Format check + minor improvements + updated readme

Author: JanPetterMG

README.md

@@ -6,8 +6,8 @@
 [![Packagist](https://img.shields.io/packagist/v/vipnytt/useragentparser.svg)](https://packagist.org/packages/vipnytt/useragentparser)
 [![Chat](https://badges.gitter.im/VIPnytt/UserAgentParser.svg)](https://gitter.im/VIPnytt/UserAgentParser)
 
-# User-Agent string parser
-PHP class to parse User-Agent strings sent by web-crawlers.
+# User-Agent parser for robot rule sets
+Parser and group determiner optimalized for ``robots.txt``, ``X-Robots-tag`` and ``Robots-meta-tag`` usage cases.
 
 [![SensioLabsInsight](https://insight.sensiolabs.com/projects/1386c14c-546c-4c42-ac55-91ea3a3a1ae1/big.png)](https://insight.sensiolabs.com/projects/1386c14c-546c-4c42-ac55-91ea3a3a1ae1)
 
@@ -23,16 +23,17 @@ The library is available for install via [Composer](https://getcomposer.org). Ju
 Then run `composer update`.
 
 ## Features
-- Strip the version tag.
-- Find different groups the User-Agent belongs to.
-- Determine the correct group of records by finding the group with the most specific user-agent that still matches.
+- Stripping of the version tag.
+- List any _rule groups_ the User-Agent belongs to.
+- Determine the correct group of records by finding the group with the most specific User-agent that still matches.
 
-### When do I need it?
-- Parsing of `robots.txt`, the rules for robots online.
-- Parsing of the _X-Robots-Tag_ HTTP-header.
-- Parsing of _Robots meta tags_ in HTML documents
+### When to use it?
+- When parsing `robots.txt` rule sets, for robots online.
+- When parsing the ``X-Robots-Tag`` HTTP-header.
+- When parsing ``Robots meta tags`` in HTML documents
 
-Note: _The library is not compatible with User-Agent strings sent by eg. web-browsers. Contributions are of course welcome._
+Note: _Full User-agent strings, like them sent by eg. web-browsers, or found in your log files, are not compatible, this is by design._
+Supported User-agent string formats are ``UserAgentName/version`` with or without the version tag. Eg. ``MyWebCrawler/2.0`` or just ``MyWebCrawler``.
 
 
 ## Getting Started
@@ -46,7 +47,7 @@ var_dump($parser->stripVersion());
 /* googlebot */
 ```
 
-### Find different groups the User-Agent belongs to
+### List different groups the User-agent belongs to
 ```php
 use vipnytt\UserAgentParser;
 
@@ -60,7 +61,7 @@ var_dump($parser->export());
 ```
 
 ### Determine the correct group
-Determine the correct group of records by finding the group with the most specific user-agent that still matches
+Determine the correct group of records by finding the group with the most specific User-agent that still matches your rule sets
 ```php
 use vipnytt\UserAgentParser;
 

composer.json

@@ -1,6 +1,7 @@
 {
   "name": "vipnytt/useragentparser",
-  "description": "User-Agent string parser class",
+  "description": "User-Agent string parser",
+  "type": "library",
   "keywords": [
     "UserAgent",
     "User-Agent",
@@ -10,33 +11,37 @@
     "Robots-meta-tag",
     "Robot",
     "Spider",
-    "Web-Crawler"
+    "Crawler"
   ],
   "homepage": "https://github.com/VIPnytt/UserAgentParser",
-  "type": "library",
   "license": "MIT",
-  "minimum-stability": "dev",
-  "prefer-stable": true,
-  "require": {
-    "php": ">=5.4.0",
-    "ext-mbstring": "*"
-  },
-  "require-dev": {
-    "phpunit/phpunit": ">=4.0",
-    "codeclimate/php-test-reporter": ">=0.2.0"
-  },
   "authors": [
     {
       "name": "VIP nytt",
       "email": "vipnytt@gmail.com",
-      "role": "creator"
+      "role": "Creator"
     },
     {
       "name": "Jan-Petter Gundersen",
       "email": "europe.jpg@gmail.com",
-      "role": "developer"
+      "role": "Developer"
     }
   ],
+  "require": {
+    "php": ">=5.4.0",
+    "ext-mbstring": "*"
+  },
+  "require-dev": {
+    "phpunit/phpunit": ">=4.0",
+    "codeclimate/php-test-reporter": ">=0.2.0"
+  },
+  "support": {
+    "email": "vipnytt@gmail.com",
+    "issues": "https://github.com/VIPnytt/UserAgentParser/issues",
+    "wiki": "https://github.com/VIPnytt/UserAgentParser/wiki",
+    "source": "https://github.com/VIPnytt/UserAgentParser",
+    "docs": "https://github.com/VIPnytt/UserAgentParser/wiki"
+  },
   "autoload": {
     "psr-4": {
       "vipnytt\\": "src/"

src/UserAgentParser.php

@@ -1,7 +1,7 @@
 <?php
 namespace vipnytt;
 
-use Exception;
+use vipnytt\UserAgentParser\Exceptions\FormatException;
 
 /**
  * Class UserAgentParser
@@ -17,29 +17,46 @@ class UserAgentParser
      * Constructor
      *
      * @param string $userAgent
-     * @throws Exception
      */
     public function __construct($userAgent)
     {
         mb_detect_encoding($userAgent);
-        $this->userAgent = mb_strtolower(trim($userAgent));
+        $this->userAgent = trim($userAgent);
+        $this->checkFormat();
         $this->explode();
     }
 
+    /**
+     * Validate the UserAgent format
+     *
+     * @throws FormatException
+     */
+    protected function checkFormat()
+    {
+        if (preg_match('/\s/', $this->userAgent)) {
+            throw new FormatException("Format not supported. Please use `name/version` or just `name`, eg. `MyUserAgent/1.0` and `MyUserAgent`.");
+        }
+    }
+
     /**
      * Parses all possible User-Agent groups to an array
      *
      * @return array
      */
     private function explode()
     {
-        $this->groups = [$this->userAgent];
-        $this->groups[] = $this->stripVersion();
-        while (mb_strpos(end($this->groups), '-') !== false) {
-            $current = end($this->groups);
-            $this->groups[] = mb_substr($current, 0, mb_strrpos($current, '-'));
+        $groups = [$this->userAgent];
+
+        $groups[] = $this->stripVersion();
+        while (mb_stripos(end($groups), '-') !== false) {
+            $current = end($groups);
+            $groups[] = mb_substr($current, 0, mb_strripos($current, '-'));
+        }
+        foreach ($groups as $group) {
+            if (!in_array($group, $this->groups)) {
+                $this->groups[] = $group;
+            }
         }
-        $this->groups = array_unique($this->groups);
     }
 
     /**
@@ -49,7 +66,7 @@ private function explode()
      */
     public function stripVersion()
     {
-        if (mb_strpos($this->userAgent, '/') !== false) {
+        if (mb_stripos($this->userAgent, '/') !== false) {
             return mb_split('/', $this->userAgent, 2)[0];
         }
         return $this->userAgent;
@@ -64,8 +81,9 @@ public function stripVersion()
      */
     public function match($array)
     {
+        $array = array_map('mb_strtolower', $array);
         foreach ($this->groups as $userAgent) {
-            if (in_array($userAgent, array_map('mb_strtolower', $array))) {
+            if (in_array(mb_strtolower($userAgent), $array)) {
                 return $userAgent;
             }
         }

src/UserAgentParser/Exceptions/FormatException.php

@@ -0,0 +1,11 @@
+<?php
+namespace vipnytt\UserAgentParser\Exceptions;
+
+/**
+ * Class FormatException
+ *
+ * @package vipnytt\UserAgentParser\Exceptions
+ */
+class FormatException extends UserAgentParserException
+{
+}

src/UserAgentParser/Exceptions/UserAgentParserException.php

@@ -0,0 +1,13 @@
+<?php
+namespace vipnytt\UserAgentParser\Exceptions;
+
+use Exception;
+
+/**
+ * Class UserAgentParserException
+ *
+ * @package vipnytt\UserAgentParser\Exceptions
+ */
+class UserAgentParserException extends Exception
+{
+}

tests/FormatTest.php

@@ -0,0 +1,43 @@
+<?php
+namespace vipnytt\UserAgentParser\Tests;
+
+use vipnytt\UserAgentParser;
+use vipnytt\UserAgentParser\Exceptions\FormatException;
+
+/**
+ * Class FormatTest
+ *
+ * @package vipnytt\UserAgentParser\Tests
+ */
+class FormatTest extends \PHPUnit_Framework_TestCase
+{
+    /**
+     * @dataProvider generateDataForTest
+     * @param string $userAgent
+     */
+    public function testFormat($userAgent)
+    {
+        $this->expectException(FormatException::class);
+        new UserAgentParser($userAgent);
+    }
+
+    /**
+     * Generate test data
+     * @return array
+     */
+    public
+    function generateDataForTest()
+    {
+        return [
+            [
+                'mybot 2.0',
+            ],
+            [
+                'my crawler',
+            ],
+            [
+                'Mozilla/5.0 (Windows NT 6.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2228.0 Safari/537.36',
+            ],
+        ];
+    }
+}