zu

Functional DOM grokking with CSS style selectors

aus außer bei mit nach seit von zu

Motivation

When reading XML/markup data on the server, I often reach for a jQuery-esque tool using CSS selectors to extract various bits of the document. There are solutions solving that problem already, however jQuery is an acquired taste, and with my latest foragings into functional programming, those API:s just feel awkward.

Installing with NPM

npm install -S zu

Use in a project

var zu = require('zu');
 
var nodes = zu.parse(<some xml string>);
var subs  = zu.find(nodes, 'div#a > span.b');

Goals

• Composable curried functions, not methods.

• Selectors return Arrays, not "array-like objects". This means functional libraries such as fnuc and Ramda can interoperate.

• Reading, not manipulating. This cuts out a lot of code as well as avoids awkward jQuery style dual-purpose functions (get/set).

• Performance. Especially matching should be fast.

• XML bias. HTML is the odd one out. Namespaces do not need escaping in selector expressions (zu.find(nodes, 'e:event')).

Example

var zu = require('zu');
var fs = require('fs');
 
// parse an xml document to array of nodes.
var xml = zu.parseXml(fs.readFileSync('./mydoc.xml', 'utf-8'));
 
// partial application to make function that does a find over the xml.
var fn = zu.findWith(xml);
 
// find some elements. this is just a plain array of
// nodes. the nodes are inspected using the api.
var events = fn('e:eventList > e:event');
 
// use regular Array.prototype.map
var myresult = events.map(function(ev) {
 
    // function bound to the events.
    var efn = zu.findWith(ev);
 
    // extract some interesting data.
    return {
        eventId:   zu.text(efn('e:eventId')),
        title:     zu.text(efn('e:title')),
        startTime: zu.text(efn('e:startTime')),
        eventXml:  zu.xml(ev)     // event as an xml string
    };
 
});

The DOM nodes

Most functions find, closest, next etc accept and return an array of nodes. Where the API accepts an array of nodes in, you can also use a single node without a wrapping array as well. The return type will however always be an array.

var html = parseHtml(someStr);
 
var dom1 = zu.find(html, 'div');
var dom2 = zu.find(html[0], 'div'); // we can use non-array
 
// dom1 and dom2 are different arrays
// with the same nodes inside.

The node's datastructure is considered opaque (it is htmlparser2 domhandler object). This implementation may change, so accessing the internal of those objects may be breaking. The way to get data out are:

• zu.text(node or nodes)
• zu.attr(node or nodes, attrname)
• zu.xml(node or nodes)
• zu.html(node or nodes)

Curry

Each zu.[something](n,e) that takes two arguments also have two curried version zu.[something]With(n) and zu.[something]With(e) Example illustrated with zu.parent.

Both arguments

:: [n], s -> [n]

No curry here.

zu.parent(nodes, exp);

One argument

:: [n] -> [n]

Equivalent to having null as second argument.

zu.parent(nodes);

Partially applied with nodes

:: [n] -> s -> [n]

Provided an array of nodes, gives a function expecting the expression.

fn = zu.parentWith(nodes);
fn(exp);

Partially applied with expression

:: s -> [n] -> [n]

Provided an expression, gives a function expecting an array of nodes.

fn = zu.parentWith(exp);
fn(nodes);

CSS Selectors

The following selectors are (currently) supported:

• Type selector div
• All selector *
• Descendant div span
• Child div > span. Also works as > span for immediate children.
• Class div.foo
• Id div#bar
• Namespace e|div, but also e:div for names not clashing with pseudo-classes.
• Attributes with or without quotes
  • exists div[foo]
  • equals div[foo=bar]
  • whitespace separated match div[foo~=bar]
  • hyphenated start match div[foo|=bar]
  • starts with div[foo^=bar]
  • ends with div[foo$=bar]
  • substring div[foo*=bar]
• Pseudo-classes
  • contains specified text :contains()
  • that have no children :empty
  • first child of parent :first-child
  • last child of parent :last-child

API

Parsing

parseXml

zu.parseXml(str)

:: s -> [n]

Parse an XML string to an array of nodes.

parseHtml

zu.parseHtml(str)

Parse an HTML string to an array of nodes. The difference from XML is that certain HTML elements get special treatment. <script> contents are not further parsed, <img> tags does not need closing etc.

:: s -> [n]

Data out

xml

zu.xml(nodes)

Turn given array of nodes into a string XML form.

:: [n] -> s

html

zu.html(nodes)

Turn given array of nodes into a string HTML form.

:: [n] -> s

text

zu.text(nodes)

Turn given array of text nodes into a string, where the contents of each node is concatenated together.

:: [n] -> s

attr

zu.attr(nodes, name)

Return the attribute value for name from the first element in the given array of nodes.

:: [n], s -> s

Also zu.attrWith(nodes or name)

:: [n] -> s -> s
:: s -> [n] -> s

attrList

zu.attr(nodes, name)

Enumerates the attribute names for the first element in the given array of nodes.

:: [n] -> [s]

hasClass

Test whether any node in the given array of nodes has a name as a class.

zu.hasClass(nodes, name)

:: [n], s -> s

Also zu.hasClassWith(nodes or name)

:: [n] -> s -> s
:: s -> [n] -> s

Selectors

find

zu.find(nodes, exp)

Match the given nodes, and any descendants of the given nodes against the given expression.

:: [n], s -> [n]

Also zu.findWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

children

Match the immediate children of the given nodes against the given expression. Also zu.children(nodes) to get immediate children without any filtering expression.

zu.children(nodes, exp)

:: [n], s -> [n]
:: [n] -> [n]

Also zu.childrenWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

closest

zu. closest(nodes, exp)

Test the given set of nodes and recursively parent nodes against expression and return the first match.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.closestWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

filter

zu.filter(nodes, exp)

Filter the given set of nodes using the expression. Also zu.filter(nodes) just returns the same nodes (however in a new array).

:: [n], s -> [n]
:: [n] -> [n]

Also filterWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

is

zu.is(nodes, exp)

Filter the given set of nodes using the expression and tell whether any matched.

:: [n], s -> bool
:: [n] -> bool

Also zu.isWith(nodes or exp)

:: s -> [n] -> bool
:: [n] -> s -> bool

next

zu.next(nodes, exp)

Select immediate sibling nodes to the right of given nodes, optionally apply a filter expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.nextWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

nextAll

zu.nextAll(nodes, exp)

Select all sibling nodes to the right of the given nodes, optionally filtered by an expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.nextAllWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

parent

zu.parent(nodes, exp)

Select immediate parent nodes of given nodes, optionally filtered by an expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.parentWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

parents

zu.parents(nodes, exp)

Select all parent nodes of given nodes, optionally filtered by an expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.parentsWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

prev

zu.prev(nodes, exp)

Select immediate sibling nodes to the left of given nodes, optionally apply a filter expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.prevWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

prevAll

zu.prevAll(nodes, exp)

Select all sibling nodes to the left of the given nodes, optionally filtered by an expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.prevAllWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

siblings

zu.siblings(nodes, exp)

Select all sibling nodes both to the left and right, optionally filtered by an expression.

:: [n], s -> [n]
:: [n] -> [n]

Also zu.siblingsWith(nodes or exp)

:: s -> [n] -> [n]
:: [n] -> s -> [n]

License

The MIT License (MIT)

Copyright © 2015 Martin Algesten

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.