More micro-optimizations

CHANGELOG.md

@@ -11,6 +11,7 @@ Updates should follow the [Keep a CHANGELOG](https://keepachangelog.com/) princi
  - Any leading UTF-8 BOM will be stripped from the input
  - Mark the `getEnvironment()` method of `CommonMarkConverter` and `GithubFlavoredMarkdownConverter` as always returning the concrete, configurable `Environment` for upgrading convenience
  - Optimized AST iteration
+ - Lots of small micro-optimizations
 
 ## [2.0.0-beta2]
 

src/Extension/CommonMark/Parser/Block/FencedCodeParser.php

@@ -52,7 +52,10 @@ public function tryContinue(Cursor $cursor, BlockContinueParserInterface $active
         }
 
         // Skip optional spaces of fence offset
-        $cursor->match('/^ {0,' . $this->block->getOffset() . '}/');
+        // Optimization: don't attempt to match if we're at a non-space position
+        if ($cursor->getNextNonSpacePosition() > $cursor->getPosition()) {
+            $cursor->match('/^ {0,' . $this->block->getOffset() . '}/');
+        }
 
         return BlockContinue::at($cursor);
     }

src/Extension/CommonMark/Parser/Block/FencedCodeStartParser.php

@@ -22,12 +22,7 @@ final class FencedCodeStartParser implements BlockStartParserInterface
 {
     public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserState): ?BlockStart
     {
-        if ($cursor->isIndented()) {
-            return BlockStart::none();
-        }
-
-        $c = $cursor->getCurrentCharacter();
-        if ($c !== ' ' && $c !== "\t" && $c !== '`' && $c !== '~') {
+        if ($cursor->isIndented() || ! \in_array($cursor->getNextNonSpaceCharacter(), ['`', '~'], true)) {
             return BlockStart::none();
         }
 
@@ -38,9 +33,8 @@ public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserSta
         }
 
         // fenced code block
-        $fence       = \ltrim($fence, " \t");
-        $fenceLength = \strlen($fence);
+        $fence = \ltrim($fence, " \t");
 
-        return BlockStart::of(new FencedCodeParser($fenceLength, $fence[0], $indent))->at($cursor);
+        return BlockStart::of(new FencedCodeParser(\strlen($fence), $fence[0], $indent))->at($cursor);
     }
 }

src/Extension/CommonMark/Parser/Block/HeadingStartParser.php

@@ -23,7 +23,7 @@ class HeadingStartParser implements BlockStartParserInterface
 {
     public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserState): ?BlockStart
     {
-        if ($cursor->isIndented()) {
+        if ($cursor->isIndented() || ! \in_array($cursor->getNextNonSpaceCharacter(), ['#', '-', '='], true)) {
             return BlockStart::none();
         }
 
@@ -50,7 +50,7 @@ public function tryStart(Cursor $cursor, MarkdownParserStateInterface $parserSta
 
     private static function getAtxHeader(Cursor $cursor): ?HeadingParser
     {
-        $match = RegexHelper::matchFirst('/^#{1,6}(?:[ \t]+|$)/', $cursor->getLine(), $cursor->getNextNonSpacePosition());
+        $match = RegexHelper::matchFirst('/^#{1,6}(?:[ \t]+|$)/', $cursor->getRemainder());
         if (! $match) {
             return null;
         }
@@ -70,7 +70,7 @@ private static function getAtxHeader(Cursor $cursor): ?HeadingParser
 
     private static function getSetextHeadingLevel(Cursor $cursor): int
     {
-        $match = RegexHelper::matchFirst('/^(?:=+|-+)[ \t]*$/', $cursor->getLine(), $cursor->getNextNonSpacePosition());
+        $match = RegexHelper::matchFirst('/^(?:=+|-+)[ \t]*$/', $cursor->getRemainder());
         if ($match === null) {
             return 0;
         }

src/Extension/CommonMark/Parser/Block/ListItemParser.php

@@ -84,6 +84,7 @@ public function tryContinue(Cursor $cursor, BlockContinueParserInterface $active
             return BlockContinue::at($cursor);
         }
 
+        // Note: We'll hit this case for lazy continuation lines, they will get added later.
         return BlockContinue::none();
     }
 }

src/Parser/Cursor.php

@@ -142,11 +142,7 @@ public function getCharacter(?int $index = null): ?string
         }
 
         if ($this->isMultibyte) {
-            if (isset($this->charCache[$index])) {
-                return $this->charCache[$index];
-            }
-
-            return $this->charCache[$index] = \mb_substr($this->line, $index, 1, 'UTF-8');
+            return $this->charCache[$index] ??= \mb_substr($this->line, $index, 1, 'UTF-8');
         }
 
         return $this->line[$index];
@@ -162,11 +158,7 @@ public function getCurrentCharacter(): ?string
         }
 
         if ($this->isMultibyte) {
-            if (isset($this->charCache[$this->currentPosition])) {
-                return $this->charCache[$this->currentPosition];
-            }
-
-            return $this->charCache[$this->currentPosition] = \mb_substr($this->line, $this->currentPosition, 1, 'UTF-8');
+            return $this->charCache[$this->currentPosition] ??= \mb_substr($this->line, $this->currentPosition, 1, 'UTF-8');
         }
 
         return $this->line[$this->currentPosition];

src/Parser/InlineParserEngine.php

@@ -160,7 +160,7 @@ private function matchParsers(string $contents): array
                 \assert(\is_int($offset));
 
                 // Remove the offsets, keeping only the matched text
-                $m = \array_map(static fn (array $s): string => (string) $s[0], $match);
+                $m = \array_column($match, 0);
 
                 if ($m === []) {
                     continue;

src/Reference/ReferenceMap.php

@@ -57,11 +57,7 @@ public function get(string $label): ?ReferenceInterface
     {
         $label = $this->normalizer->normalize($label);
 
-        if (! isset($this->references[$label])) {
-            return null;
-        }
-
-        return $this->references[$label];
+        return $this->references[$label] ?? null;
     }
 
     /**

src/Util/HtmlElement.php

@@ -71,11 +71,7 @@ public function getAllAttributes(): array
      */
     public function getAttribute(string $key)
     {
-        if (! isset($this->attributes[$key])) {
-            return null;
-        }
-
-        return $this->attributes[$key];
+        return $this->attributes[$key] ?? null;
     }
 
     /**