Xpath金手指

最近在搞puppeteer的自动化，有很多地方需要用到CSS Selector，但是某些地方又是用Xpath方便，金手指放一下，备查：

Test queries in the Xpath test bed:

Xpath test bed (whitebeam.org)

Browser console

1$x("//div")

Works in Firefox and Chrome.

# Selectors

Descendant selectors

`h1`	`//h1`
`div p`	`//div//p
`ul > li`	`//ul/li
`ul > li > a`	`//ul/li/a`
`div > *`	`//div/*`
`:root`	`/
`:root > body`	`/body`

Attribute selectors

`#id`	`//*[@id="id"]`
`.class`	`//[@class="class"]` …kinda*
`input[type="submit"]`	`//input[@type="submit"]`
`a#abc[for="xyz"]`	`//a[@id=“abc”][@for=“xyz”]
`a[rel]`	`//a[@rel]`
`a[href^='/']`	`//a[starts-with(@href, ‘/’)]
`a[href$='pdf']`	`//a[ends-with(@href, '.pdf')]`
`a[href*='://']`	`//a[contains(@href, '://')]`
`a[rel~='help']`	`//a[contains(@rel, 'help')]` …kinda

Order selectors

`ul > li:first-of-type`	`//ul/li[1]
`ul > li:nth-of-type(2)`	`//ul/li[2]`
`ul > li:last-of-type`	`//ul/li[last()]`
`li#id:first-of-type`	`//li[1][@id=“id”]
`a:first-child`	`//*[1][name()="a"]`
`a:last-child`	`//*[last()][name()="a"]`

Siblings

`h1 ~ ul`	`//h1/following-sibling::ul
`h1 + ul`	`//h1/following-sibling::ul[1]`
`h1 ~ #id`	`//h1/following-sibling::[@id="id"]`

jQuery

`$('ul > li').parent()`	`//ul/li/..
`$('li').closest('section')`	`//li/ancestor-or-self::section`
`$('a').attr('href')`	`//a/@href
`$('span').text()`	`//span/text()`

Other things

`h1:not([id])`	`//h1[not(@id)]
Text match	`//button[text()=“Submit”]
Text match (substring)	`//button[contains(text(),"Go")]`
Arithmetic	`//product[@price > 2.50]`
Has children	`//ul[*]`
Has children (specific)	`//ul[li]`
Or logic	`//a[@name or @href]
Union (joins results)	`//a \| //div

Class check

1//div[contains(concat(' ',normalize-space(@class),' '),' foobar ')]

Xpath doesn’t have the “check if part of space-separated list” operator, so this is the workaround (source ).

# Expressions

Steps and axes

`//`	`ul`	`/`	`a[@id='link']`
Axis	Step	Axis	Step

Prefixes

Prefix	Example	What
`//`	`//hr[@class='edge']`	Anywhere
`./`	`./a`	Relative
`/`	`/html/body/div`	Root

Begin your expression with any of these.

Axes

Axis	Example	What
`/`	`//ul/li/a`	Child
`//`	`//[@id="list"]//a`	Descendant

Separate your steps with /. Use two (//) if you don’t want to select direct children.

Steps

1//div
2//div[@name='box']
3//[@id='link']

A step may have an element name (div) and predicates ([...]). Both are optional. They can also be these other things:

1//a/text()     #=> "Go home"
2//a/@href      #=> "index.html"
3//a/*          #=> All a's child elements

# Predicates

Predicates

1//div[true()]
2//div[@class="head"]
3//div[@class="head"][@id="top"]

Restricts a nodeset only if some condition is true. They can be chained.

Operators

1# Comparison
2//a[@id = "xyz"]
3//a[@id != "xyz"]
4//a[@price > 25]
5# Logic (and/or)
6//div[@id="head" and position()=2]
7//div[(x and y) or not(z)]

Use comparison and logic operators to make conditionals.

Using nodes

1# Use them inside functions
2//ul[count(li) > 2]
3//ul[count(li[@class='hide']) > 0]
4# This returns `<ul>` that has a `<li>` child
5//ul[li]

You can use nodes inside predicates.

Indexing

1//a[1]                  # first <a>
2//a[last()]             # last <a>
3//ol/li[2]              # second <li>
4//ol/li[position()=2]   # same as above
5//ol/li[position()>1]   # :not(:first-of-type)

Use [] with a number, or last() or position().

Chaining order

1a[1][@href='/']
2a[@href='/'][1]

Order is significant, these two are different.

Nesting predicates

1//section[.//h1[@id='hi']]

This returns <section> if it has an <h1> descendant with id='hi'.

# Functions

Node functions

1name()                     # //[starts-with(name(), 'h')]
2text()                     # //button[text()="Submit"]
3                           # //button/text()
4lang(str)
5namespace-uri()
6count()                    # //table[count(tr)=1]
7position()                 # //ol/li[position()=2]

Boolean functions

1not(expr)                  # button[not(starts-with(text(),"Submit"))]

String functions

 1contains()                 # font[contains(@class,"head")]
 2starts-with()              # font[starts-with(@class,"head")]
 3ends-with()                # font[ends-with(@class,"head")]
 4concat(x,y)
 5substring(str, start, len)
 6substring-before("01/02", "/")  #=> 01
 7substring-after("01/02", "/")   #=> 02
 8translate()
 9normalize-space()
10string-length()

Type conversion

1string()
2number()
3boolean()

# Axes

Using axes

1//ul/li                       # ul > li
2//ul/child::li                # ul > li (same)
3//ul/following-sibling::li    # ul ~ li
4//ul/descendant-or-self::li   # ul li
5//ul/ancestor-or-self::li     # $('ul').closest('li')

Steps of an expression are separated by /, usually used to pick child nodes. That’s not always true: you can specify a different “axis” with ::.

`//`	`ul`	`/child::`	`li`
Axis	Step	Axis	Step

Child axis

1# both the same
2//ul/li/a
3//child::ul/child::li/child::a

child:: is the default axis. This makes //a/b/c work.

1# both the same
2# this works because `child::li` is truthy, so the predicate succeeds
3//ul[li]
4//ul[child::li]
5# both the same
6//ul[count(li) > 2]
7//ul[count(child::li) > 2]

Descendant-or-self axis

1# both the same
2//div//h4
3//div/descendant-or-self::h4

// is short for the descendant-or-self:: axis.

1# both the same
2//ul//[last()]
3//ul/descendant-or-self::[last()]

Other axes

Axis	Abbrev	Notes
`ancestor`
`ancestor-or-self`
`attribute`	`@`	`@href` is short for `attribute::href`
`child`		`div` is short for `child::div`
`descendant`
`descendant-or-self`	`//`	`//` is short for `/descendant-or-self::node()/`
`namespace`
`self`	`.`	`.` is short for `self::node()`
`parent`	`..`	`..` is short for `parent::node()`
`following`
`following-sibling`
`preceding`
`preceding-sibling`

There are other axes you can use.

Unions

1//a | //span

Use | to join two expressions.

# More examples

Examples

1//*                 # all elements
2count(//*)          # count all elements
3(//h1)[1]/text()    # text of the first h1 heading
4//li[span]          # find a <li> with an <span> inside it
5                    # ...expands to //li[child::span]
6//ul/li/..          # use .. to select a parent

Find a parent

1//section[h1[@id='section-name']]

Finds a <section> that directly contains h1#section-name

1//section[//h1[@id='section-name']]

Finds a <section> that contains h1#section-name. (Same as above, but uses descendant-or-self instead of child)

Closest

1./ancestor-or-self::[@class="box"]

Works like jQuery’s $().closest('.box').

Attributes

1//item[@price > 2*@discount]

Finds <item> and check its attributes

# References

Xpath test bed (whitebeam.org)