MensBeam
/
Microformats
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
Browse Source

Implement textContent parsing properly

master
J. King 11 months ago
parent
650d1c7e98
commit
65bbf61579
2 changed files with 80 additions and 41 deletions
Whitespace
Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. 92
      lib/Parser.php
  2. 29
      tests/cases/StandardTest.php

92
lib/Parser.php
View File

@ -313,7 +313,7 @@ class Parser {
}
// sort and clean rel microformats
foreach ($out['rels'] as $k => $v) {
$out['rels'][$k] = array_unique($v);
$out['rels'][$k] = array_values(array_unique($v));
}
foreach ($out['rel-urls'] as $k => $v) {
$out['rel-urls'][$k]['rels'] = array_unique($v['rels']);
@ -1040,7 +1040,19 @@ class Parser {
if ($this->options['basicTrim']) {
return $this->getCleanTextBasic($node, $prefix);
} else {
return $this->getCleanTextThorough($node, $prefix);
// https://microformats.org/wiki/textcontent-parsing
# Plain text of element
# To get the plain text for an Element input:
# Let output be the result of running [Element to string] on input
$output = $this->getCleanTextThorough($node, $prefix);
# Remove any sequence of one or more consecutive U+0020 SPACE code points directly before and after an U+000A LF code point from output
$output = preg_replace('/^\s+|\s+$/m', "", $output);
# Strip leading and trailing ASCII whitespace from output
$output = trim($output);
# Replace any sequence of one or more consecutive U+0020 SPACE code points in output with a single U+0020 SPACE code point
$output = preg_replace('/ {2,}/', " ", $output);
# Return output
return $output;
}
}
@ -1060,7 +1072,7 @@ class Parser {
// NOTE: This ought to include FORM FEED characters
$value = strtr($value, "\t\n\r\f", " ");
# Append value to output
$output .= $value;
$output[] = $value;
} elseif ($n instanceof \DOMElement) {
# If child is an Element, switch on its tagName:
// NOTE: we switch on localName instead to avoid silly case folding
@ -1068,37 +1080,59 @@ class Parser {
case "script":
case "style":
case "template":
# SCRIPT
# STYLE
// TEMPLATE as well
# SCRIPT
# STYLE
// TEMPLATE as well
# Continue
continue 2;
# IMG
# If child has an alt attribute, then:
# Let value be the contents of the alt attribute
# Strip leading and trailing ASCII whitespace from value
# Else if child has a src attribute, then:
# Let value be the contents of the src attribute
# Strip leading and trailing ASCII whitespace from value
# Set value to the absolute URL created by resolving value following the containing document’s language’s rules
# Else continue
# Append and prepend a single U+0020 SPACE code point to value
# Append value to output
# BR
# Append a string containing a single U+000A LF code point to output
# P
# Let value be the result of running this algorithm on child
# Prepend a single U+000A LF code point to value
# Append value to output
# Any other value
# Let value be the result of running this algorithm on child
# Append value to output
case "img":
# IMG
if ($n->hasAttribute("alt")) {
# If child has an alt attribute, then:
# Let value be the contents of the alt attribute
# Strip leading and trailing ASCII whitespace from value
$value = trim($n->getAttribute("alt"));
} elseif ($n->hasAttribute("src")) {
# Else if child has a src attribute, then:
# Let value be the contents of the src attribute
# Strip leading and trailing ASCII whitespace from value
$value = trim($n->getAttribute("src"));
# Set value to the absolute URL created by resolving value following the containing document’s language’s rules
$value = $this->normalizeUrl($value);
} else {
# Else continue
continue 2;
}
# Append and prepend a single U+0020 SPACE code point to value
# Append value to output
$output[] = " ".$value." ";
break;
case "br":
# BR
# Append a string containing a single U+000A LF code point to output
$output[] = "\n";
break;
case "p":
# P
# Let value be the result of running this algorithm on child
# Prepend a single U+000A LF code point to value
# Append value to output
$output[] = "\n".$this->getCleanTextThorough($n, $prefix);
break;
default:
# Any other value
# Let value be the result of running this algorithm on child
# Append value to output
$output[] = $this->getCleanTextThorough($n, $prefix);
break;
}
# Else continue
} else {
# Else continue
continue;
}
# Return the concatenation of output
}
# Return the concatenation of output
return implode("", $output);
}
protected function getCleanTextBasic(\DOMElement $node, string $prefix): string {

29
tests/cases/StandardTest.php
View File

@ -22,7 +22,7 @@ class StandardTest extends \PHPUnit\Framework\TestCase {
];
/** @dataProvider provideStandardTests */
public function testStandardTests(string $name, string $path): void {
public function testStandardTests(string $name, string $path, $options): void {
if (isset(self::SUPPRESSED[$name])) {
$this->markTestIncomplete(self::SUPPRESSED[$name]);
}
@ -32,17 +32,23 @@ class StandardTest extends \PHPUnit\Framework\TestCase {
// fix up expectation where necessary
array_walk_recursive($exp, function(&$v) {
// URLs differ trivially from output of our normalization library
if (preg_match('#^https?://[^/]+$#', $v)) {
$v .= "/";
}
$v = preg_replace('#^https?://[^/]+$#', "$0/", $v);
});
// URLs also need fixing as keys in rel-urls
foreach ($exp['rel-urls'] as $k => $v) {
$fixed = preg_replace('#^https?://[^/]+$#', "$0/", $k);
$exp['rel-urls'][$fixed] = $v;
if ($fixed !== $k) {
unset($exp['rel-urls'][$k]);
}
}
// perform some further monkey-patching on specific tests
$exp = $this->fixTests($exp, $name);
// parse input
$dom = new DOMParser;
$parser = new Parser;
$doc = $dom->parseFromString($html, "text/html; charset=UTF-8");
$act = $parser->parseElement($doc->documentElement, "http://example.com");
$act = $parser->parseElement($doc->documentElement, "http://example.com", $options);
// sort both arrays
$this->ksort($exp);
$this->ksort($act);
@ -55,14 +61,13 @@ class StandardTest extends \PHPUnit\Framework\TestCase {
}
public function provideStandardTests(): \Generator {
return $this->provideTestList(\MensBeam\Microformats\BASE."vendor-bin/phpunit/vendor/mf2/tests/tests/");
// the standard tests
yield from $this->provideTestList([\MensBeam\Microformats\BASE."vendor-bin/phpunit/vendor/mf2/tests/tests/"], ['basicTrim' => true]);
// tests from php-mf2
yield from $this->provideTestList([\MensBeam\Microformats\BASE."tests/cases/json/"], null);
}
protected function provideTestList(): \Generator {
$tests = [
\MensBeam\Microformats\BASE."vendor-bin/phpunit/vendor/mf2/tests/tests/", // standard tests
\MensBeam\Microformats\BASE."tests/cases/json/", // additional tests
];
protected function provideTestList(array $tests, ?array $options = null): \Generator {
foreach ($tests as $base) {
$base = strtr($base, "\\", "/");
foreach (new \RegexIterator(new \RecursiveIteratorIterator(new \RecursiveDirectoryIterator($base )), '/\.json$/') as $file) {
@ -70,7 +75,7 @@ class StandardTest extends \PHPUnit\Framework\TestCase {
$path = preg_replace('/\.json$/', '', $path);
$name = strtr($path, "\\", "/");
$name = str_replace(strtr($base, "\\", "/"), "", $name);
yield $name => [$name, $path];
yield $name => [$name, $path, $options];
}
}
}

Write Preview
Loading…
Cancel
Save