The assertions from dxw

Use separate functions for whitespace modes

Currently it's not immediately obvious what the difference between these two function calls is:

$this->assertHTMLEquals($expected, $actual, true);
$this->assertHTMLEquals($expected, $actual, false);

So let's use two separate functions:

$this->assertHTMLEquals($expected, $actual);
$this->assertHTMLEqualsIgnoringWhitespace($expected, $actual);

To follow PHPUnit conventions, we could use assertHTMLEquals/assertHTMLSame.

Fix whitespace handling

Even outside of $ignoreWhitespace mode, IIRC in HTML all whitespace is equivalent. So " " and "\n" should be treated as equal.

Doesn't support any HTML elements invented after 1999

DOMDocument::loadHTML(): Tag article invalid in Entity, line: 3

Tab characters are being stripped

assertions/tests/html_test.php

Lines 40 to 43 in 18a9bef

    
           $h->assertHTMLEquals("<a href='aaa'>\nbbb   <br>\t</a>", '<a href="aaa">bbb<br></a>'); 
        
           $this->assertEquals([ 
        
               "<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.0 Transitional//EN\" \"http://www.w3.org/TR/REC-html40/loose.dtd\">\n<html><body><a href=\"aaa\">\nbbb   <br></a></body></html>\n",

Likely to be either a bug in my understanding of HTML, or PHP's understanding of HTML.

Implement \Dxw\Assertions\XML

API should be identical to \Dxw\Assertions\HTML.

Ideally we should convert both documents to Canonical XML.

Here's a bit of code we used on another project:

    protected function assertXMLEquals($expected, $actual)
    {
        call_user_func_array(
            [$this, 'assertEquals'],
            array_map(
                function ($x) {
                    $d = new DOMDocument;
                    $d->loadXML($x);

                    $t = $d->C14N();
                    $t = preg_replace('/^\s+/m', '', $t);
                    $t = preg_replace('/\n/', '', $t);
                    $t = preg_replace('/></', ">\n<", $t);
                    return $t;
                },
                [$expected, $actual]
            )
        );
    }

Allow empty strings

DOMDocument::loadHTML(): Empty string supplied as input

An empty string is a valid HTML fragment. So it should be allowed by this library.

Use a better HTML parser

This would probably fix several open issues.

This one may be a viable option: https://github.com/Masterminds/html5-php

Fix ignoreWhitespace mode

assertions/src/HTML.php

Lines 16 to 19 in 18a9bef

    
           if ($ignoreWhitespace) { 
        
               $out = preg_replace('/>\s+/m', '>', $out); 
        
               $out = preg_replace('/\s+</m', '<', $out); 
        
           }

This is just wrong. We should be able to iterate through the DOM and when we encounter a text node we should remove all leading/trailing whitespace.

Add negative assertions for existing assertions

i.e. There's a $this->assertRegExp() but no $this->assertNotRegExp().

Make the output nicer

The HTML should be pretty-printed before handing over to assertEquals() (and we could strip the DTD and other boilerplate HTML).

We should perhaps also directly call assertThat(), so that we can have it print "Failed asserting that two HTML fragments are equal."

Bad output:

2) CategoryWidgetTest::testGetOutput2
Failed asserting that two strings are equal.
--- Expected
+++ Actual
@@ @@
-'<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"><html><body><div class="category-widget"><div class="current-cat"><ul><li><a href="/category/meow/">Meowz</a><ul><li><a href="/category/mew/">meW</a></li><li><a href="/category/mew2/">mew2</a></li></ul></li></ul></div><div class="other-cats"><ul><li>Other contents<ul><li><a href="/category/meowth/">Meowthhhh</a></li><li><a href="/category/marth/">Marth</a></li></ul></li></ul></div></div></body></html>'
+'<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd"><html><body><div class="category-widget"><div class="current-cat"><ul><li><a href="/category/cat/">Cat</a><ul><li><a href="/category/oof/">Oof</a></li><li><a href="/category/woo/">Woo</a></li></ul></li></ul></div><div class="other-cats"><ul><li>Other contents<ul><li><a href="/category/cat/">Cat</a></li><li><a href="/category/cog/">Cog</a></li></ul></li></ul></div></div></body></html>'

dxw / assertions Goto Github PK

assertions's People

Contributors

Watchers

Forkers

assertions's Issues

Use separate functions for whitespace modes

Fix whitespace handling

Doesn't support any HTML elements invented after 1999

Tab characters are being stripped

Implement \Dxw\Assertions\XML

Allow empty strings

Use a better HTML parser

Fix ignoreWhitespace mode

Add negative assertions for existing assertions

Make the output nicer

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent

	$h->assertHTMLEquals("<a href='aaa'>\nbbb <br>\t</a>", '<a href="aaa">bbb<br></a>');

	$this->assertEquals([
	"<!DOCTYPE html PUBLIC \"-//W3C//DTD HTML 4.0 Transitional//EN\" \"http://www.w3.org/TR/REC-html40/loose.dtd\">\n<html><body><a href=\"aaa\">\nbbb <br></a></body></html>\n",

	if ($ignoreWhitespace) {
	$out = preg_replace('/>\s+/m', '>', $out);
	$out = preg_replace('/\s+</m', '<', $out);
	}