Issue codeplea#1 - update your utility

* Updates the code to use PSR-2 formatting (https://www.php-fig.org/psr/psr-2/). This requires some changes to formatting of class and method names.
* Adds composer to the project. Currently only pulls in `phpunit` as a development requirement.
* Adds `phpunit` tests created from the `example.php` code along with testing the various exceptions that can be thrown.
* Some code polish based on static analysis.

.gitignore

@@ -0,0 +1 @@
+/vendor/

AhoCorasick.php

@@ -0,0 +1,151 @@
+<?php
+
+/*
+ * ahocorasick - fast string searching in php
+ *
+ * Copyright (c) 2017-2018 Lewis Van Winkle
+ *
+ * http://CodePlea.com
+ *
+ * This software is provided 'as-is', without any express or implied
+ * warranty. In no event will the authors be held liable for any damages
+ * arising from the use of this software.
+ *
+ * Permission is granted to anyone to use this software for any purpose,
+ * including commercial applications, and to alter it and redistribute it
+ * freely, subject to the following restrictions:
+ *
+ * 1. The origin of this software must not be misrepresented; you must not
+ *    claim that you wrote the original software. If you use this software
+ *    in a product, an acknowledgement in the product documentation would be
+ *    appreciated but is not required.
+ * 2. Altered source versions must be plainly marked as such, and must not be
+ *    misrepresented as being the original software.
+ * 3. This notice may not be removed or altered from any source distribution.
+ *
+ */
+
+
+
+class AhoCorasick
+{
+
+    private $nodes = [ [] ];
+
+    private $final = 0;
+
+    /**
+     * Add a keyword to search for.
+     *
+     * @param string $needle
+     * @throws Exception
+     */
+    public function addNeedle(string $needle)
+    {
+        if ($this->final) {
+            throw new Exception('Cannot add word to finalized ahocorasick.');
+        }
+
+        $nodes = &$this->nodes;
+        $n = 0;
+
+        $needleLength = strlen($needle);
+
+        for ($i = 0; $i < $needleLength; ++$i) {
+            $c = $needle[$i];
+
+            if (!isset($nodes[$n][$c])) {
+                $nodes[$n][$c] = count($nodes);
+                $nodes[] = [];
+            }
+            $n = $nodes[$n][$c];
+        }
+
+        $nodes[$n][0][] = $needle;
+    }
+
+    /**
+     * Create the structure needed to search text for the given keywords.
+     * Once you call this, you cannot add additional keywords via addNeedle().
+     */
+    public function finalize()
+    {
+        $nodes = &$this->nodes;
+        $queue = [];
+
+        foreach ($nodes[0] as $j => $_) {
+            $nodes[$nodes[0][$j]][1] = 0;
+            $queue[] = $nodes[0][$j];
+        }
+
+        while (count($queue)) {
+            $r = $queue[0];
+            $queue = array_slice($queue, 1);
+
+            foreach ($nodes[$r] as $j => $_) {
+                if ($j === 0 || $j === 1) {
+                    continue;
+                }
+                $v = $nodes[$r][1];
+                $u = $nodes[$r][$j];
+                while ($v > 0 && !isset($nodes[$v][$j])) {
+                    $v = $nodes[$v][1];
+                }
+                $nodes[$u][1] = $nodes[$v][$j] ?? $v;
+                if (isset($nodes[$nodes[$u][1]][0])) {
+                    if (!isset($nodes[$u][0])) {
+                        $nodes[$u][0] = [];
+                    }
+                    $nodes[$u][0] = array_merge($nodes[$u][0], $nodes[$nodes[$u][1]][0]);
+                }
+                $queue[] = $u;
+            }
+        }
+
+        $this->final = 1;
+    }
+
+    /**
+     * Search the text for the given keywords.
+     *
+     * @param string $haystack
+     * @return array
+     * @throws Exception
+     */
+    public function search(string $haystack):array
+    {
+        if (!$this->final) {
+            throw new Exception('Must call finalize() before search.');
+        }
+
+        $nodes = &$this->nodes;
+        $found = [];
+        $n = 0;
+
+        $haystackLength = strlen($haystack);
+
+        for ($i = 0; $i < $haystackLength; ++$i) {
+            $c = $haystack[$i];
+
+            while (!array_key_exists($c, $nodes[$n]) && $n) {
+                $n = $nodes[$n][1];
+                if ($n === null) {
+                    die();
+                }
+            }
+
+            if (isset($nodes[$n][$c])) {
+                $n = $nodes[$n][$c];
+            }
+
+            if (isset($nodes[$n][0])) {
+                $z = $nodes[$n][0];
+                foreach ($z as $w) {
+                    $found[] = [$w, $i - strlen($w) + 1];
+                }
+            }
+        }
+
+        return $found;
+    }
+}

README.md

@@ -24,13 +24,13 @@ search text.
 Create, add keywords, and `finalize()`:
 
 ```php
-require('ahocorasick.php');
+require('AhoCorasick.php');
 
-$ac = new ahocorasick();
+$ac = new AhoCorasick();
 
-$ac->add_needle('art');
-$ac->add_needle('cart');
-$ac->add_needle('ted');
+$ac->addNeedle('art');
+$ac->addNeedle('cart');
+$ac->addNeedle('ted');
 
 $ac->finalize();
 
@@ -97,10 +97,17 @@ time: 0.054709911346436
 
 ```
 
-Note: the regex solutions are actually slightly broken. They won't work if you
+**Note:** the regex solutions are actually slightly broken. They won't work if you
 have a keyword that is a prefix or suffix of another. But hey, who really uses
 regex when it's not slightly broken?
 
-Also keep in mind that building the search tree (the `add_needle()` and
+Also keep in mind that building the search tree (the `addNeedle()` and
 `finalize()` calls) takes time. So you'll get the best speed-up if you're
 reusing the same keywords and calling `search()` many times.
+
+# Running tests
+
+```$php
+$ composer install
+$ ./vendor/bin/phpunit
+```

benchmark.php

@@ -4,57 +4,48 @@
 /* It compares our ahocorasick method with regex and strpos. */
 
 
-require('ahocorasick.php');
-require('benchmark_setup.php'); /* keywords and text */
+require 'AhoCorasick.php';
+require 'benchmark_setup.php'; /* keywords and text */
 
 $loops = 10;
 
-print("Loaded " . count($needles) . " keywords to search on a text of " .
-  strlen($haystack) . " characters.\n");
+print('Loaded ' . count($needles) . ' keywords to search on a text of ' .
+    strlen($haystack) . " characters.\n");
 
 print("\nSearching with strpos...\n");
 
 $st = microtime(1);
 for ($loop = 0; $loop < $loops; ++$loop) {
-  $found = array();
-  foreach($needles as $n) {
-    $k = 0;
-    while(($k = strpos($haystack, $n, $k)) !== FALSE) {
-      $found[] = array($n, $k);
-      ++$k;
+    $found = array();
+    foreach ($needles as $n) {
+        $k = 0;
+        while (($k = strpos($haystack, $n, $k)) !== false) {
+            $found[] = array($n, $k);
+            ++$k;
+        }
     }
-  }
 }
 $et = microtime(1);
-print("time: " . ($et - $st) . "\n");
+print('time: ' . ($et - $st) . "\n");
 $found_strpos = $found;
 
 
-
-
-
-
 print("\nSearching with preg_match...\n");
 //Note, this actually sucks and misses cases where one needle is a prefix or
 //suffix of another.
 $regex = '/' . implode('|', $needles) . '/';
 
 $st = microtime(1);
 for ($loop = 0; $loop < $loops; ++$loop) {
-  $found = array();
-  $k = 0;
-  while(preg_match($regex, $haystack, $m, PREG_OFFSET_CAPTURE, $k)) {
-    $found[] = $m[0];
-    $k = $m[0][1] + 1;
-  }
+    $found = array();
+    $k = 0;
+    while (preg_match($regex, $haystack, $m, PREG_OFFSET_CAPTURE, $k)) {
+        $found[] = $m[0];
+        $k = $m[0][1] + 1;
+    }
 }
 $et = microtime(1);
-print("time: " . ($et - $st) . "\n");
-//print_r($found);
-
-
-
-
+print('time: ' . ($et - $st) . "\n");
 
 
 print("\nSearching with preg_match_all...\n");
@@ -64,57 +55,51 @@
 
 $st = microtime(1);
 for ($loop = 0; $loop < $loops; ++$loop) {
-  $found = array();
-  $k = 0;
-  preg_match_all($regex, $haystack, $found, PREG_OFFSET_CAPTURE);
-  $found = $found[0];
+    $found = array();
+    $k = 0;
+    preg_match_all($regex, $haystack, $found, PREG_OFFSET_CAPTURE);
+    $found = $found[0];
 }
 $et = microtime(1);
-print("time: " . ($et - $st) . "\n");
-
-
-
+print('time: ' . ($et - $st) . "\n");
 
 
 print("\nSearching with aho corasick...\n");
-$ac = new ahocorasick();
-foreach ($needles as $n) $ac->add_needle($n);
+$ac = new AhoCorasick();
+foreach ($needles as $n) {
+    $ac->addNeedle($n);
+}
 $ac->finalize();
 
 $st = microtime(1);
 for ($loop = 0; $loop < $loops; ++$loop) {
-  $found = array();
-  $found = $ac->search($haystack);
+    $found = $ac->search($haystack);
 }
 $et = microtime(1);
-print("time: " . ($et - $st) . "\n");
-
-
-
-
+print('time: ' . ($et - $st) . "\n");
 
 
 //Check that the answers match.
 //First sort the arrays.
-$comp = function($a, $b) {return ($a[1] === $b[1]) ? ($a[0] > $b[0]) : ($a[1] > $b[1]);};
+$comp = function ($a, $b) {
+    return ($a[1] === $b[1]) ? ($a[0] > $b[0]) : ($a[1] > $b[1]);
+};
 usort($found, $comp);
 usort($found_strpos, $comp);
 
 if ($found_strpos !== $found) {
-  print("ERROR - Aho Corasick got the wrong result.\n");
+    print("ERROR - Aho Corasick got the wrong result.\n");
+
+    print('strpos size: ' . count($found_strpos) . "\n");
+    print('aho corasick size: ' . count($found) . "\n");
 
-  print("strpos size: " . count($found_strpos) . "\n");
-  print("aho corasick size: " . count($found) . "\n");
+    $numberFound = count($found);
 
-  for ($i = 0; $i < count($found); ++$i) {
-    if ($found_strpos[$i] !== $found[$i]) {
-      print("Mismatch $i\n");
-      print_r($found_strpos[$i]);
-      print_r($found[$i]);
+    for ($i = 0; $i < $numberFound; ++$i) {
+        if ($found_strpos[$i] !== $found[$i]) {
+            print("Mismatch $i\n");
+            print_r($found_strpos[$i]);
+            print_r($found[$i]);
+        }
     }
-  }
 }
-
-
-
-

benchmark_setup.php

@@ -1,6 +1,6 @@
 <?php
 
-$needles = array('abandonment', 'abashed', 'abashments', 'abduction',
+$needles = ['abandonment', 'abashed', 'abashments', 'abduction',
 'aberrant', 'abiding', 'abidingly', 'abjures', 'ablution', 'abolishes',
 'abominably', 'aborted', 'abrasion', 'abridgment', 'abscesses', 'absconds',
 'absences', 'absinthe', 'absolves', 'absorbingly', 'abundant', 'abused',
@@ -532,7 +532,7 @@
 'remonstrates', 'remorse', 'removes', 'remunerated', 'rendering',
 'renditions', 'reneged', 'renominate', 'renovators', 'reorders',
 'repatriates', 'repave', 'repaying', 'repeatedly', 'repertoires', 'replied',
-'reprehend', 'reprieves', 'reprimanded');
+'reprehend', 'reprieves', 'reprimanded'];
 
 $haystack = 'unscathed grampus antinuclear avenged waste oversee doggies spumes
 senators balk gooseberries grilles respelled ceramists outlaid maladroitly

composer.json

@@ -0,0 +1,15 @@
+{
+    "name": "codeplea/ahocorasickphp",
+    "description": "Aho-Corasick multi-keyword string searching library in PHP.",
+    "type": "library",
+    "config": {
+        "sort-packages": true
+    },
+    "require": {
+        "php": ">=7.0"
+    },
+    "require-dev": {
+        "phpunit/phpunit": "^7.3"
+    },
+    "license": "zlib"
+}