From 900e462bcdf4b3dec97eccde6f92abc7e091394d Mon Sep 17 00:00:00 2001
From: Dustin Wilson <dustin@dustinwilson.com>
Date: Thu, 1 Jul 2021 17:01:15 -0500
Subject: [PATCH] Wrote JSON to Grammar converter, successfully parses all
 grammars tested

---
 data/xml/xml.json                    |  12 +-
 lib/FauxReadOnly.php                 |   6 +-
 lib/Grammar.php                      | 168 ++++++++++++++++++++++-----
 lib/Grammar/CaptureList.php          |  18 +++
 lib/Grammar/GrammarInclude.php       |  35 ++++++
 lib/Grammar/ImmutableList.php        |  31 ++++-
 lib/Grammar/Include.php              |  32 -----
 lib/Grammar/InjectionList.php        |   2 +-
 lib/Grammar/NamedPatternListList.php |   2 +-
 lib/Grammar/Pattern.php              |  10 +-
 lib/Grammar/PatternList.php          |   4 +-
 lib/Scope/Matchers/GroupMatcher.php  |   4 +-
 lib/Scope/Matchers/PathMatcher.php   |   4 +-
 lib/Scope/Parser.php                 |   4 +-
 14 files changed, 241 insertions(+), 91 deletions(-)
 create mode 100644 lib/Grammar/CaptureList.php
 create mode 100644 lib/Grammar/GrammarInclude.php
 delete mode 100644 lib/Grammar/Include.php

diff --git a/data/xml/xml.json b/data/xml/xml.json
index 921fdf0..036bc80 100644
--- a/data/xml/xml.json
+++ b/data/xml/xml.json
@@ -431,12 +431,12 @@
         {
           "begin": "<%--",
           "captures": {
-            "0": {
-              "name": "punctuation.definition.comment.xml"
-            },
-            "end": "--%>",
-            "name": "comment.block.xml"
-          }
+              "0": {
+                  "name": "punctuation.definition.comment.xml"
+              }
+          },
+          "end": "--%>",
+          "name": "comment.block.xml"
         },
         {
           "begin": "<!--",
diff --git a/lib/FauxReadOnly.php b/lib/FauxReadOnly.php
index 8cd977e..0018a12 100644
--- a/lib/FauxReadOnly.php
+++ b/lib/FauxReadOnly.php
@@ -8,11 +8,11 @@ namespace dW\Lit;
 
 trait FauxReadOnly {
     public function __get(string $name) {
-        if ($name[0] !== '_') {
+        if ($name[0] === '_') {
             return;
         }
 
-        $name = substr($name, 1);
-        return $this->$name;
+        $prop = "_$name";
+        return $this->$prop;
     }
 }
\ No newline at end of file
diff --git a/lib/Grammar.php b/lib/Grammar.php
index ba09ff7..a8014c5 100644
--- a/lib/Grammar.php
+++ b/lib/Grammar.php
@@ -5,23 +5,26 @@
 
 declare(strict_types=1);
 namespace dW\Lit;
-use dW\Lit\Grammar\InjectionList,
+use dW\Lit\Grammar\CaptureList,
+    dW\Lit\Grammar\GrammarInclude,
+    dW\Lit\Grammar\InjectionList,
+    dW\Lit\Grammar\Pattern,
     dW\Lit\Grammar\PatternList,
     dW\Lit\Grammar\Repository;
 
 class Grammar {
     use FauxReadOnly;
 
-    protected string|null $_contentRegex;
-    protected string|null $_firstLineMatch;
-    protected InjectionList|null $_injections;
-    protected string $_name;
+    protected ?string $_contentRegex;
+    protected ?string $_firstLineMatch;
+    protected ?InjectionList $_injections;
+    protected ?string $_name;
     protected PatternList $_patterns;
-    protected Repository|null $_repository;
+    protected ?Repository $_repository;
     protected string $_scopeName;
 
 
-    public function __construct(string $name, string $scopeName, PatternList $patterns, string|null $contentRegex = null, string|null $firstLineMatch = null, InjectionList|null $injections = null, Repository|null $repository = null) {
+    public function __construct(string $scopeName, PatternList $patterns, ?string $name = null, ?string $contentRegex = null, ?string $firstLineMatch = null, ?InjectionList $injections = null, ?Repository $repository = null) {
         $this->_name = $name;
         $this->_scopeName = $scopeName;
         $this->_patterns = $patterns;
@@ -35,58 +38,159 @@ class Grammar {
     public static function fromJSON(string $jsonPath): self {
         assert(is_file($jsonPath), new \Exception("\"$jsonPath\" is either not a file or you do not have permission to read the file\n"));
 
-        $json = json_decode($jsonPath, true);
-        assert($json, new \Exception("\"$jsonPath\" is not a valid JSON file.\n"));
+        $json = json_decode(file_get_contents($jsonPath), true);
+        if ($json === null) {
+            $message = "Parsing \"$jsonPath\" failed with the following error: ";
+            switch (json_last_error()) {
+                case JSON_ERROR_DEPTH:
+                    $message .= 'Maximum stack depth exceeded';
+                break;
+                case JSON_ERROR_STATE_MISMATCH:
+                    $message .= 'Underflow or mode mismatch';
+                break;
+                case JSON_ERROR_CTRL_CHAR:
+                    $message .= 'Unexpected control character found';
+                break;
+                case JSON_ERROR_SYNTAX:
+                    $message .= 'Syntax error, malformed JSON';
+                break;
+                case JSON_ERROR_UTF8:
+                    $message .= 'Malformed UTF-8 characters, possibly incorrectly encoded';
+                break;
+                default:
+                    $message .= 'Unknown error';
+                break;
+            }
+
+            throw new \Exception("$message\n");
+        }
 
-        assert(isset($json['name']), new \Exception("\"$jsonPath\" does not have the required name property"));
         assert(isset($json['scopeName']), new \Exception("\"$jsonPath\" does not have the required scopeName property"));
         assert(isset($json['patterns']), new \Exception("\"$jsonPath\" does not have the required patterns property"));
 
-        $name = $json['name'];
+        $name = $json['name'] ?? null;
         $scopeName = $json['scopeName'];
         $contentRegex = (isset($json['contentRegex'])) ? "/{$json['contentRegex']}/" : null;
         $firstLineMatch = (isset($json['firstLineMatch'])) ? "/{$json['firstLineMatch']}/" : null;
 
-        $patterns = [];
-        foreach ($json['patterns'] as $pattern) {
-            foreach ($pattern as $key => $p) {
-
-            }
-        }
-
-        if (count($patterns) > 0) {
-            $patterns = new PatternList(...$patterns);
-        } else {
-            $patterns = null;
-        }
+        $patterns = self::parseJSONPatternList($json['patterns']);
 
+        $injections = null;
         if (isset($json['injections'])) {
             $injections = [];
-            foreach ($json['injections'] as $injection) {
-
+            foreach ($json['injections'] as $key => $injection) {
+                $injsections[$key] = (count($injection) === 1 && key($injection) === 'patterns') ?  self::parseJSONPatternList($injection['patterns']) : self::parseJSONPattern($injection);
             }
 
             if (count($injections) > 0) {
                 $injections = new InjectionList($injections);
             } else {
-                $patterns = null;
+                $injections = null;
             }
         }
 
-
+        $repository = null;
         if (isset($json['repository'])) {
             $respository = [];
-            foreach ($json['repository'] as $r) {
-
+            foreach ($json['repository'] as $key => $r) {
+                $repository[$key] = (count($r) === 1 && key($r) === 'patterns') ? self::parseJSONPatternList($r['patterns']) : self::parseJSONPattern($r);
             }
 
             if (count($repository) > 0) {
-                $repository = new InjectionList($repository);
+                $repository = new Repository($repository);
             } else {
                 $repository = null;
             }
         }
-        
-        return new self($name, $scopeName, $patterns, $contentRegex, $firstLineMatch, $injections, $repository);
+
+        return new self($scopeName, $patterns, $name, $contentRegex, $firstLineMatch, $injections, $repository);
+    }
+
+
+    protected static function parseJSONPattern(array $pattern): GrammarInclude|Pattern|null {
+        if (array_keys($pattern) === [ 'include' ]) {
+            return new GrammarInclude($pattern['include']);
+        }
+
+        $p = [
+            'name' => null,
+            'contentName' => null,
+            'begin' => null,
+            'end' => null,
+            'match' => null,
+            'patterns' => null,
+            'captures' => null,
+            'beginCaptures' => null,
+            'endCaptures' => null,
+            'applyEndPatternLast' => false
+        ];
+
+        $modified = false;
+        foreach ($pattern as $key => $value) {
+            switch ($key) {
+                case 'applyEndPatternLast':
+                    assert(is_bool($value) || (is_int($value) && ($value === 0 || $value === 1)), new \Exception("The value for applyEndPatternLast must be either a boolean, 0, or 1\n"));
+                    $value = (bool)$value;
+                case 'name':
+                case 'contentName':
+                    $p[$key] = $value;
+                    $modified = true;
+                break;
+                case 'begin':
+                case 'end':
+                case 'match':
+                    $p[$key] = "/$value/";
+                    $modified = true;
+                break;
+                case 'captures':
+                case 'beginCaptures':
+                case 'endCaptures':
+                    assert(is_array($value), new \Exception("Array value expected for '$key', found " . gettype($value) . "\n"));
+                    if (count($value) === 0) {
+                        continue 2;
+                    }
+
+                    $kk = array_keys($value);
+                    $v = array_values($value);
+
+                    // Skipping that bad three k variable name here... :)
+                    foreach ($kk as &$kkkk) {
+                        if (is_int($kkkk)) {
+                            continue;
+                        }
+
+                        assert(strspn($kkkk, '0123456789') === strlen($kkkk), new \Exception("\"$kkkk\" is not castable to an integer for use in a capture list\n"));
+                        $kkk = (int)$kkkk;
+                    }
+
+                    $v = array_map(function ($n) {
+                        return (count($n) === 1 && key($n) === 'patterns') ? self::parseJSONPatternList($n['patterns']) : self::parseJSONPattern($n);
+                    }, $v);
+
+                    $p[$key] = new CaptureList(array_combine($kk, $v));
+                    $modified = true;
+                break;
+                case 'patterns':
+                    assert(is_array($value), new \Exception("Array value expected for '$key', found " . gettype($value) . "\n"));
+                    $p[$key] = self::parseJSONPatternList($value);
+                    $modified = true;
+                break;
+            }
+        }
+
+        return ($modified) ? new Pattern(...$p) : null;
+    }
+
+
+    protected static function parseJSONPatternList(array $list): ?PatternList {
+        $result = [];
+        foreach ($list as $pattern) {
+            $p = self::parseJSONPattern($pattern);
+            if ($p !== null) {
+                $result[] = $p;
+            }
+        }
+
+        return (count($result) > 0) ? new PatternList(...$result) : null;
     }
 }
\ No newline at end of file
diff --git a/lib/Grammar/CaptureList.php b/lib/Grammar/CaptureList.php
new file mode 100644
index 0000000..f001f87
--- /dev/null
+++ b/lib/Grammar/CaptureList.php
@@ -0,0 +1,18 @@
+<?php
+/** @license MIT
+ * Copyright 2021 Dustin Wilson et al.
+ * See LICENSE file for details */
+
+declare(strict_types=1);
+namespace dW\Lit\Grammar;
+
+class CaptureList extends ImmutableList {
+    public function __construct(array $array) {
+        foreach ($array as $k => $v) {
+            assert(is_int($k), new \Exception('Integer index expected for supplied array, found ' . gettype($k) . "\n"));
+            assert($v instanceof GrammarInclude || $v instanceof Pattern || $v instanceof PatternList, new \Exception(__NAMESPACE__ . '\GrammarInclude, ' . __NAMESPACE__ . '\Pattern, or ' . __NAMESPACE__ . '\PatternList value expected for supplied array, found ' . gettype($v) . "\n"));
+        }
+
+        $this->storage = $array;
+    }
+}
\ No newline at end of file
diff --git a/lib/Grammar/GrammarInclude.php b/lib/Grammar/GrammarInclude.php
new file mode 100644
index 0000000..45bde0c
--- /dev/null
+++ b/lib/Grammar/GrammarInclude.php
@@ -0,0 +1,35 @@
+<?php
+/** @license MIT
+ * Copyright 2021 Dustin Wilson et al.
+ * See LICENSE file for details */
+
+declare(strict_types=1);
+namespace dW\Lit\Grammar;
+use dW\Lit\FauxReadOnly;
+
+class GrammarInclude {
+    use FauxReadOnly;
+
+    const BASE_TYPE = 0;
+    const REPOSITORY_TYPE = 1;
+    const SCOPE_TYPE = 2;
+    const SELF_TYPE = 3;
+
+    protected ?string $_name;
+    protected int $_type;
+
+
+    public function __construct(string $string) {
+        if ($string[0] === '#') {
+            $this->_type = self::REPOSITORY_TYPE;
+            $this->_name = substr($string, 1);
+        } elseif ($string === '$base') {
+            $this->_type = self::BASE_TYPE;
+        } elseif ($string === '$self') {
+            $this->_type = self::SELF_TYPE;
+        } else {
+            $this->_type = self::SCOPE_TYPE;
+            $this->_name = $string;
+        }
+    }
+}
\ No newline at end of file
diff --git a/lib/Grammar/ImmutableList.php b/lib/Grammar/ImmutableList.php
index 6801ac4..d5a2ed1 100644
--- a/lib/Grammar/ImmutableList.php
+++ b/lib/Grammar/ImmutableList.php
@@ -6,12 +6,14 @@
 declare(strict_types=1);
 namespace dW\Lit\Grammar;
 
-abstract class ImmutableList implements \ArrayAccess, \Countable {
-    protected $storage = [];
-    protected $count = 0;
+abstract class ImmutableList implements \ArrayAccess, \Countable, \Iterator {
+    protected int $count = 0;
+    protected int|string|null $position;
+    protected array $storage = [];
 
     public function __construct(...$values) {
         $this->storage = $values;
+        $this->count = count($this->storage);
     }
 
     public function offsetSet($offset, $value) {
@@ -31,6 +33,29 @@ abstract class ImmutableList implements \ArrayAccess, \Countable {
         return $this->storage[$offset];
     }
 
+    public function rewind() {
+        reset($this->storage);
+        $this->position = key($this->storage);
+    }
+
+    public function current() {
+        return current($this->storage);
+    }
+
+    public function key(){
+        $this->position = key($this->storage);
+        return $this->position;
+    }
+
+    public function next() {
+        next($this->storage);
+        $this->position = key($this->storage);
+    }
+
+    public function valid() {
+        return $this->offsetExists($this->position);
+    }
+
     public function count(): int {
         return $this->count;
     }
diff --git a/lib/Grammar/Include.php b/lib/Grammar/Include.php
deleted file mode 100644
index 4eb07ef..0000000
--- a/lib/Grammar/Include.php
+++ /dev/null
@@ -1,32 +0,0 @@
-<?php
-/** @license MIT
- * Copyright 2021 Dustin Wilson et al.
- * See LICENSE file for details */
-
-declare(strict_types=1);
-namespace dW\Lit\Grammar;
-use dW\Lit\FauxReadOnly;
-
-class Include {
-    use FauxReadOnly;
-
-    const REPOSITORY_TYPE = 0;
-    const SCOPE_TYPE = 1;
-    const SELF_TYPE = 2;
-
-    protected ?string $name;
-    protected int $type;
-
-
-    public function __construct(string $string) {
-        if ($string[0] === '#') {
-            $this->type = self::REPOSITORY_TYPE;
-            $this->name = substr($string, 1);
-        } elseif ($string === '$self') {
-            $this->type = self::SELF_TYPE;
-        }
-
-        $this->type = self::SCOPE_TYPE;
-        $this->name = $string;
-    }
-}
\ No newline at end of file
diff --git a/lib/Grammar/InjectionList.php b/lib/Grammar/InjectionList.php
index 60ffb72..55da791 100644
--- a/lib/Grammar/InjectionList.php
+++ b/lib/Grammar/InjectionList.php
@@ -6,4 +6,4 @@
 declare(strict_types=1);
 namespace dW\Lit\Grammar;
 
-class Repository extends NamedPatternListList {}
\ No newline at end of file
+class InjectionList extends NamedPatternListList {}
\ No newline at end of file
diff --git a/lib/Grammar/NamedPatternListList.php b/lib/Grammar/NamedPatternListList.php
index ffadaa8..0bf79e1 100644
--- a/lib/Grammar/NamedPatternListList.php
+++ b/lib/Grammar/NamedPatternListList.php
@@ -10,7 +10,7 @@ abstract class NamedPatternListList extends ImmutableList {
     public function __construct(array $array) {
         foreach ($array as $k => $v) {
             assert(is_string($k), new \Exception('String index expected for supplied array, found ' . gettype($k) . "\n"));
-            assert($v instanceof PatternList, new \Exception(__NAMESPACE__ . '\PatternList value expected for supplied array, found ' . gettype($v) . "\n"));
+            assert($v instanceof GrammarInclude || $v instanceof Pattern || $v instanceof PatternList, new \Exception(__NAMESPACE__ . '\GrammarInclude, ' . __NAMESPACE__ . '\Pattern, or ' . __NAMESPACE__ . '\PatternList value expected for supplied array, found ' . gettype($v) . "\n"));
         }
 
         $this->storage = $array;
diff --git a/lib/Grammar/Pattern.php b/lib/Grammar/Pattern.php
index 7e3aacc..77a24dd 100644
--- a/lib/Grammar/Pattern.php
+++ b/lib/Grammar/Pattern.php
@@ -13,26 +13,26 @@ class Pattern {
 
     protected bool $_applyEndPatternLast = false;
     protected ?string $_begin;
-    protected ?array $_beginCaptures;
-    protected ?array $_captures;
+    protected ?CaptureList $_beginCaptures;
+    protected ?CaptureList $_captures;
     protected ?string $_contentName;
     protected ?string $_end;
-    protected ?array $_endCaptures;
+    protected ?CaptureList $_endCaptures;
     protected ?string $_match;
     protected ?string $_name;
     protected ?PatternList $_patterns;
 
 
-    public function __construct(?string $name = null, ?string $contentName = null, ?string $begin = null, ?string $end = null, ?string $match = null, ?PatternList $patterns = null, ?string $include = null, ?array $captures = null, ?array $beginCaptures = null, ?array $endCaptures = null) {
+    public function __construct(?string $name = null, ?string $contentName = null, ?string $begin = null, ?string $end = null, ?string $match = null, ?PatternList $patterns = null, ?CaptureList $captures = null, ?CaptureList $beginCaptures = null, ?CaptureList $endCaptures = null, bool $applyEndPatternLast = false) {
         $this->_name = $name;
         $this->_contentName = $contentName;
         $this->_begin = $begin;
         $this->_end = $end;
         $this->_match = $match;
-
         $this->_patterns = $patterns;
         $this->_captures = $captures;
         $this->_beginCaptures = $beginCaptures;
         $this->_endCaptures = $endCaptures;
+        $this->_applyEndPatternLast = $applyEndPatternLast;
     }
 }
\ No newline at end of file
diff --git a/lib/Grammar/PatternList.php b/lib/Grammar/PatternList.php
index 4561595..74b9538 100644
--- a/lib/Grammar/PatternList.php
+++ b/lib/Grammar/PatternList.php
@@ -7,7 +7,7 @@ declare(strict_types=1);
 namespace dW\Lit\Grammar;
 
 class PatternList extends ImmutableList {
-    public function __construct(Pattern|Include ...$values) {
-        parent::__construct($values);
+    public function __construct(Pattern|GrammarInclude ...$values) {
+        parent::__construct(...$values);
     }
 }
\ No newline at end of file
diff --git a/lib/Scope/Matchers/GroupMatcher.php b/lib/Scope/Matchers/GroupMatcher.php
index e16f7b6..c171e36 100644
--- a/lib/Scope/Matchers/GroupMatcher.php
+++ b/lib/Scope/Matchers/GroupMatcher.php
@@ -7,10 +7,10 @@ declare(strict_types=1);
 namespace dW\Lit\Scope;
 
 class GroupMatcher extends Matcher {
-    protected string|null $prefix;
+    protected ?string $prefix;
     protected Matcher $selector;
 
-    public function __construct(string|null $prefix, Matcher $selector) {
+    public function __construct(?string $prefix, Matcher $selector) {
         $this->prefix = ($prefix !== null) ? $prefix[0] : null;
         $this->selector = $selector;
     }
diff --git a/lib/Scope/Matchers/PathMatcher.php b/lib/Scope/Matchers/PathMatcher.php
index 0dfe555..eea0f67 100644
--- a/lib/Scope/Matchers/PathMatcher.php
+++ b/lib/Scope/Matchers/PathMatcher.php
@@ -7,10 +7,10 @@ declare(strict_types=1);
 namespace dW\Lit\Scope;
 
 class PathMatcher extends Matcher {
-    protected string|null $prefix;
+    protected ?string $prefix;
     protected array $matchers;
 
-    public function __construct(string|null $prefix, ScopeMatcher ...$matchers) {
+    public function __construct(?string $prefix, ScopeMatcher ...$matchers) {
         $this->prefix = ($prefix !== null) ? $prefix[0] : null;
         $this->matchers = $matchers;
     }
diff --git a/lib/Scope/Parser.php b/lib/Scope/Parser.php
index 4d3f984..0cdc2c2 100644
--- a/lib/Scope/Parser.php
+++ b/lib/Scope/Parser.php
@@ -103,7 +103,7 @@ class Parser {
         return $result;
     }
 
-    protected static function parseGroup(string|null $prefix = null): Matcher {
+    protected static function parseGroup(?string $prefix = null): Matcher {
         if (self::$debug) {
             self::debug();
         }
@@ -123,7 +123,7 @@ class Parser {
         return $result;
     }
 
-    protected static function parsePath(string|null $prefix = null): PathMatcher|ScopeMatcher {
+    protected static function parsePath(?string $prefix = null): PathMatcher|ScopeMatcher {
         if (self::$debug) {
             self::debug();
         }