mirror of
https://github.com/silverstripe/silverstripe-framework
synced 2024-10-06 08:08:33 +02:00
Merge pull request #10641 from creative-commoners/pulls/5/change-diff-lib
API Replace thirdparty difflib with maintained lib
This commit is contained in:
Sabina Talipova
GitHub
commit
6a6a7849c0
@ -33,6 +33,7 @@
|
||||
"monolog/monolog": "^3.2.0",
|
||||
"nikic/php-parser": "^4.15.0",
|
||||
"psr/container": "^1.1 || ^2.0",
|
||||
"sebastian/diff": "^4.0",
|
||||
"silverstripe/config": "^2",
|
||||
"silverstripe/assets": "^2",
|
||||
"silverstripe/vendor-plugin": "^2",
|
||||
|
||||
@ -1,203 +0,0 @@
|
||||
<?php
|
||||
|
||||
namespace SilverStripe\View\Parsers;
|
||||
|
||||
use InvalidArgumentException;
|
||||
use SilverStripe\Core\Convert;
|
||||
use SilverStripe\Core\Injector\Injector;
|
||||
|
||||
require_once 'difflib/difflib.php';
|
||||
|
||||
/**
|
||||
* Class representing a 'diff' between two sequences of strings.
|
||||
*/
|
||||
class Diff extends \Diff
|
||||
{
|
||||
public static $html_cleaner_class = null;
|
||||
|
||||
/**
|
||||
* Attempt to clean invalid HTML, which messes up diffs.
|
||||
* This cleans code if possible, using an instance of HTMLCleaner
|
||||
*
|
||||
* NB: By default, only extremely simple tidying is performed,
|
||||
* by passing through DomDocument::loadHTML and saveXML
|
||||
*
|
||||
* @param string $content HTML content
|
||||
* @param HTMLCleaner $cleaner Optional instance of a HTMLCleaner class to
|
||||
* use, overriding self::$html_cleaner_class
|
||||
* @return mixed|string
|
||||
*/
|
||||
public static function cleanHTML($content, $cleaner = null)
|
||||
{
|
||||
if (!$cleaner) {
|
||||
if (self::$html_cleaner_class && class_exists(self::$html_cleaner_class)) {
|
||||
$cleaner = Injector::inst()->create(self::$html_cleaner_class);
|
||||
} else {
|
||||
//load cleaner if the dependent class is available
|
||||
$cleaner = HTMLCleaner::inst();
|
||||
}
|
||||
}
|
||||
|
||||
if ($cleaner) {
|
||||
$content = $cleaner->cleanHTML($content);
|
||||
} else {
|
||||
// At most basic level of cleaning, use DOMDocument to save valid XML.
|
||||
$doc = HTMLValue::create($content);
|
||||
$content = $doc->getContent();
|
||||
}
|
||||
|
||||
// Remove empty <ins /> and <del /> tags because browsers hate them
|
||||
$content = preg_replace('/<(ins|del)[^>]*\/>/', '', $content ?? '');
|
||||
|
||||
return $content;
|
||||
}
|
||||
|
||||
/**
|
||||
* @param string $from
|
||||
* @param string $to
|
||||
* @param bool $escape
|
||||
* @return string
|
||||
*/
|
||||
public static function compareHTML($from, $to, $escape = false)
|
||||
{
|
||||
// First split up the content into words and tags
|
||||
$set1 = self::getHTMLChunks($from);
|
||||
$set2 = self::getHTMLChunks($to);
|
||||
|
||||
// Diff that
|
||||
$diff = new Diff($set1, $set2);
|
||||
|
||||
$tagStack[1] = $tagStack[2] = 0;
|
||||
$rechunked[1] = $rechunked[2] = [];
|
||||
|
||||
// Go through everything, converting edited tags (and their content) into single chunks. Otherwise
|
||||
// the generated HTML gets crusty
|
||||
foreach ($diff->edits as $edit) {
|
||||
$lookForTag = false;
|
||||
$stuffFor = [];
|
||||
switch ($edit->type) {
|
||||
case 'copy':
|
||||
$lookForTag = false;
|
||||
$stuffFor[1] = $edit->orig;
|
||||
$stuffFor[2] = $edit->orig;
|
||||
break;
|
||||
|
||||
case 'change':
|
||||
$lookForTag = true;
|
||||
$stuffFor[1] = $edit->orig;
|
||||
$stuffFor[2] = $edit->final;
|
||||
break;
|
||||
|
||||
case 'add':
|
||||
$lookForTag = true;
|
||||
$stuffFor[1] = null;
|
||||
$stuffFor[2] = $edit->final;
|
||||
break;
|
||||
|
||||
case 'delete':
|
||||
$lookForTag = true;
|
||||
$stuffFor[1] = $edit->orig;
|
||||
$stuffFor[2] = null;
|
||||
break;
|
||||
}
|
||||
|
||||
foreach ($stuffFor as $listName => $chunks) {
|
||||
if ($chunks) {
|
||||
foreach ($chunks as $item) {
|
||||
// $tagStack > 0 indicates that we should be tag-building
|
||||
if ($tagStack[$listName]) {
|
||||
$rechunked[$listName][sizeof($rechunked[$listName])-1] .= ' ' . $item;
|
||||
} else {
|
||||
$rechunked[$listName][] = $item;
|
||||
}
|
||||
|
||||
if ($lookForTag
|
||||
&& !$tagStack[$listName]
|
||||
&& isset($item[0])
|
||||
&& $item[0] == "<"
|
||||
&& substr($item ?? '', 0, 2) != "</"
|
||||
) {
|
||||
$tagStack[$listName] = 1;
|
||||
} elseif ($tagStack[$listName]) {
|
||||
if (substr($item ?? '', 0, 2) == "</") {
|
||||
$tagStack[$listName]--;
|
||||
} elseif (isset($item[0]) && $item[0] == "<") {
|
||||
$tagStack[$listName]++;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// Diff the re-chunked data, turning it into maked up HTML
|
||||
$diff = new Diff($rechunked[1], $rechunked[2]);
|
||||
$content = '';
|
||||
foreach ($diff->edits as $edit) {
|
||||
$orig = ($escape) ? Convert::raw2xml($edit->orig) : $edit->orig;
|
||||
$final = ($escape) ? Convert::raw2xml($edit->final) : $edit->final;
|
||||
|
||||
switch ($edit->type) {
|
||||
case 'copy':
|
||||
$content .= " " . implode(" ", $orig) . " ";
|
||||
break;
|
||||
|
||||
case 'change':
|
||||
$content .= " <ins>" . implode(" ", $final) . "</ins> ";
|
||||
$content .= " <del>" . implode(" ", $orig) . "</del> ";
|
||||
break;
|
||||
|
||||
case 'add':
|
||||
$content .= " <ins>" . implode(" ", $final) . "</ins> ";
|
||||
break;
|
||||
|
||||
case 'delete':
|
||||
$content .= " <del>" . implode(" ", $orig) . "</del> ";
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
return self::cleanHTML($content);
|
||||
}
|
||||
|
||||
/**
|
||||
* @param string|bool|array $content If passed as an array, values will be concatenated with a comma.
|
||||
* @return array
|
||||
*/
|
||||
public static function getHTMLChunks($content)
|
||||
{
|
||||
if ($content && !is_string($content) && !is_array($content) && !is_numeric($content) && !is_bool($content)) {
|
||||
throw new InvalidArgumentException('$content parameter needs to be a string or array');
|
||||
}
|
||||
if (is_bool($content)) {
|
||||
// Convert boolean to strings
|
||||
$content = $content ? "true" : "false";
|
||||
}
|
||||
if (is_array($content)) {
|
||||
$content = array_filter($content ?? [], 'is_scalar');
|
||||
// Convert array to CSV
|
||||
$content = implode(',', $content);
|
||||
}
|
||||
|
||||
$content = str_replace([" ", "<", ">"], [" "," <", "> "], $content ?? '');
|
||||
$candidateChunks = preg_split("/[\t\r\n ]+/", $content ?? '');
|
||||
$chunks = [];
|
||||
for ($i = 0; $i < count($candidateChunks ?? []); $i++) {
|
||||
$item = $candidateChunks[$i];
|
||||
if (isset($item[0]) && $item[0] == "<") {
|
||||
$newChunk = $item;
|
||||
while ($item[strlen($item)-1] != ">") {
|
||||
if (++$i >= count($candidateChunks ?? [])) {
|
||||
break;
|
||||
}
|
||||
$item = $candidateChunks[$i];
|
||||
$newChunk .= ' ' . $item;
|
||||
}
|
||||
$chunks[] = $newChunk;
|
||||
} else {
|
||||
$chunks[] = $item;
|
||||
}
|
||||
}
|
||||
return $chunks;
|
||||
}
|
||||
}
|
||||
292
src/View/Parsers/HtmlDiff.php
Normal file
292
src/View/Parsers/HtmlDiff.php
Normal file
@ -0,0 +1,292 @@
|
||||
<?php
|
||||
|
||||
namespace SilverStripe\View\Parsers;
|
||||
|
||||
use LogicException;
|
||||
use Masterminds\HTML5\Elements;
|
||||
use SebastianBergmann\Diff\Differ;
|
||||
use SilverStripe\Core\Convert;
|
||||
use SilverStripe\Core\Injector\Injector;
|
||||
|
||||
/**
|
||||
* Class representing a 'diff' between two sequences of HTML strings.
|
||||
*/
|
||||
class HtmlDiff
|
||||
{
|
||||
private const OLD_VAL = 'old';
|
||||
private const NEW_VAL = 'new';
|
||||
|
||||
private static ?Differ $differ = null;
|
||||
|
||||
public static ?string $html_cleaner_class = null;
|
||||
|
||||
/**
|
||||
* Get a diff between two sets of HTML content. The result is an HTML fragment which can be added directly
|
||||
* into the DOM. <ins> elements are used to indicate new content, and <del> elements are used to indicate
|
||||
* removed content.
|
||||
*
|
||||
* @param bool $escape If true, the HTML in $from and $to will be escaped after the diff operation is performed.
|
||||
*/
|
||||
public static function compareHtml(string|array $from, string|array $to, bool $escape = false): string
|
||||
{
|
||||
// Get HTML chunks even if we're going to escape it later
|
||||
// The diff algorithm sees "<span>some" as a single piece rather than "<span>" and "some" being separate
|
||||
$from = self::explodeToHtmlChunks($from);
|
||||
$to = self::explodeToHtmlChunks($to);
|
||||
|
||||
// Diff the chunks
|
||||
$differ = self::getDiffer();
|
||||
$diff = $differ->diffToArray($from, $to);
|
||||
|
||||
// If we aren't escaping the HTML, convert the first diff into clean HTML blocks and then run a new diff
|
||||
// on the blocks to get an end result that doesn't have broken HTML
|
||||
if (!$escape) {
|
||||
$diifAsBlocks = self::convertDiffToHtmlBlocks($diff);
|
||||
$diff = $differ->diffToArray($diifAsBlocks[self::OLD_VAL], $diifAsBlocks[self::NEW_VAL]);
|
||||
}
|
||||
|
||||
$diff = self::createFinalDiffBlocks($diff, $escape);
|
||||
|
||||
// Take the diff and slap the appropriate <ins> and <del> tags in place
|
||||
$content = '';
|
||||
foreach ($diff as $edit) {
|
||||
list($value, $type) = $edit;
|
||||
if (is_array($value)) {
|
||||
$value = implode(' ', $value);
|
||||
}
|
||||
if ($escape) {
|
||||
$value = Convert::raw2xml($value);
|
||||
}
|
||||
|
||||
switch ($type) {
|
||||
case Differ::OLD:
|
||||
$content .= ' ' . $value . ' ';
|
||||
break;
|
||||
|
||||
case Differ::ADDED:
|
||||
$content .= ' <ins>' . $value . '</ins> ';
|
||||
break;
|
||||
|
||||
case Differ::REMOVED:
|
||||
$content .= ' <del>' . $value . '</del> ';
|
||||
break;
|
||||
|
||||
default:
|
||||
throw new LogicException('Unexpected type encountered: "' . (string)$type . '"');
|
||||
}
|
||||
}
|
||||
|
||||
return self::cleanHTML($content);
|
||||
}
|
||||
|
||||
/**
|
||||
* Takes a final diff and pulls the distinct tokens into related blocks
|
||||
* i.e. we avoid having multiple separate additions/subtractions in a row
|
||||
*
|
||||
* Similar to SebastianBergmann\Diff\Output\AbstractChunkOutputBuilder::getCommonChunks but it's HTML aware.
|
||||
*/
|
||||
private static function createFinalDiffBlocks(array $diff, bool $escaped): array
|
||||
{
|
||||
$blocks = [];
|
||||
$building = null;
|
||||
$openTagsInBlock = 0;
|
||||
|
||||
foreach ($diff as $edit) {
|
||||
list($value, $type) = $edit;
|
||||
$isClosingTag = !$escaped && str_starts_with($value, '</');
|
||||
$isOpeningNonVoidTag = !$escaped && self::isOpeningNonVoidTag($value);
|
||||
|
||||
// If we were building a DIFFERENT type of block, or we've run out of open tags and are closing something
|
||||
// earlier in the chain, close the previous block and start a new one
|
||||
if ($building !== $type || ($isClosingTag && $openTagsInBlock <= 0)) {
|
||||
$building = $type;
|
||||
$openTagsInBlock = $isOpeningNonVoidTag ? 1 : 0;
|
||||
$blocks[] = [$value, $type];
|
||||
continue;
|
||||
}
|
||||
|
||||
// Mark opened or closed blocks
|
||||
if ($isOpeningNonVoidTag) {
|
||||
$openTagsInBlock++;
|
||||
}
|
||||
if ($isClosingTag) {
|
||||
$openTagsInBlock--;
|
||||
}
|
||||
|
||||
// Add this value to the current block
|
||||
$blocks[count($blocks) - 1][0] .= ' ' . $value;
|
||||
}
|
||||
|
||||
return $blocks;
|
||||
}
|
||||
|
||||
/**
|
||||
* Convert an intermediate diff into clean HTML blocks of changes.
|
||||
*
|
||||
* e.g. if making this change:
|
||||
* - <p>
|
||||
* <span>
|
||||
* - Some text
|
||||
* + Other text
|
||||
* </span>
|
||||
* - </p>
|
||||
*
|
||||
* We don't want to end up breaking up the HTML like this:
|
||||
* <del><p></del><span><del>Some</del><ins>Other</ins> text</span><del></p></del>
|
||||
* Instead we want to retain the valid HTML like this:
|
||||
* <del><p><span>Some text</span></p></del>
|
||||
* <ins><p>Other text</p></ins>
|
||||
*/
|
||||
private static function convertDiffToHtmlBlocks(array $diff): array
|
||||
{
|
||||
$openTagsInBlock[self::OLD_VAL] = $openTagsInBlock[self::NEW_VAL] = 0;
|
||||
$htmlBlocks[self::OLD_VAL] = $htmlBlocks[self::NEW_VAL] = [];
|
||||
|
||||
foreach ($diff as $edit) {
|
||||
list($value, $type) = $edit;
|
||||
switch ($type) {
|
||||
case Differ::OLD:
|
||||
if ($value === '') {
|
||||
break;
|
||||
}
|
||||
self::addToHtmlBlocks($htmlBlocks, $openTagsInBlock, self::OLD_VAL, false, $value);
|
||||
self::addToHtmlBlocks($htmlBlocks, $openTagsInBlock, self::NEW_VAL, false, $value);
|
||||
break;
|
||||
|
||||
case Differ::ADDED:
|
||||
self::addToHtmlBlocks($htmlBlocks, $openTagsInBlock, self::NEW_VAL, true, $value);
|
||||
break;
|
||||
|
||||
case Differ::REMOVED:
|
||||
self::addToHtmlBlocks($htmlBlocks, $openTagsInBlock, self::OLD_VAL, true, $value);
|
||||
break;
|
||||
}
|
||||
}
|
||||
return $htmlBlocks;
|
||||
}
|
||||
|
||||
/**
|
||||
* Add an intermediate diff value to the appropriate HTML block
|
||||
*/
|
||||
private static function addToHtmlBlocks(
|
||||
array &$htmlBlocks,
|
||||
array &$openTagsInBlock,
|
||||
string $oldOrNew,
|
||||
bool $lookForTag,
|
||||
string $value
|
||||
): void {
|
||||
$alreadyMidBlock = $openTagsInBlock[$oldOrNew] > 0;
|
||||
$canAddTagsToBlock = $lookForTag || $alreadyMidBlock;
|
||||
|
||||
if ($alreadyMidBlock) {
|
||||
// If we haven't closed all tags in the block, this value is part of the previous HTML block
|
||||
$htmlBlocks[$oldOrNew][count($htmlBlocks[$oldOrNew]) - 1] .= ' ' . $value;
|
||||
} else {
|
||||
// Otherwise it's part of a new block
|
||||
$htmlBlocks[$oldOrNew][] = $value;
|
||||
}
|
||||
|
||||
if ($canAddTagsToBlock && self::isOpeningNonVoidTag($value)) {
|
||||
// If we're mid block or explicitly looking for new tags, we should add any new non-void tags to the block
|
||||
$openTagsInBlock[$oldOrNew]++;
|
||||
} elseif ($alreadyMidBlock && str_starts_with($value, '</')) {
|
||||
// If we're mid block and closing a tag, that's one less tag to close before the block ends
|
||||
$openTagsInBlock[$oldOrNew]--;
|
||||
}
|
||||
}
|
||||
|
||||
private static function isOpeningNonVoidTag(string $value): bool
|
||||
{
|
||||
preg_match('/^<(\w*)[ >]/', $value, $matches);
|
||||
return isset($matches[1]) && Elements::isElement($matches[1]) && !Elements::isA($matches[1], Elements::VOID_TAG);
|
||||
}
|
||||
|
||||
/**
|
||||
* Takes a long HTML string (or array of strings) and breaks it up into chunks of HTML
|
||||
* e.g. '<div class="something">Some Text</div>' becomes ['<div class="something">', 'Some', 'Text', '</div>']
|
||||
*
|
||||
* @param string|array $content If passed as an array, values will be concatenated with a comma.
|
||||
*/
|
||||
private static function explodeToHtmlChunks(string|array $content): array
|
||||
{
|
||||
if (is_array($content)) {
|
||||
$content = array_filter($content, 'is_scalar');
|
||||
// Convert array to CSV
|
||||
$content = implode(',', $content);
|
||||
}
|
||||
|
||||
$content = str_replace([' ', '<', '>'], [' ',' <', '> '], $content);
|
||||
$candidateChunks = preg_split('/[\s]+/', $content);
|
||||
$chunks = [];
|
||||
$currentChunk = '';
|
||||
|
||||
foreach ($candidateChunks as $item) {
|
||||
if ($item === '') {
|
||||
continue;
|
||||
}
|
||||
// If we've started a chunk, keep going until we close the tag.
|
||||
if ($currentChunk !== '') {
|
||||
$currentChunk .= ' ' . $item;
|
||||
if (!str_ends_with($item, '>')) {
|
||||
continue;
|
||||
} else {
|
||||
$chunks[] = $currentChunk;
|
||||
$currentChunk = '';
|
||||
continue;
|
||||
}
|
||||
}
|
||||
|
||||
// If we open a tag, start a new chunk.
|
||||
if (str_starts_with($item, '<') && !str_ends_with($item, '>')) {
|
||||
$currentChunk = $item;
|
||||
continue;
|
||||
}
|
||||
|
||||
// If we're not starting or continuing a tag chunk, just add this as its own chunk.
|
||||
$chunks[] = $item;
|
||||
}
|
||||
|
||||
return $chunks;
|
||||
}
|
||||
|
||||
/**
|
||||
* Attempt to clean invalid HTML, which messes up diffs.
|
||||
* This cleans code if possible, using an instance of HTMLCleaner
|
||||
*
|
||||
* @param ?HTMLCleaner $cleaner Optional instance of a HTMLCleaner class to
|
||||
* use, overriding self::$html_cleaner_class
|
||||
*/
|
||||
private static function cleanHTML(string $content, ?HTMLCleaner $cleaner = null): string
|
||||
{
|
||||
if (!$cleaner) {
|
||||
if (self::$html_cleaner_class && class_exists(self::$html_cleaner_class)) {
|
||||
$cleaner = Injector::inst()->create(self::$html_cleaner_class);
|
||||
} else {
|
||||
//load cleaner if the dependent class is available
|
||||
$cleaner = HTMLCleaner::inst();
|
||||
}
|
||||
}
|
||||
|
||||
/** @var HTMLCleaner $cleaner */
|
||||
if ($cleaner) {
|
||||
$content = $cleaner->cleanHTML($content);
|
||||
} else {
|
||||
// At most basic level of cleaning, use DOMDocument to save valid XML.
|
||||
$doc = HTMLValue::create($content);
|
||||
$content = $doc->getContent();
|
||||
}
|
||||
|
||||
// Remove empty <ins /> and <del /> tags because browsers hate them
|
||||
$content = preg_replace('/<(ins|del)[^>]*\/>/', '', $content);
|
||||
|
||||
return $content;
|
||||
}
|
||||
|
||||
private static function getDiffer(): Differ
|
||||
{
|
||||
if (!self::$differ) {
|
||||
self::$differ = new Differ();
|
||||
}
|
||||
return self::$differ;
|
||||
}
|
||||
}
|
||||
@ -1,90 +0,0 @@
|
||||
<?php
|
||||
|
||||
namespace SilverStripe\View\Tests\Parsers;
|
||||
|
||||
use SilverStripe\Dev\SapphireTest;
|
||||
use SilverStripe\View\Parsers\Diff;
|
||||
|
||||
class DiffTest extends SapphireTest
|
||||
{
|
||||
|
||||
/**
|
||||
* @see https://groups.google.com/forum/#!topic/silverstripe-dev/yHcluCvuszo
|
||||
*/
|
||||
public function testTableDiff()
|
||||
{
|
||||
if (!class_exists('DOMDocument')) {
|
||||
$this->markTestSkipped('"DOMDocument" required');
|
||||
return;
|
||||
}
|
||||
|
||||
$from = "<table>
|
||||
<tbody>
|
||||
<tr class=\"blah\">
|
||||
<td colspan=\"2\">Row 1</td>
|
||||
</tr>
|
||||
<tr class=\"foo\">
|
||||
<td>Row 2</td>
|
||||
<td>Row 2</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>Row 3</td>
|
||||
<td>Row 3</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>";
|
||||
|
||||
$to = "<table class=\"new-class\">
|
||||
<tbody>
|
||||
<tr class=\"blah\">
|
||||
<td colspan=\"2\">Row 1</td>
|
||||
</tr>
|
||||
<tr class=\"foo\">
|
||||
<td>Row 2</td>
|
||||
<td>Row 2</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>";
|
||||
|
||||
$expected = "<ins>" . $to . "</ins>" . "<del>" . $from . "</del>";
|
||||
$compare = Diff::compareHTML($from, $to);
|
||||
|
||||
// Very hard to debug this way, wouldn't need to do this if PHP had an *actual* DOM parsing lib,
|
||||
// and not just the poor excuse that is DOMDocument
|
||||
$compare = preg_replace('/[\s\t\n\r]*/', '', $compare ?? '');
|
||||
$expected = preg_replace('/[\s\t\n\r]*/', '', $expected ?? '');
|
||||
|
||||
$this->assertEquals($expected, $compare);
|
||||
}
|
||||
|
||||
/**
|
||||
* @see https://github.com/silverstripe/silverstripe-framework/issues/8053
|
||||
*/
|
||||
public function testLegacyEachStatement()
|
||||
{
|
||||
$sentenceOne =
|
||||
'Lorem ipsum dolor sit amet, consectetur adipiscing elit.';
|
||||
$sentenceTwo =
|
||||
'Nulla porttitor, ex quis commodo pharetra, diam dui efficitur justo, eu gravida elit eros vel libero.';
|
||||
|
||||
$from = "$sentenceOne $sentenceTwo";
|
||||
$to = "$sentenceTwo $sentenceOne";
|
||||
|
||||
// We're cheating our test a little bit here, because depending on what HTML cleaner you have, you'll get
|
||||
// spaces added or not added around the tags.
|
||||
$expected = "/^ *<del>$sentenceOne<\/del> *$sentenceTwo *<ins>$sentenceOne<\/ins> *$/";
|
||||
$actual = Diff::compareHTML($from, $to);
|
||||
|
||||
$this->assertMatchesRegularExpression($expected, $actual);
|
||||
}
|
||||
|
||||
public function testDiffArray()
|
||||
{
|
||||
$from = ['Lorem', ['array here please ignore'], 'ipsum dolor'];
|
||||
$to = 'Lorem,ipsum';
|
||||
$expected = "/^Lorem,ipsum *<del>dolor<\/del> *$/";
|
||||
$actual = Diff::compareHTML($from, $to);
|
||||
|
||||
$this->assertMatchesRegularExpression($expected, $actual);
|
||||
}
|
||||
}
|
||||
241
tests/php/View/Parsers/HtmlDiffTest.php
Normal file
241
tests/php/View/Parsers/HtmlDiffTest.php
Normal file
@ -0,0 +1,241 @@
|
||||
<?php
|
||||
|
||||
namespace SilverStripe\View\Tests\Parsers;
|
||||
|
||||
use SebastianBergmann\Diff\Differ;
|
||||
use SilverStripe\Dev\SapphireTest;
|
||||
use SilverStripe\View\Parsers\HtmlDiff;
|
||||
|
||||
class HtmlDiffTest extends SapphireTest
|
||||
{
|
||||
|
||||
/**
|
||||
* Most if not all other tests strip out the whitespace from comparisons to avoid complexities with checking
|
||||
* if the actual HTML content is correct, since whitespace in HTML isn't all that vital and the algorithm
|
||||
* can add some extra spaces where they're not stricly necessary but don't affect anything.
|
||||
*
|
||||
* This test is here to ensure that spaces _are_ kept where they're actually needed (i.e. between text)
|
||||
*/
|
||||
public function testKeepsSpacesBetweenText()
|
||||
{
|
||||
$from = '<span>Some text</span> <span>more text</span>';
|
||||
$to = '<span>Other text</span> <span>more text</span>';
|
||||
$diff = HtmlDiff::compareHtml($from, $to);
|
||||
$this->assertEquals('<span><del>Some</del> <ins>Other</ins> text</span> <span>more text</span>', $diff, false);
|
||||
|
||||
// Note that the end result here isn't perfect (there are new spaces where there weren't before)...
|
||||
// If we make improvements later on that keep only the original spaces, that would be preferred.
|
||||
// This test is more here to protect against any unexpected changes to the spacing, so that we can make an intentional
|
||||
// decision as to whether those changes are desirable.
|
||||
$diff = HtmlDiff::compareHtml($from, $to, true);
|
||||
$this->assertEquals('<span> <del>Some</del> <ins>Other</ins> text </span> <span> more text </span>', $diff, true);
|
||||
}
|
||||
|
||||
/**
|
||||
* The underlying SebastianBergmann\Diff\Differ class has a special constant for end of line differences
|
||||
* but we shouldn't ever encounter that because of the way we're modifying the values before passing them
|
||||
* in to that class.
|
||||
*/
|
||||
public function testEndOfLineDoesntNeedHandling()
|
||||
{
|
||||
$from = 'some change' . "\r";
|
||||
$to = 'some change' . "\n";
|
||||
|
||||
// If we encounter an end-of-line difference, these should throw exceptions
|
||||
$this->assertSame('some change', HtmlDiff::compareHtml($from, $to, true));
|
||||
$this->assertSame('some change', HtmlDiff::compareHtml($from, $to));
|
||||
$this->assertSame('some change', HtmlDiff::compareHtml($to, $from, true));
|
||||
$this->assertSame('some change', HtmlDiff::compareHtml($to, $from));
|
||||
|
||||
// Ensure that this test is valid and that those changes would include an end-of-line warning
|
||||
// in a direct call to the underlying differ
|
||||
$differ = new Differ();
|
||||
$expected = [
|
||||
[
|
||||
'#Warning: Strings contain different line endings!' . "\n",
|
||||
Differ::DIFF_LINE_END_WARNING,
|
||||
],
|
||||
[
|
||||
$from,
|
||||
Differ::REMOVED,
|
||||
],
|
||||
[
|
||||
$to,
|
||||
Differ::ADDED,
|
||||
],
|
||||
];
|
||||
$this->assertSame($expected, $differ->diffToArray($from, $to));
|
||||
}
|
||||
|
||||
public function provideCompareHtml(): array
|
||||
{
|
||||
return [
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<p><span>Other text</span></p>',
|
||||
'escape' => false,
|
||||
'expected' => '<p><span><del>Some</del><ins>Other</ins> text</span></p>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<span>Other text</span>',
|
||||
'escape' => false,
|
||||
'expected' => '<del><p><span>Some text</span></p></del><ins><span>Other text</span></ins>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<p>Other text</p>',
|
||||
'escape' => false,
|
||||
'expected' => '<p><del><span>Some text</span></del><ins>Other text</ins></p>',
|
||||
],
|
||||
[
|
||||
'from' => '<h2 class="mb-3 h4">About</h2>',
|
||||
'to' => '<h2 class="mb-3 h2">About</h2>',
|
||||
'escape' => false,
|
||||
'expected' => '<del><h2 class="mb-3 h4">About</h2></del><ins><h2 class="mb-3 h2">About</h2></ins>',
|
||||
],
|
||||
[
|
||||
'from' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><p class="f4 my-3">A comprehensive Library</p></div>',
|
||||
'to' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><span class="etc"><p class="f4 my-3">A comprehensive</p></span></div>',
|
||||
'escape' => false,
|
||||
'expected' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><del><p class="f4 my-3">A comprehensive Library</p></del><ins><p class="f4 my-3"><span class="etc">A comprehensive</span></p></ins></div>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span><span>more stuff</span></p>',
|
||||
'to' => '<p><span>Some text</span></p><p>more stuff</p>',
|
||||
'escape' => false,
|
||||
'expected' => '<p><span>Sometext</span><del><span>morestuff</span></del></p><ins><p>morestuff</p></ins>',
|
||||
],
|
||||
// Same examples as above, but with escaped HTML
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<p><span>Other text</span></p>',
|
||||
'escape' => true,
|
||||
'expected' => '<p><span><del>Some</del><ins>Other</ins> text</span></p>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<span>Other text</span>',
|
||||
'escape' => true,
|
||||
'expected' => '<del><p></del><span><del>Some</del><ins>Other</ins> text</span><del></p></del>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span></p>',
|
||||
'to' => '<p>Other text</p>',
|
||||
'escape' => true,
|
||||
'expected' => '<p><del><span>Some</del><ins>Other</ins> text<del></span></del></p>',
|
||||
],
|
||||
[
|
||||
'from' => '<h2 class="mb-3 h4">About</h2>',
|
||||
'to' => '<h2 class="mb-3 h2">About</h2>',
|
||||
'escape' => true,
|
||||
// Note: This sees the whole h2 tag as being changed because of the initial call to explodeToHtmlChunks.
|
||||
// There is room to improve this in the future, but care would have to be taken not to aversely affect other scenarios.
|
||||
'expected' => '<del><h2 class="mb-3 h4"></del><ins><h2 class="mb-3 h2"></ins>About</h2>',
|
||||
],
|
||||
[
|
||||
'from' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><p class="f4 my-3">A comprehensive Library</p></div>',
|
||||
'to' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><span class="etc"><p class="f4 my-3">A comprehensive</p></span></div>',
|
||||
'escape' => true,
|
||||
'expected' => '<div class="BorderGrid-cell"><h2 class="mb-3 h4">About</h2><ins><span class="etc"></ins><p class="f4 my-3">A comprehensive<del> Library</del></p><ins></span></ins></div>',
|
||||
],
|
||||
[
|
||||
'from' => '<p><span>Some text</span><span>more stuff</span></p>',
|
||||
'to' => '<p><span>Some text</span></p><p>more stuff</p>',
|
||||
'escape' => true,
|
||||
'expected' => '<p><span>Some text</span><del><span></del><ins></p><p></ins>more stuff<del></span></del></p>',
|
||||
],
|
||||
];
|
||||
}
|
||||
|
||||
/**
|
||||
* @dataProvider provideCompareHtml
|
||||
*/
|
||||
public function testCompareHTML(string|array $from, string|array $to, bool $escape, string $expected)
|
||||
{
|
||||
$diff = HtmlDiff::compareHtml($from, $to, $escape);
|
||||
$this->assertEquals($this->removeWhiteSpace($expected), $this->removeWhiteSpace($diff));
|
||||
}
|
||||
|
||||
/**
|
||||
* @see https://groups.google.com/forum/#!topic/silverstripe-dev/yHcluCvuszo
|
||||
*/
|
||||
public function testTableDiff()
|
||||
{
|
||||
if (!class_exists('DOMDocument')) {
|
||||
$this->markTestSkipped('"DOMDocument" required');
|
||||
return;
|
||||
}
|
||||
|
||||
$from = '<table>
|
||||
<tbody>
|
||||
<tr class="blah">
|
||||
<td colspan="2">Row 1</td>
|
||||
</tr>
|
||||
<tr class="foo">
|
||||
<td>Row 2</td>
|
||||
<td>Row 2</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>Row 3</td>
|
||||
<td>Row 3</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>';
|
||||
|
||||
$to = '<table class="new-class">
|
||||
<tbody>
|
||||
<tr class="blah">
|
||||
<td colspan="2">Row 1</td>
|
||||
</tr>
|
||||
<tr class="foo">
|
||||
<td>Row 2</td>
|
||||
<td>Row 2</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>';
|
||||
|
||||
$expected = '<del>' . $from . '</del>' . '<ins>' . $to . '</ins>';
|
||||
$compare = HtmlDiff::compareHtml($from, $to);
|
||||
|
||||
$this->assertEquals($this->removeWhiteSpace($expected), $this->removeWhiteSpace($compare));
|
||||
}
|
||||
|
||||
/**
|
||||
* @see https://github.com/silverstripe/silverstripe-framework/issues/8053
|
||||
*/
|
||||
public function testLegacyEachStatement()
|
||||
{
|
||||
$sentenceOne =
|
||||
'Lorem ipsum dolor sit amet, consectetur adipiscing elit.';
|
||||
$sentenceTwo =
|
||||
'Nulla porttitor, ex quis commodo pharetra, diam dui efficitur justo, eu gravida elit eros vel libero.';
|
||||
|
||||
$from = "$sentenceOne $sentenceTwo";
|
||||
$to = "$sentenceTwo $sentenceOne";
|
||||
|
||||
// We're cheating our test a little bit here, because depending on what HTML cleaner you have, you'll get
|
||||
// spaces added or not added around the tags.
|
||||
$quotedOne = preg_quote($sentenceOne, '/');
|
||||
$quotedTwo = preg_quote($sentenceTwo, '/');
|
||||
$expected = '/^ *<del>' . $quotedOne . '<\/del> *' . $quotedTwo . ' *<ins>' . $quotedOne . '<\/ins> *$/';
|
||||
$actual = HtmlDiff::compareHtml($from, $to);
|
||||
|
||||
$this->assertMatchesRegularExpression($expected, $actual);
|
||||
}
|
||||
|
||||
public function testDiffArray()
|
||||
{
|
||||
$from = ['Lorem', ['array here please ignore'], 'ipsum dolor'];
|
||||
$to = 'Lorem,ipsum';
|
||||
$expected = '/^Lorem,ipsum *<del>dolor<\/del> *$/';
|
||||
$actual = HtmlDiff::compareHtml($from, $to);
|
||||
|
||||
$this->assertMatchesRegularExpression($expected, $actual);
|
||||
}
|
||||
|
||||
private function removeWhiteSpace(string $value): string
|
||||
{
|
||||
return preg_replace('/[\s]*/', '', $value);
|
||||
}
|
||||
}
|
||||
911
thirdparty/difflib/difflib.php
vendored
911
thirdparty/difflib/difflib.php
vendored
@ -1,911 +0,0 @@
|
||||
<?php
|
||||
/*
|
||||
* Copyright (C) 2000, 2001 Geoffrey T. Dairiki <dairiki@dairiki.org>
|
||||
*
|
||||
* This file is part of PhpWiki.
|
||||
*
|
||||
* PhpWiki is free software; you can redistribute it and/or modify
|
||||
* it under the terms of the GNU General Public License as published by
|
||||
* the Free Software Foundation; either version 2 of the License, or
|
||||
* (at your option) any later version.
|
||||
*
|
||||
* PhpWiki is distributed in the hope that it will be useful,
|
||||
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
||||
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
||||
* GNU General Public License for more details.
|
||||
*
|
||||
* You should have received a copy of the GNU General Public License along
|
||||
* with PhpWiki; if not, write to the Free Software Foundation, Inc.,
|
||||
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
|
||||
*/
|
||||
|
||||
// difflib.php
|
||||
//
|
||||
// A PHP diff engine for phpwiki.
|
||||
|
||||
abstract class DiffOp
|
||||
{
|
||||
public $type;
|
||||
public $orig;
|
||||
public $final;
|
||||
|
||||
abstract public function reverse();
|
||||
|
||||
public function norig()
|
||||
{
|
||||
return $this->orig ? sizeof($this->orig) : 0;
|
||||
}
|
||||
|
||||
public function nfinal()
|
||||
{
|
||||
return $this->final ? sizeof($this->final) : 0;
|
||||
}
|
||||
}
|
||||
|
||||
class DiffOp_Copy extends DiffOp
|
||||
{
|
||||
public $type = 'copy';
|
||||
|
||||
function __construct($orig, $final = false)
|
||||
{
|
||||
if (!is_array($final))
|
||||
$final = $orig;
|
||||
$this->orig = $orig;
|
||||
$this->final = $final;
|
||||
}
|
||||
|
||||
public function reverse()
|
||||
{
|
||||
return new DiffOp_Copy($this->final, $this->orig);
|
||||
}
|
||||
}
|
||||
|
||||
class DiffOp_Delete extends DiffOp
|
||||
{
|
||||
public $type = 'delete';
|
||||
|
||||
function __construct($lines)
|
||||
{
|
||||
$this->orig = $lines;
|
||||
$this->final = false;
|
||||
}
|
||||
|
||||
public function reverse()
|
||||
{
|
||||
return new DiffOp_Add($this->orig);
|
||||
}
|
||||
}
|
||||
|
||||
class DiffOp_Add extends DiffOp
|
||||
{
|
||||
public $type = 'add';
|
||||
|
||||
function __construct($lines)
|
||||
{
|
||||
$this->final = $lines;
|
||||
$this->orig = false;
|
||||
}
|
||||
|
||||
public function reverse()
|
||||
{
|
||||
return new DiffOp_Delete($this->final);
|
||||
}
|
||||
}
|
||||
|
||||
class DiffOp_Change extends DiffOp
|
||||
{
|
||||
public $type = 'change';
|
||||
|
||||
function __construct($orig, $final)
|
||||
{
|
||||
$this->orig = $orig;
|
||||
$this->final = $final;
|
||||
}
|
||||
|
||||
public function reverse()
|
||||
{
|
||||
return new DiffOp_Change($this->final, $this->orig);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Class used internally by Diff to actually compute the diffs.
|
||||
*
|
||||
* The algorithm used here is mostly lifted from the perl module
|
||||
* Algorithm::Diff (version 1.06) by Ned Konz, which is available at:
|
||||
* http://www.perl.com/CPAN/authors/id/N/NE/NEDKONZ/Algorithm-Diff-1.06.zip
|
||||
*
|
||||
* More ideas are taken from:
|
||||
* http://www.ics.uci.edu/~eppstein/161/960229.html
|
||||
*
|
||||
* Some ideas are (and a bit of code) are from from analyze.c, from GNU
|
||||
* diffutils-2.7, which can be found at:
|
||||
* ftp://gnudist.gnu.org/pub/gnu/diffutils/diffutils-2.7.tar.gz
|
||||
*
|
||||
* Finally, some ideas (subdivision by NCHUNKS > 2, and some optimizations)
|
||||
* are my own.
|
||||
*
|
||||
* @author Geoffrey T. Dairiki
|
||||
* @access private
|
||||
*/
|
||||
class DiffEngine
|
||||
{
|
||||
public $xchanged;
|
||||
public $ychanged;
|
||||
public $xv;
|
||||
public $yv;
|
||||
public $xind;
|
||||
public $yind;
|
||||
public $lcs;
|
||||
public $seq;
|
||||
public $in_seq;
|
||||
|
||||
public function diff($from_lines, $to_lines)
|
||||
{
|
||||
$n_from = sizeof($from_lines ?? []);
|
||||
$n_to = sizeof($to_lines ?? []);
|
||||
|
||||
$this->xchanged = $this->ychanged = array();
|
||||
$this->xv = $this->yv = array();
|
||||
$this->xind = $this->yind = array();
|
||||
unset($this->seq);
|
||||
unset($this->in_seq);
|
||||
unset($this->lcs);
|
||||
|
||||
// Skip leading common lines.
|
||||
for ($skip = 0; $skip < $n_from && $skip < $n_to; $skip++) {
|
||||
if ($from_lines[$skip] != $to_lines[$skip])
|
||||
break;
|
||||
$this->xchanged[$skip] = $this->ychanged[$skip] = false;
|
||||
}
|
||||
// Skip trailing common lines.
|
||||
$xi = $n_from;
|
||||
$yi = $n_to;
|
||||
for ($endskip = 0; --$xi > $skip && --$yi > $skip; $endskip++) {
|
||||
if ($from_lines[$xi] != $to_lines[$yi])
|
||||
break;
|
||||
$this->xchanged[$xi] = $this->ychanged[$yi] = false;
|
||||
}
|
||||
|
||||
// Ignore lines which do not exist in both files.
|
||||
for ($xi = $skip; $xi < $n_from - $endskip; $xi++)
|
||||
$xhash[$from_lines[$xi]] = 1;
|
||||
for ($yi = $skip; $yi < $n_to - $endskip; $yi++) {
|
||||
$line = $to_lines[$yi];
|
||||
if (($this->ychanged[$yi] = empty($xhash[$line])))
|
||||
continue;
|
||||
$yhash[$line] = 1;
|
||||
$this->yv[] = $line;
|
||||
$this->yind[] = $yi;
|
||||
}
|
||||
for ($xi = $skip; $xi < $n_from - $endskip; $xi++) {
|
||||
$line = $from_lines[$xi];
|
||||
if (($this->xchanged[$xi] = empty($yhash[$line])))
|
||||
continue;
|
||||
$this->xv[] = $line;
|
||||
$this->xind[] = $xi;
|
||||
}
|
||||
|
||||
// Find the LCS.
|
||||
$this->compareseq(0, sizeof($this->xv ?? []), 0, sizeof($this->yv ?? []));
|
||||
|
||||
// Merge edits when possible
|
||||
$this->shift_boundaries($from_lines, $this->xchanged, $this->ychanged);
|
||||
$this->shift_boundaries($to_lines, $this->ychanged, $this->xchanged);
|
||||
|
||||
// Compute the edit operations.
|
||||
$edits = array();
|
||||
$xi = $yi = 0;
|
||||
while ($xi < $n_from || $yi < $n_to) {
|
||||
assert($yi < $n_to || $this->xchanged[$xi]);
|
||||
assert($xi < $n_from || $this->ychanged[$yi]);
|
||||
|
||||
// Skip matching "snake".
|
||||
$copy = array();
|
||||
while ($xi < $n_from && $yi < $n_to
|
||||
&& !$this->xchanged[$xi] && !$this->ychanged[$yi]) {
|
||||
$copy[] = $from_lines[$xi++];
|
||||
++$yi;
|
||||
}
|
||||
if ($copy)
|
||||
$edits[] = new DiffOp_Copy($copy);
|
||||
|
||||
// Find deletes & adds.
|
||||
$delete = array();
|
||||
while ($xi < $n_from && $this->xchanged[$xi])
|
||||
$delete[] = $from_lines[$xi++];
|
||||
|
||||
$add = array();
|
||||
while ($yi < $n_to && $this->ychanged[$yi])
|
||||
$add[] = $to_lines[$yi++];
|
||||
|
||||
if ($delete && $add)
|
||||
$edits[] = new DiffOp_Change($delete, $add);
|
||||
elseif ($delete)
|
||||
$edits[] = new DiffOp_Delete($delete);
|
||||
elseif ($add)
|
||||
$edits[] = new DiffOp_Add($add);
|
||||
}
|
||||
return $edits;
|
||||
}
|
||||
|
||||
/* Divide the Largest Common Subsequence (LCS) of the sequences
|
||||
* [XOFF, XLIM) and [YOFF, YLIM) into NCHUNKS approximately equally
|
||||
* sized segments.
|
||||
*
|
||||
* Returns (LCS, PTS). LCS is the length of the LCS. PTS is an
|
||||
* array of NCHUNKS+1 (X, Y) indexes giving the diving points between
|
||||
* sub sequences. The first sub-sequence is contained in [X0, X1),
|
||||
* [Y0, Y1), the second in [X1, X2), [Y1, Y2) and so on. Note
|
||||
* that (X0, Y0) == (XOFF, YOFF) and
|
||||
* (X[NCHUNKS], Y[NCHUNKS]) == (XLIM, YLIM).
|
||||
*
|
||||
* This function assumes that the first lines of the specified portions
|
||||
* of the two files do not match, and likewise that the last lines do not
|
||||
* match. The caller must trim matching lines from the beginning and end
|
||||
* of the portions it is going to specify.
|
||||
*/
|
||||
private function diag($xoff, $xlim, $yoff, $ylim, $nchunks)
|
||||
{
|
||||
$flip = false;
|
||||
|
||||
if ($xlim - $xoff > $ylim - $yoff) {
|
||||
// Things seems faster (I'm not sure I understand why)
|
||||
// when the shortest sequence in X.
|
||||
$flip = true;
|
||||
list ($xoff, $xlim, $yoff, $ylim)
|
||||
= array($yoff, $ylim, $xoff, $xlim);
|
||||
}
|
||||
|
||||
if ($flip)
|
||||
for ($i = $ylim - 1; $i >= $yoff; $i--)
|
||||
$ymatches[$this->xv[$i]][] = $i;
|
||||
else
|
||||
for ($i = $ylim - 1; $i >= $yoff; $i--)
|
||||
$ymatches[$this->yv[$i]][] = $i;
|
||||
|
||||
$this->lcs = 0;
|
||||
$this->seq[0] = $yoff - 1;
|
||||
$this->in_seq = array();
|
||||
$ymids[0] = array();
|
||||
|
||||
$numer = $xlim - $xoff + $nchunks - 1;
|
||||
$x = $xoff;
|
||||
for ($chunk = 0; $chunk < $nchunks; $chunk++) {
|
||||
if ($chunk > 0)
|
||||
for ($i = 0; $i <= $this->lcs; $i++)
|
||||
$ymids[$i][$chunk - 1] = $this->seq[$i];
|
||||
|
||||
$x1 = $xoff + (int)(($numer + ($xlim - $xoff) * $chunk) / $nchunks);
|
||||
for (; $x < $x1; $x++) {
|
||||
$line = $flip ? $this->yv[$x] : $this->xv[$x];
|
||||
if (empty($ymatches[$line]))
|
||||
continue;
|
||||
$matches = $ymatches[$line];
|
||||
reset($matches);
|
||||
$pointer = 0;
|
||||
foreach ($matches as $y) {
|
||||
$pointer++;
|
||||
if (empty($this->in_seq[$y])) {
|
||||
$k = $this->lcs_pos($y);
|
||||
assert($k > 0);
|
||||
$ymids[$k] = $ymids[$k - 1];
|
||||
break;
|
||||
}
|
||||
}
|
||||
foreach (array_slice($matches ?? [], $pointer ?? 0) as $y) {
|
||||
if ($y > $this->seq[$k - 1]) {
|
||||
assert($y < $this->seq[$k]);
|
||||
// Optimization: this is a common case:
|
||||
// next match is just replacing previous match.
|
||||
$this->in_seq[$this->seq[$k]] = false;
|
||||
$this->seq[$k] = $y;
|
||||
$this->in_seq[$y] = 1;
|
||||
} elseif (empty($this->in_seq[$y])) {
|
||||
$k = $this->lcs_pos($y);
|
||||
assert($k > 0);
|
||||
$ymids[$k] = $ymids[$k - 1];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
$seps[] = $flip ? array($yoff, $xoff) : array($xoff, $yoff);
|
||||
$ymid = $ymids[$this->lcs];
|
||||
for ($n = 0; $n < $nchunks - 1; $n++) {
|
||||
$x1 = $xoff + (int)(($numer + ($xlim - $xoff) * $n) / $nchunks);
|
||||
$y1 = $ymid[$n] + 1;
|
||||
$seps[] = $flip ? array($y1, $x1) : array($x1, $y1);
|
||||
}
|
||||
$seps[] = $flip ? array($ylim, $xlim) : array($xlim, $ylim);
|
||||
|
||||
return array($this->lcs, $seps);
|
||||
}
|
||||
|
||||
private function lcs_pos($ypos)
|
||||
{
|
||||
$end = $this->lcs;
|
||||
if ($end == 0 || $ypos > $this->seq[$end]) {
|
||||
$this->seq[++$this->lcs] = $ypos;
|
||||
$this->in_seq[$ypos] = 1;
|
||||
return $this->lcs;
|
||||
}
|
||||
|
||||
$beg = 1;
|
||||
while ($beg < $end) {
|
||||
$mid = (int)(($beg + $end) / 2);
|
||||
if ($ypos > $this->seq[$mid])
|
||||
$beg = $mid + 1;
|
||||
else
|
||||
$end = $mid;
|
||||
}
|
||||
|
||||
assert($ypos != $this->seq[$end]);
|
||||
|
||||
$this->in_seq[$this->seq[$end]] = false;
|
||||
$this->seq[$end] = $ypos;
|
||||
$this->in_seq[$ypos] = 1;
|
||||
return $end;
|
||||
}
|
||||
|
||||
/* Find LCS of two sequences.
|
||||
*
|
||||
* The results are recorded in the vectors $this->{x,y}changed[], by
|
||||
* storing a 1 in the element for each line that is an insertion
|
||||
* or deletion (ie. is not in the LCS).
|
||||
*
|
||||
* The subsequence of file 0 is [XOFF, XLIM) and likewise for file 1.
|
||||
*
|
||||
* Note that XLIM, YLIM are exclusive bounds.
|
||||
* All line numbers are origin-0 and discarded lines are not counted.
|
||||
*/
|
||||
private function compareseq($xoff, $xlim, $yoff, $ylim)
|
||||
{
|
||||
// Slide down the bottom initial diagonal.
|
||||
while ($xoff < $xlim && $yoff < $ylim
|
||||
&& $this->xv[$xoff] == $this->yv[$yoff]) {
|
||||
++$xoff;
|
||||
++$yoff;
|
||||
}
|
||||
|
||||
// Slide up the top initial diagonal.
|
||||
while ($xlim > $xoff && $ylim > $yoff
|
||||
&& $this->xv[$xlim - 1] == $this->yv[$ylim - 1]) {
|
||||
--$xlim;
|
||||
--$ylim;
|
||||
}
|
||||
|
||||
if ($xoff == $xlim || $yoff == $ylim)
|
||||
$lcs = 0;
|
||||
else {
|
||||
// This is ad hoc but seems to work well.
|
||||
//$nchunks = sqrt(min($xlim - $xoff, $ylim - $yoff) / 2.5);
|
||||
//$nchunks = max(2,min(8,(int)$nchunks));
|
||||
$nchunks = min(7, $xlim - $xoff, $ylim - $yoff) + 1;
|
||||
list ($lcs, $seps)
|
||||
= $this->diag($xoff, $xlim, $yoff, $ylim, $nchunks);
|
||||
}
|
||||
|
||||
if ($lcs == 0) {
|
||||
// X and Y sequences have no common subsequence:
|
||||
// mark all changed.
|
||||
while ($yoff < $ylim)
|
||||
$this->ychanged[$this->yind[$yoff++]] = 1;
|
||||
while ($xoff < $xlim)
|
||||
$this->xchanged[$this->xind[$xoff++]] = 1;
|
||||
} else {
|
||||
// Use the partitions to split this problem into subproblems.
|
||||
reset($seps);
|
||||
$pt1 = $seps[0];
|
||||
while ($pt2 = next($seps)) {
|
||||
$this->compareseq($pt1[0], $pt2[0], $pt1[1], $pt2[1]);
|
||||
$pt1 = $pt2;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/* Adjust inserts/deletes of identical lines to join changes
|
||||
* as much as possible.
|
||||
*
|
||||
* We do something when a run of changed lines include a
|
||||
* line at one end and has an excluded, identical line at the other.
|
||||
* We are free to choose which identical line is included.
|
||||
* `compareseq' usually chooses the one at the beginning,
|
||||
* but usually it is cleaner to consider the following identical line
|
||||
* to be the "change".
|
||||
*
|
||||
* This is extracted verbatim from analyze.c (GNU diffutils-2.7).
|
||||
*/
|
||||
private function shift_boundaries($lines, &$changed, $other_changed)
|
||||
{
|
||||
$i = 0;
|
||||
$j = 0;
|
||||
|
||||
assert(sizeof($lines ?? []) == sizeof($changed ?? []));
|
||||
$len = sizeof($lines ?? []);
|
||||
$other_len = sizeof($other_changed ?? []);
|
||||
|
||||
while (1) {
|
||||
/*
|
||||
* Scan forwards to find beginning of another run of changes.
|
||||
* Also keep track of the corresponding point in the other file.
|
||||
*
|
||||
* Throughout this code, $i and $j are adjusted together so that
|
||||
* the first $i elements of $changed and the first $j elements
|
||||
* of $other_changed both contain the same number of zeros
|
||||
* (unchanged lines).
|
||||
* Furthermore, $j is always kept so that $j == $other_len or
|
||||
* $other_changed[$j] == false.
|
||||
*/
|
||||
while ($j < $other_len && $other_changed[$j])
|
||||
$j++;
|
||||
|
||||
while ($i < $len && !$changed[$i]) {
|
||||
assert($j < $other_len && ! $other_changed[$j]);
|
||||
$i++;
|
||||
$j++;
|
||||
while ($j < $other_len && $other_changed[$j])
|
||||
$j++;
|
||||
}
|
||||
|
||||
if ($i == $len)
|
||||
break;
|
||||
|
||||
$start = $i;
|
||||
|
||||
// Find the end of this run of changes.
|
||||
while (++$i < $len && $changed[$i])
|
||||
continue;
|
||||
|
||||
do {
|
||||
/*
|
||||
* Record the length of this run of changes, so that
|
||||
* we can later determine whether the run has grown.
|
||||
*/
|
||||
$runlength = $i - $start;
|
||||
|
||||
/*
|
||||
* Move the changed region back, so long as the
|
||||
* previous unchanged line matches the last changed one.
|
||||
* This merges with previous changed regions.
|
||||
*/
|
||||
while ($start > 0 && $lines[$start - 1] == $lines[$i - 1]) {
|
||||
$changed[--$start] = 1;
|
||||
$changed[--$i] = false;
|
||||
while ($start > 0 && $changed[$start - 1])
|
||||
$start--;
|
||||
assert($j > 0);
|
||||
while ($other_changed[--$j])
|
||||
continue;
|
||||
assert($j >= 0 && !$other_changed[$j]);
|
||||
}
|
||||
|
||||
/*
|
||||
* Set CORRESPONDING to the end of the changed run, at the last
|
||||
* point where it corresponds to a changed run in the other file.
|
||||
* CORRESPONDING == LEN means no such point has been found.
|
||||
*/
|
||||
$corresponding = $j < $other_len ? $i : $len;
|
||||
|
||||
/*
|
||||
* Move the changed region forward, so long as the
|
||||
* first changed line matches the following unchanged one.
|
||||
* This merges with following changed regions.
|
||||
* Do this second, so that if there are no merges,
|
||||
* the changed region is moved forward as far as possible.
|
||||
*/
|
||||
while ($i < $len && $lines[$start] == $lines[$i]) {
|
||||
$changed[$start++] = false;
|
||||
$changed[$i++] = 1;
|
||||
while ($i < $len && $changed[$i])
|
||||
$i++;
|
||||
|
||||
assert($j < $other_len && ! $other_changed[$j]);
|
||||
$j++;
|
||||
if ($j < $other_len && $other_changed[$j]) {
|
||||
$corresponding = $i;
|
||||
while ($j < $other_len && $other_changed[$j])
|
||||
$j++;
|
||||
}
|
||||
}
|
||||
} while ($runlength != $i - $start);
|
||||
|
||||
/*
|
||||
* If possible, move the fully-merged run of changes
|
||||
* back to a corresponding run in the other file.
|
||||
*/
|
||||
while ($corresponding < $i) {
|
||||
$changed[--$start] = 1;
|
||||
$changed[--$i] = 0;
|
||||
assert($j > 0);
|
||||
while ($other_changed[--$j])
|
||||
continue;
|
||||
assert($j >= 0 && !$other_changed[$j]);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Class representing a 'diff' between two sequences of strings.
|
||||
*/
|
||||
class Diff
|
||||
{
|
||||
public $edits;
|
||||
|
||||
/**
|
||||
* Computes diff between sequences of strings.
|
||||
*
|
||||
* @param $from_lines array An array of strings.
|
||||
* (Typically these are lines from a file.)
|
||||
* @param $to_lines array An array of strings.
|
||||
*/
|
||||
function __construct($from_lines, $to_lines)
|
||||
{
|
||||
$eng = new DiffEngine();
|
||||
$this->edits = $eng->diff($from_lines, $to_lines);
|
||||
}
|
||||
|
||||
/**
|
||||
* Check for empty diff.
|
||||
*
|
||||
* @return bool True iff two sequences were identical.
|
||||
*/
|
||||
public function isEmpty()
|
||||
{
|
||||
foreach ($this->edits as $edit) {
|
||||
if ($edit->type != 'copy')
|
||||
return false;
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
/**
|
||||
* Get the original set of lines.
|
||||
*
|
||||
* This reconstructs the $from_lines parameter passed to the
|
||||
* constructor.
|
||||
*
|
||||
* @return array The original sequence of strings.
|
||||
*/
|
||||
public function orig()
|
||||
{
|
||||
$lines = array();
|
||||
|
||||
foreach ($this->edits as $edit) {
|
||||
if ($edit->orig)
|
||||
array_splice($lines, sizeof($lines ?? []), 0, $edit->orig);
|
||||
}
|
||||
return $lines;
|
||||
}
|
||||
|
||||
/**
|
||||
* Get the final set of lines.
|
||||
*
|
||||
* This reconstructs the $to_lines parameter passed to the
|
||||
* constructor.
|
||||
*
|
||||
* @return array The sequence of strings.
|
||||
*/
|
||||
public function finalize()
|
||||
{
|
||||
$lines = array();
|
||||
|
||||
foreach ($this->edits as $edit) {
|
||||
if ($edit->final)
|
||||
array_splice($lines, sizeof($lines ?? []), 0, $edit->final);
|
||||
}
|
||||
return $lines;
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* FIXME: bad name.
|
||||
*/
|
||||
class MappedDiff
|
||||
extends Diff
|
||||
{
|
||||
/**
|
||||
* Computes diff between sequences of strings.
|
||||
*
|
||||
* This can be used to compute things like
|
||||
* case-insensitve diffs, or diffs which ignore
|
||||
* changes in white-space.
|
||||
*
|
||||
* @param $from_lines array An array of strings.
|
||||
* (Typically these are lines from a file.)
|
||||
*
|
||||
* @param $to_lines array An array of strings.
|
||||
*
|
||||
* @param $mapped_from_lines array This array should
|
||||
* have the same size number of elements as $from_lines.
|
||||
* The elements in $mapped_from_lines and
|
||||
* $mapped_to_lines are what is actually compared
|
||||
* when computing the diff.
|
||||
*
|
||||
* @param $mapped_to_lines array This array should
|
||||
* have the same number of elements as $to_lines.
|
||||
*/
|
||||
function __construct($from_lines, $to_lines,
|
||||
$mapped_from_lines, $mapped_to_lines)
|
||||
{
|
||||
|
||||
assert(sizeof($from_lines ?? []) == sizeof($mapped_from_lines ?? []));
|
||||
assert(sizeof($to_lines ?? []) == sizeof($mapped_to_lines ?? []));
|
||||
|
||||
parent::__construct($mapped_from_lines, $mapped_to_lines);
|
||||
|
||||
$xi = $yi = 0;
|
||||
// Optimizing loop invariants:
|
||||
// http://phplens.com/lens/php-book/optimizing-debugging-php.php
|
||||
for ($i = 0, $max = sizeof($this->edits); $i < $max; $i++) {
|
||||
$orig = &$this->edits[$i]->orig;
|
||||
if (is_array($orig)) {
|
||||
$orig = array_slice($from_lines ?? [], $xi ?? 0, sizeof($orig ?? []));
|
||||
$xi += sizeof($orig ?? []);
|
||||
}
|
||||
|
||||
$final = &$this->edits[$i]->final;
|
||||
if (is_array($final)) {
|
||||
$final = array_slice($to_lines ?? [], $yi ?? 0, sizeof($final ?? []));
|
||||
$yi += sizeof($final ?? []);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* A class to format Diffs
|
||||
*
|
||||
* This class formats the diff in classic diff format.
|
||||
* It is intended that this class be customized via inheritance,
|
||||
* to obtain fancier outputs.
|
||||
*/
|
||||
class DiffFormatter
|
||||
{
|
||||
/**
|
||||
* Number of leading context "lines" to preserve.
|
||||
*
|
||||
* This should be left at zero for this class, but subclasses
|
||||
* may want to set this to other values.
|
||||
*/
|
||||
public $leading_context_lines = 0;
|
||||
|
||||
/**
|
||||
* Number of trailing context "lines" to preserve.
|
||||
*
|
||||
* This should be left at zero for this class, but subclasses
|
||||
* may want to set this to other values.
|
||||
*/
|
||||
public $trailing_context_lines = 0;
|
||||
|
||||
/**
|
||||
* Format a diff.
|
||||
*
|
||||
* @param $diff object A Diff object.
|
||||
* @return string The formatted output.
|
||||
*/
|
||||
public function format($diff)
|
||||
{
|
||||
|
||||
$xi = $yi = 1;
|
||||
$block = false;
|
||||
$context = array();
|
||||
|
||||
$nlead = $this->leading_context_lines;
|
||||
$ntrail = $this->trailing_context_lines;
|
||||
|
||||
$this->start_diff();
|
||||
$x0 = 0;
|
||||
$y0 = 0;
|
||||
|
||||
foreach ($diff->edits as $edit) {
|
||||
if ($edit->type == 'copy') {
|
||||
if (is_array($block)) {
|
||||
if (sizeof($edit->orig ?? []) <= $nlead + $ntrail) {
|
||||
$block[] = $edit;
|
||||
} else {
|
||||
if ($ntrail) {
|
||||
$context = array_slice($edit->orig ?? [], 0, $ntrail);
|
||||
$block[] = new DiffOp_Copy($context);
|
||||
}
|
||||
$this->block($x0, $ntrail + $xi - $x0,
|
||||
$y0, $ntrail + $yi - $y0,
|
||||
$block);
|
||||
$block = false;
|
||||
}
|
||||
}
|
||||
$context = $edit->orig;
|
||||
} else {
|
||||
if (!is_array($block)) {
|
||||
$context = array_slice($context ?? [], max(0, sizeof($context ?? []) - $nlead) ?? 0);
|
||||
$x0 = $xi - sizeof($context ?? []);
|
||||
$y0 = $yi - sizeof($context ?? []);
|
||||
$block = array();
|
||||
if ($context)
|
||||
$block[] = new DiffOp_Copy($context);
|
||||
}
|
||||
$block[] = $edit;
|
||||
}
|
||||
|
||||
if ($edit->orig)
|
||||
$xi += sizeof($edit->orig ?? []);
|
||||
if ($edit->final)
|
||||
$yi += sizeof($edit->final ?? []);
|
||||
}
|
||||
|
||||
if (is_array($block))
|
||||
$this->block($x0, $xi - $x0,
|
||||
$y0, $yi - $y0,
|
||||
$block);
|
||||
|
||||
return $this->end_diff();
|
||||
}
|
||||
|
||||
private function block($xbeg, $xlen, $ybeg, $ylen, &$edits)
|
||||
{
|
||||
$this->start_block($this->block_header($xbeg, $xlen, $ybeg, $ylen));
|
||||
foreach ($edits as $edit) {
|
||||
if ($edit->type == 'copy')
|
||||
$this->context($edit->orig);
|
||||
elseif ($edit->type == 'add')
|
||||
$this->added($edit->final);
|
||||
elseif ($edit->type == 'delete')
|
||||
$this->deleted($edit->orig);
|
||||
elseif ($edit->type == 'change')
|
||||
$this->changed($edit->orig, $edit->final);
|
||||
else
|
||||
trigger_error("Unknown edit type", E_USER_ERROR);
|
||||
}
|
||||
$this->end_block();
|
||||
}
|
||||
|
||||
protected function start_diff()
|
||||
{
|
||||
ob_start();
|
||||
}
|
||||
|
||||
protected function end_diff()
|
||||
{
|
||||
$val = ob_get_contents();
|
||||
ob_end_clean();
|
||||
return $val;
|
||||
}
|
||||
|
||||
protected function block_header($xbeg, $xlen, $ybeg, $ylen)
|
||||
{
|
||||
if ($xlen > 1)
|
||||
$xbeg .= "," . ($xbeg + $xlen - 1);
|
||||
if ($ylen > 1)
|
||||
$ybeg .= "," . ($ybeg + $ylen - 1);
|
||||
|
||||
return $xbeg . ($xlen ? ($ylen ? 'c' : 'd') : 'a') . $ybeg;
|
||||
}
|
||||
|
||||
protected function start_block($header)
|
||||
{
|
||||
echo $header;
|
||||
}
|
||||
|
||||
protected function end_block()
|
||||
{
|
||||
}
|
||||
|
||||
protected function lines($lines, $prefix = ' ')
|
||||
{
|
||||
foreach ($lines as $line)
|
||||
echo "$prefix $line\n";
|
||||
}
|
||||
|
||||
protected function context($lines)
|
||||
{
|
||||
$this->lines($lines);
|
||||
}
|
||||
|
||||
protected function added($lines)
|
||||
{
|
||||
$this->lines($lines, ">");
|
||||
}
|
||||
|
||||
protected function deleted($lines)
|
||||
{
|
||||
$this->lines($lines, "<");
|
||||
}
|
||||
|
||||
protected function changed($orig, $final)
|
||||
{
|
||||
$this->deleted($orig);
|
||||
echo "---\n";
|
||||
$this->added($final);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* "Unified" diff formatter.
|
||||
*
|
||||
* This class formats the diff in classic "unified diff" format.
|
||||
*/
|
||||
class UnifiedDiffFormatter extends DiffFormatter
|
||||
{
|
||||
function __construct($context_lines = 4)
|
||||
{
|
||||
$this->leading_context_lines = $context_lines;
|
||||
$this->trailing_context_lines = $context_lines;
|
||||
}
|
||||
|
||||
protected function block_header($xbeg, $xlen, $ybeg, $ylen)
|
||||
{
|
||||
if ($xlen != 1)
|
||||
$xbeg .= "," . $xlen;
|
||||
if ($ylen != 1)
|
||||
$ybeg .= "," . $ylen;
|
||||
return "@@ -$xbeg +$ybeg @@\n";
|
||||
}
|
||||
|
||||
protected function added($lines)
|
||||
{
|
||||
$this->lines($lines, "+");
|
||||
}
|
||||
|
||||
protected function deleted($lines)
|
||||
{
|
||||
$this->lines($lines, "-");
|
||||
}
|
||||
|
||||
protected function changed($orig, $final)
|
||||
{
|
||||
$this->deleted($orig);
|
||||
$this->added($final);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* block conflict diff formatter.
|
||||
*
|
||||
* This class will format a diff identical to Diff3 (i.e. editpage
|
||||
* conflicts), but when there are only two source files. To be used by
|
||||
* future enhancements to reloading / upgrading pgsrc.
|
||||
*
|
||||
* Functional but not finished yet, need to eliminate redundant block
|
||||
* suffixes (i.e. "=======" immediately followed by another prefix)
|
||||
* see class LoadFileConflictPageEditor
|
||||
*/
|
||||
class BlockDiffFormatter extends DiffFormatter
|
||||
{
|
||||
function __construct($context_lines = 4)
|
||||
{
|
||||
$this->leading_context_lines = $context_lines;
|
||||
$this->trailing_context_lines = $context_lines;
|
||||
}
|
||||
|
||||
protected function lines($lines, $prefix = '')
|
||||
{
|
||||
if (!$prefix == '')
|
||||
echo "$prefix\n";
|
||||
foreach ($lines as $line)
|
||||
echo "$line\n";
|
||||
if (!$prefix == '')
|
||||
echo "$prefix\n";
|
||||
}
|
||||
|
||||
protected function added($lines)
|
||||
{
|
||||
$this->lines($lines, ">>>>>>>");
|
||||
}
|
||||
|
||||
protected function deleted($lines)
|
||||
{
|
||||
$this->lines($lines, "<<<<<<<");
|
||||
}
|
||||
|
||||
protected function block_header($xbeg, $xlen, $ybeg, $ylen)
|
||||
{
|
||||
return "";
|
||||
}
|
||||
|
||||
protected function changed($orig, $final)
|
||||
{
|
||||
$this->deleted($orig);
|
||||
$this->added($final);
|
||||
}
|
||||
}
|
||||
Loading…
Reference in New Issue
Block a user