2013-05-27 54 views
1

我听说.NET手册是从代码构建的,并且最重要的部分后来被手动改进。从PHP代码构建文档

到目前为止Googleing还没有出现。我发现的大部分内容都与编码和记录最佳实践有关。但没有记录可以帮助以后编写手册,特别是不能做到这一点的解决方案。即使不限制搜索到PHP。

有谁知道一个解决方案,最好使用PHP构建,可以从PHP代码构建类似的文档?

回答

0

既然这看起来很有趣,我建立了这个实践和教育。

我发现它实际上比其他所有东西都更方便,因为它可以让我根据需要构建文档。

我也为它建立了一个HTML生成器,如果有人想要它,我会发布它。

请不要在评论中激进。

<?php 

/**moDO(Classes)(Modo)(Modo_Parse) 

    @(Description) 
    Parses Many One standard code comments and produces an array that can be used to build web manuals. 


    @(Syntax){quote} 
    object `Modo_Parse` (string ~$contents~) 


    @(Parameters) 
    @(Parameters)($contents)(1){info} 
    Contents of the code files that contain code comments to be parsed. 

    @(Parameters)($contents)(2){note warn} 
    Please note very long contents can cause performance issues. 
    It is recommended that you run the parser file by file. 


    @(Return) 
    Returns an object that contains a variable `parsed` that contains the resulting array. 


    @(Examples) 
    @(Examples)(1){info} 
    `Example 1:` Basic usage example. 

    @(Examples)(2){code php} 
    $contents = file_get_contents("Modo_Parse.php"); 
    $parser = new Modo_Parse($contents); 
    print_r($parser->parsed); 


    @(Changelog){list} 
    (1.0) ~Initial release.~ 

*/ 

    /** 
    * Parses Many One standard code comments and produces an array that can be used to build manuals. 
    * @syntax new Modo_Parse($contents); 
    * @contents string containing codes with comments to be parsed 
    */ 
    class Modo_Parse { 
    /** 
    * Takes the parameter and calls the parser function. 
    */ 
    function __construct($contents) { 
     // standardise line breaks 
     $contents = str_replace(Array("\r\n", "\n\r", "\n"), Array("\n", "\n", "\r\n"), $contents); 

     $this->parsed = $this->parser((string)$contents); 
    } 

    /** 
    * The parser function that does all the work. 
    */ 
    private function parser($contents) { 
     $return = Array(); 

     // Identify docs 
     preg_match_all('/(\/\*\*moDO(\([a-z \$_-]+\))+\s+)(.*?)(\s*\r\n*\*\/\r\n)/is', $contents, $docs); 

     foreach($docs[0] AS $doc) { 
     $records = Array(); 

     // Extract and prepare log header 
     $header = explode("\n", trim($doc)); 
     $header = trim(rtrim(str_replace("/**moDO", "", $header[0]))); 
     $header = explode("|", str_replace(Array(")(", "(", ")"), Array("|", "", ""), $header)); 

     // Identify log records 
     preg_match_all('/((@\([a-z0-9 \$_-]+\)+(\{[a-z0-9 _-]+\})?))(.*?)+([email protected]\([a-z0-9 \$_-]+\)|\*\/)/is', $doc, $log_records); 

     foreach($log_records[0] AS $record) { 
      // Separate header and text 
      preg_match("/(@(\([a-z0-9 \$_-]+\))+(\{[a-z0-9 _-]+\})?)(.*)/is", trim($record), $separated); 

      // Extract and prepare record header and style 
      $record_head = str_replace(Array("@", ")(", "(", ")", "{", "}"), Array("", "|", "", "", "~", ""), $separated[1]); 
      $record_head = explode("~", $record_head); 
      if(count($record_head) == 1) { 
      $record_head[] = ""; 
      } 
      $record_head = Array(explode("|", $record_head[0]), $record_head[1]); 

      // Prepare record text 
      if(!strstr($record_head[1], "code")) { 
      $separated[4] = trim(implode("\n", array_map('trim', explode("\n", $separated[4])))); 
      } 
      $separated[4] = preg_replace("/^(\r\n)(.*)/", "$2", rtrim($separated[4])); 

      $record_text = preg_replace(Array("/</", "/>/", "/`(.*?)`/", "/~(.*?)~/"), Array("&lt;", "&gt;", "<b>$1</b>", "<i>$1</i>"), $separated[4]); 

      // Find and set Links 
      preg_match_all("/(\&gt\;)([a-z-.:\/]+)(()([a-z .,;:-_]+))?(\&lt\;)/i", $record_text, $anchors, PREG_SET_ORDER); 
      if(!empty($anchors)) { 
      foreach($anchors AS $anchor) { 
       if(empty($anchor[5])) { 
       $anchor[5] = $anchor[2]; 
       } 
       $record_text = str_replace($anchor[0], '<a href="' . $anchor[2] . '" target="_blank">' . $anchor[5] . '</a>', $record_text); 
      } 
      } 

      // Merge the data together 
      $records = $this->array_merge_recursive_distinct($records, $this->array_chain($record_head[0], Array("style" => $record_head[1], "text" => $record_text))); 
     } 

     $return[] = Array("log" => $header, "data" => $records); 
     } 

     return $return; 
    } 

    /** 
    * Turns a list into a chain of keys with the value in the last key. 
    */ 
    private function array_chain(Array $keys, $value) { 
     $first = array_shift($keys); 

     if (count($keys) === 0) { 
     $return = Array($first => $value); 
     } 
     else { 
     $return = Array($first => $this->array_chain($keys, $value)); 
     } 

     return $return; 
    } 

    /** 
    * Distinct recursive array merge. 
    */ 
    private function array_merge_recursive_distinct(Array &$array1, Array &$array2) { 
     $merged = $array1; 

     foreach($array2 as $key => &$value) { 
     if(is_array($value) && isset($merged[$key]) && is_array($merged[$key])) { 
      $merged[$key] = $this->array_merge_recursive_distinct($merged[$key], $value); 
     } 
     else { 
      $merged[$key] = $value; 
     } 
     } 

     return $merged; 
    } 
    } 
2

我相信phpdoc是您正在寻找的术语:

http://en.wikipedia.org/wiki/PHPDoc

我本人来说使用http://www.phpdoc.org/实施

例如:你定义一个函数

/** 
    * This is the short description for a DocBlock. 
    * 
    * This is the long description for a DocBlock. This text may contain 
    * multiple lines and even some _markdown_. 
    * 
    * 
    * @author Mike van Riel 
    * 
    * @since 1.0 
    * 
    * @param int $example This is an example function/method parameter description. 
    * @param string $example2 This is a second example. 
    * @return int 
    */ 
    function docBlock($example, $example2){ 
     return $example*intval($example2); 
    } 
+0

你能为我提供了一个链接到包括复杂的文档和示例:瓦片说明,语法,参数,返回,更新日志,笔记,例子吗? – transilvlad

+0

也许在http://www.phpdoc.org/docs/latest/for-users/phpdoc-reference.html文件的前几行将给你主要的想法是如何工作的。 – ulkas

+0

我得到了一般想法,但没有深入的想法:)我一直在寻找他们的网站,但没有找到一个与使用示例,更新日志,笔记,警告等复杂的例子。会产生这样的事情:http:// php.net/manual/en/function.mail.php – transilvlad

0

之前把这个PhpDocumentorApiGen是很好去。然而,PHP.net可能会使用一个特殊的命令,仅仅为他们(或者从数据库中获取额外的信息,比如更新日志,以及用户示例和投票)。

非常重要的是,请记住,没有文档标准,并且每个解决方案都可以根据自己的需求或愿景实施自己的解决方案。幸运的是,他们都试图像Javadoc一样做。

我已经Google了一下,没有看到Javadoc的@changelog标签,所以也许唯一的办法就是在每个docblock中做自己的结构,然后在生成文档后用一些样式解析它。