Merge branch 'hotfix/112-performance-chunked-decoding' into develop

Forward port #112

Author: weierophinney

CHANGELOG.md

@@ -40,6 +40,8 @@ All notable changes to this project will be documented in this file, in reverse
 
 ### Fixed
 
+- [#112](https://github.com/zendframework/zend-http/pull/112) provides performance improvements when parsing large chunked messages.
+
 - introduces changes to `Response::fromString()` to pull the next line of the response
   and parse it for the status when a 100 status code is initially encountered, per https://tools.ietf.org/html/rfc7231\#section-6.2.1
 

src/Response.php

@@ -524,17 +524,24 @@ protected function decodeChunkedBody($body)
     {
         $decBody = '';
 
-        while (trim($body)) {
-            if (! preg_match("/^([\da-fA-F]+)[^\r\n]*\r\n/sm", $body, $m)) {
-                throw new Exception\RuntimeException(
-                    "Error parsing body - doesn't seem to be a chunked message"
-                );
+        $offset = 0;
+
+        while (true) {
+            if (! preg_match("/^([\da-fA-F]+)[^\r\n]*\r\n/sm", $body, $m, 0, $offset)) {
+                if (trim(substr($body, $offset))) {
+                    // Message was not consumed completely!
+                    throw new Exception\RuntimeException(
+                        'Error parsing body - doesn\'t seem to be a chunked message'
+                    );
+                }
+                // Message was consumed completely
+                break;
             }
 
             $length   = hexdec(trim($m[1]));
             $cut      = strlen($m[0]);
-            $decBody .= substr($body, $cut, $length);
-            $body     = substr($body, $cut + $length + 2);
+            $decBody .= substr($body, $offset + $cut, $length);
+            $offset += $cut + $length + 2;
         }
 
         return $decBody;

test/ResponseTest.php

@@ -1,7 +1,7 @@
 <?php
 /**
  * @see       https://github.com/zendframework/zend-http for the canonical source repository
- * @copyright Copyright (c) 2005-2017 Zend Technologies USA Inc. (http://www.zend.com)
+ * @copyright Copyright (c) 2005-2018 Zend Technologies USA Inc. (http://www.zend.com)
  * @license   https://github.com/zendframework/zend-http/blob/master/LICENSE.md New BSD License
  */
 
@@ -179,6 +179,67 @@ public function testChunkedResponseCaseInsensitiveZF5438()
         $this->assertEquals('c0cc9d44790fa2a58078059bab1902a9', md5($res->getContent()));
     }
 
+    /**
+     * @param int $chunksize the data size of the chunk to create
+     * @return string a chunk of data for embedding inside a chunked response
+     */
+    private function makeChunk($chunksize)
+    {
+        return sprintf("%d\r\n%s\r\n", $chunksize, str_repeat('W', $chunksize));
+    }
+
+    /**
+     * @param Response $response
+     * @return float the time that calling the getBody function took on the response
+     */
+    private function getTimeForGetBody(Response $response)
+    {
+        $timeStart = microtime(true);
+        $response->getBody();
+        return microtime(true) - $timeStart;
+    }
+
+    /**
+     * @small
+     */
+    public function testChunkedResponsePerformance()
+    {
+        $headers = new Headers();
+        $headers->addHeaders([
+            'Date'              => 'Sun, 25 Jun 2006 19:55:19 GMT',
+            'Server'            => 'Apache',
+            'X-powered-by'      => 'PHP/5.1.4-pl3-gentoo',
+            'Connection'        => 'close',
+            'Transfer-encoding' => 'chunked',
+            'Content-type'      => 'text/html',
+        ]);
+
+        $response = new Response();
+        $response->setHeaders($headers);
+
+        // avoid flakiness, repeat test
+        $timings = [];
+        for ($i = 0; $i < 4; $i++) {
+            // get baseline for timing: 2000 x 1 Byte chunks
+            $responseData = str_repeat($this->makeChunk(1), 2000);
+            $response->setContent($responseData);
+            $time1 = $this->getTimeForGetBody($response);
+
+            // 'worst case' response, where 2000 1 Byte chunks are followed by a 10 MB Chunk
+            $responseData2 = $responseData . $this->makeChunk(10000000);
+            $response->setContent($responseData2);
+            $time2 = $this->getTimeForGetBody($response);
+
+            $timings[] = floor($time2 / $time1);
+        }
+
+        array_shift($timings); // do not measure first iteration
+
+        // make sure that the worst case packet will have an equal timing as the baseline
+        $errMsg = 'Chunked response is not parsing large packets efficiently! Timings:';
+        $this->assertLessThan(20, min($timings), $errMsg . print_r($timings, true));
+    }
+
     public function testLineBreaksCompatibility()
     {
         $responseTestLf = $this->readResponse('response_lfonly');