Another fix for unescaped double quotes

crwlrsoft · Nov 20, 2024 · 89d58f4 · 89d58f4
1 parent e2ed0ec
commit 89d58f4
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 8 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -6,6 +6,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [1.1.2] - 2024-11-20
+### Fixed
+- Another improvement for parsing invalid JSON with unescaped double quote characters inside string values.
+
 ## [1.1.1] - 2024-11-20
 ### Fixed
 - Improve parsing invalid JSON by detecting and fixing unescaped double quote characters inside string values.

diff --git a/src/Json.php b/src/Json.php
@@ -66,14 +66,18 @@ function ($match) {
         ) ?? $jsonString;
 
         // If JSON string contains unescaped double quotes inside a string value, try to fix it.
-        if (preg_match('/".+?": "(.+(?<!\\\\)".+)"\s*?(,|})/', $jsonString)) {
-            $jsonString = preg_replace_callback('/".+?": "(.+(?<!\\\\)".+)"\s*?(,|})/', function ($match) {
-                return str_replace(
-                    $match[1],
-                    str_replace('"', '\"', $match[1]),
-                    $match[0],
-                );
-            }, $jsonString) ?? $jsonString;
+        if (preg_match('/"[^",}]*?":\s*?"((?:[^",}]*?(?<!\\\\)")+[^"]*?)"\s*?(,|})/', $jsonString)) {
+            $jsonString = preg_replace_callback(
+                '/"[^",}]*?":\s*?"((?:[^",}]*?(?<!\\\\)")+[^"]*?)"\s*?(,|})/',
+                function ($match) {
+                    return str_replace(
+                        $match[1],
+                        str_replace('"', '\"', $match[1]),
+                        $match[0],
+                    );
+                },
+                $jsonString,
+            ) ?? $jsonString;
         }
 
         return $jsonString;

diff --git a/tests/JsonTest.php b/tests/JsonTest.php
@@ -75,6 +75,18 @@
     ]);
 });
 
+it('correctly fixes unescaped double quotes inside a string value in one-line JSON', function () {
+    $jsonString = <<<JSON
+        { "foo": "bar", "description": "lorem "ipsum" asdf", "baz": "quz "yo" "" "lo"" }
+        JSON;
+
+    expect(Json::stringToArray($jsonString))->toBe([
+        'foo' => 'bar',
+        'description' => 'lorem "ipsum" asdf',
+        'baz' => 'quz "yo" "" "lo"',
+    ]);
+});
+
 it('throws an exception when the string is not a (valid) JSON string', function () {
     Json::stringToArray('{ foo: bar ]');
 })->throws(InvalidJsonException::class);