Add new renderer option: wordGlues

Signed-off-by: Jack Cherng <jfcherng@gmail.com>

Author: jfcherng

README.md

@@ -112,6 +112,11 @@ $rendererOptions = [
     // it controls how the output JSON is formatted.
     // see availabe options on https://www.php.net/manual/en/function.json-encode.php
     'jsonEncodeFlags' => \JSON_UNESCAPED_SLASHES | \JSON_UNESCAPED_UNICODE,
+    // this option is currently effective when the "detailLevel" is "word"
+    // characters listed in this array can be used to make diff segments into a whole
+    // for example, making "<del>good</del>-<del>looking</del>" into "<del>good-looking</del>"
+    // this should bring better readability but set this to empty array if you do not want it
+    'wordGlues' => [' ', '-'],
     // change this value to a string as the returned diff if the two input strings are identical
     'resultForIdenticals' => null,
     // extra HTML classes added to the DOM of the diff container

example/demo.php

@@ -80,6 +80,11 @@
             // it controls how the output JSON is formatted.
             // see availabe options on https://www.php.net/manual/en/function.json-encode.php
             'jsonEncodeFlags' => \JSON_UNESCAPED_SLASHES | \JSON_UNESCAPED_UNICODE,
+            // this option is currently effective when the "detailLevel" is "word"
+            // characters listed in this array can be used to make diff segments into a whole
+            // for example, making "<del>good</del>-<del>looking</del>" into "<del>good-looking</del>"
+            // this should bring better readability but set this to empty array if you do not want it
+            'wordGlues' => [' ', '-'],
             // change this value to a string as the returned diff if the two input strings are identical
             'resultForIdenticals' => null,
             // extra HTML classes added to the DOM of the diff container

example/new_file.txt

@@ -1,4 +1,4 @@
-<div>Hello World!</div>
+<div>Hello World! Bad-tempered.</div>
 ~~~~~~~~~~~~~~~~~~~
 Let's add a new line here.
 X

example/old_file.txt

@@ -1,4 +1,4 @@
-<p>Hello World!</p>
+<p>Hello World! Good-looking.</p>
 ~~~~~~~~~~~~~~~~~~~
 X
 Y

src/Renderer/AbstractRenderer.php

@@ -71,6 +71,11 @@ abstract class AbstractRenderer implements RendererInterface
         // it controls how the output JSON is formatted.
         // see availabe options on https://www.php.net/manual/en/function.json-encode.php
         'jsonEncodeFlags' => \JSON_UNESCAPED_SLASHES | \JSON_UNESCAPED_UNICODE,
+        // this option is currently effective when the "detailLevel" is "word"
+        // characters listed in this array can be used to make diff segments into a whole
+        // for example, making "<del>good</del>-<del>looking</del>" into "<del>good-looking</del>"
+        // this should bring better readability but set this to empty array if you do not want it
+        'wordGlues' => ['-', ' '],
         // change this value to a string as the returned diff if the two input strings are identical
         'resultForIdenticals' => null,
         // extra HTML classes added to the DOM of the diff container

src/Renderer/Html/LineRenderer/Word.php

@@ -7,6 +7,7 @@
 use Jfcherng\Diff\Renderer\RendererConstant;
 use Jfcherng\Diff\SequenceMatcher;
 use Jfcherng\Diff\Utility\ReverseIterator;
+use Jfcherng\Diff\Utility\Str;
 use Jfcherng\Utility\MbString;
 
 final class Word extends AbstractLineRenderer
@@ -18,9 +19,14 @@ final class Word extends AbstractLineRenderer
      */
     public function render(MbString $mbOld, MbString $mbNew): LineRendererInterface
     {
-        static $splitRegex = '/([' . RendererConstant::PUNCTUATIONS_RANGE . ']++)/uS';
+        static $splitRegex = '/([' . RendererConstant::PUNCTUATIONS_RANGE . '])/uS';
 
-        $pregFlag = \PREG_SPLIT_DELIM_CAPTURE | \PREG_SPLIT_NO_EMPTY;
+        // using PREG_SPLIT_NO_EMPTY will make "wordGlues" work wrongly under some rare cases
+        // failure case:
+        //     old: "good-looking-"
+        //     new: "good--"
+        // notice that after glueing, the 1st "-" in the new should be in the diff segment
+        $pregFlag = \PREG_SPLIT_DELIM_CAPTURE;
         $oldWords = $mbOld->toArraySplit($splitRegex, -1, $pregFlag);
         $newWords = $mbNew->toArraySplit($splitRegex, -1, $pregFlag);
 
@@ -39,9 +45,58 @@ public function render(MbString $mbOld, MbString $mbNew): LineRendererInterface
             }
         }
 
+        if (!empty($hunk) && !empty($this->rendererOptions['wordGlues'])) {
+            $regexGlues = \array_map(
+                function (string $glue): string {
+                    return \preg_quote($glue, '/');
+                },
+                $this->rendererOptions['wordGlues']
+            );
+
+            $gluePattern = '/^(?:' . \implode('|', $regexGlues) . ')+$/uS';
+
+            $this->glueWordsResult($oldWords, $gluePattern);
+            $this->glueWordsResult($newWords, $gluePattern);
+        }
+
         $mbOld->set(\implode('', $oldWords));
         $mbNew->set(\implode('', $newWords));
 
         return $this;
     }
+
+    /**
+     * Beautify diff result by glueing words.
+     *
+     * What this function does is basically making
+     *     ["<diff_begin>good<diff_end>", "-", "<diff_begin>looking<diff_end>"]
+     * into
+     *     ["<diff_begin>good", "-", "looking<diff_end>"].
+     *
+     * @param array  $words       the words
+     * @param string $gluePattern the regex to determine a string is purely glue or not
+     */
+    protected function glueWordsResult(array &$words, string $gluePattern): void
+    {
+        /** @var int index of the word which has the trailing closure */
+        $endClosureIdx = -1;
+
+        foreach ($words as $idx => &$word) {
+            if ($word === '') {
+                continue;
+            }
+
+            if ($endClosureIdx < 0) {
+                if (Str::endsWith($word, RendererConstant::HTML_CLOSURES[1])) {
+                    $endClosureIdx = $idx;
+                }
+            } elseif (Str::startsWith($word, RendererConstant::HTML_CLOSURES[0])) {
+                $words[$endClosureIdx] = \substr($words[$endClosureIdx], 0, -\strlen(RendererConstant::HTML_CLOSURES[1]));
+                $word = \substr($word, \strlen(RendererConstant::HTML_CLOSURES[0]));
+                $endClosureIdx = $idx;
+            } elseif (!\preg_match($gluePattern, $word)) {
+                $endClosureIdx = -1;
+            }
+        }
+    }
 }

src/Utility/Str.php

@@ -0,0 +1,30 @@
+<?php
+
+declare(strict_types=1);
+
+namespace Jfcherng\Diff\Utility;
+
+final class Str
+{
+    /**
+     * Determine if a given string starts with a given substring.
+     *
+     * @param string $haystack the haystack
+     * @param string $needle   the needle
+     */
+    public static function startsWith(string $haystack, string $needle): bool
+    {
+        return \substr($haystack, 0, \strlen($needle)) === $needle;
+    }
+
+    /**
+     * Determine if a given string ends with a given substring.
+     *
+     * @param string $haystack the haystack
+     * @param string $needle   the needle
+     */
+    public static function endsWith(string $haystack, string $needle): bool
+    {
+        return \substr($haystack, -\strlen($needle)) === $needle;
+    }
+}