Normalize newlines in Comment::getReformattedText()

Normalize CRLF to LF in getReformattedText(). That was, if the comment is pretty-printed, we will use proper LF newlines, rather than inserting indentation between the CR and LF. At the same time, this also makes it easier to emit actual CRLF newlines with a custom pretty printer. Fixes #599.
nikic · May 21, 2023 · ad8daa1 · ad8daa1
1 parent 5883189
commit ad8daa1
Show file tree

Hide file tree

Showing 2 changed files with 29 additions and 37 deletions.
diff --git a/lib/PhpParser/Comment.php b/lib/PhpParser/Comment.php
@@ -150,19 +150,21 @@ public function __toString(): string {
      *
      * "Reformatted" here means that we try to clean up the whitespace at the
      * starts of the lines. This is necessary because we receive the comments
-     * without trailing whitespace on the first line, but with trailing whitespace
+     * without leading whitespace on the first line, but with leading whitespace
      * on all subsequent lines.
      *
+     * Additionally, this normalizes CRLF newlines to LF newlines.
+     *
      * @return string
      */
     public function getReformattedText(): string {
-        $text = $this->text;
+        $text = str_replace("\r\n", "\n", $this->text);
         $newlinePos = strpos($text, "\n");
         if (false === $newlinePos) {
             // Single line comments don't need further processing
             return $text;
         }
-        if (preg_match('((*BSR_ANYCRLF)(*ANYCRLF)^.*(?:\R\s+\*.*)+$)', $text)) {
+        if (preg_match('(^.*(?:\n\s+\*.*)+$)', $text)) {
             // Multi line comment of the type
             //
             //     /*
@@ -171,9 +173,9 @@ public function getReformattedText(): string {
             //      */
             //
             // is handled by replacing the whitespace sequences before the * by a single space
-            return preg_replace('(^\s+\*)m', ' *', $this->text);
+            return preg_replace('(^\s+\*)m', ' *', $text);
         }
-        if (preg_match('(^/\*\*?\s*[\r\n])', $text) && preg_match('(\n(\s*)\*/$)', $text, $matches)) {
+        if (preg_match('(^/\*\*?\s*\n)', $text) && preg_match('(\n(\s*)\*/$)', $text, $matches)) {
             // Multi line comment of the type
             //
             //    /*

diff --git a/test/PhpParser/CommentTest.php b/test/PhpParser/CommentTest.php
@@ -33,47 +33,37 @@ public function provideTestReformatting() {
             ['// Some text', '// Some text'],
             ['/* Some text */', '/* Some text */'],
             [
-                '/**
-     * Some text.
-     * Some more text.
-     */',
-                '/**
- * Some text.
- * Some more text.
- */'
+                "/**\n     * Some text.\n     * Some more text.\n     */",
+                "/**\n * Some text.\n * Some more text.\n */"
             ],
             [
-                '/*
-        Some text.
-        Some more text.
-    */',
-                '/*
-    Some text.
-    Some more text.
-*/'
+                "/**\r\n     * Some text.\r\n     * Some more text.\r\n     */",
+                "/**\n * Some text.\n * Some more text.\n */"
             ],
             [
-                '/* Some text.
-       More text.
-       Even more text. */',
-                '/* Some text.
-   More text.
-   Even more text. */'
+                "/*\n        Some text.\n        Some more text.\n    */",
+                "/*\n    Some text.\n    Some more text.\n*/"
             ],
             [
-                '/* Some text.
-       More text.
-         Indented text. */',
-                '/* Some text.
-   More text.
-     Indented text. */',
+                "/*\r\n        Some text.\r\n        Some more text.\r\n    */",
+                "/*\n    Some text.\n    Some more text.\n*/"
+            ],
+            [
+                "/* Some text.\n       More text.\n       Even more text. */",
+                "/* Some text.\n   More text.\n   Even more text. */"
+            ],
+            [
+                "/* Some text.\r\n       More text.\r\n       Even more text. */",
+                "/* Some text.\n   More text.\n   Even more text. */"
+            ],
+            [
+                "/* Some text.\n       More text.\n         Indented text. */",
+                "/* Some text.\n   More text.\n     Indented text. */",
             ],
             // invalid comment -> no reformatting
             [
-                'hallo
-    world',
-                'hallo
-    world',
+                "hello\n    world",
+                "hello\n    world",
             ],
         ];
     }