</bar>
<baz id="2"/>
<quux id="1"/>
</foo>
// Selecting all bar elements where id="1"
> xml..bar.(@id == 1)
Bar
// Literals:
// XMLList literal. No root node.
> <>Foo<br/>Baz</>
Foo
<br/>
Baz
// Interpolation.
> let x = "<foo/>"
> <foo bar={x}>{x + "<?>"}</foo>
<foo/><?>
> <foo bar={x}>{x + "<?>"}</foo>.toXMLString()
<foo bar="<foo/>"><foo/><?></foo>
> let x = <foo/>
> <foo bar={x}>{x}</foo>.toXMLString()
<foo bar="">
<foo/>
</foo>