<?php
//
// +------------------------------------------------------------------------+
// | phpDocumentor                                                          |
// +------------------------------------------------------------------------+
// | Copyright (c) 2000-2003 Joshua Eichorn, Gregory Beaver                 |
// | Email         jeichorn@phpdoc.org, cellog@phpdoc.org                   |
// | Web           http://www.phpdoc.org                                    |
// | Mirror        http://phpdocu.sourceforge.net/                          |
// | PEAR          http://pear.php.net/package-info.php?pacid=137           |
// +------------------------------------------------------------------------+
// | This source file is subject to version 3.00 of the PHP License,        |
// | that is available at http://www.php.net/license/3_0.txt.               |
// | If you did not receive a copy of the PHP license and are unable to     |
// | obtain it through the world-wide-web, please send a note to            |
// | license@php.net so we can mail you a copy immediately.                 |
// +------------------------------------------------------------------------+
//

/**
 * @author Greg Beaver <cellog@users.sourceforge.net>
 * @package phpDocumentor
 * @subpackage WordParsers
 * @since 1.2
 */
/**
 * Like WordParser, but expects an array of tokens from the tokenizer instead
 * of a string.
 * @author Greg Beaver <cellog@users.sourceforge.net>
 * @package phpDocumentor
 * @subpackage WordParsers
 * @since 1.2
 */
class phpDocumentorTWordParser extends WordParser
{
    /**#@+
     * @access private
     */
    /**
     * tokenized array from {@link token_get_all()}
     * @var array
     */
    var $_all;
    /**
     * List of tokens that can contain a newline
     * @var array
     */
    var $_nl_check = array(
        T_WHITESPACE,
        T_ENCAPSED_AND_WHITESPACE,
        T_COMMENT,
        T_DOC_COMMENT,
        T_OPEN_TAG,
        T_CLOSE_TAG,
        T_INLINE_HTML);
    /**
     * @var array
     */
    var $_global_search;
    /**
     * current source line number (relative)
     * @var integer
     */
    var $_sourceline;
    /**
     * Source of the entire file, parsed into arrays of tokens on each line
     * @var array
     */
    var $_file_source = array();
    /**
     * Line number the last comment was on
     * @var integer
     */
    var $_docblock_linenum;
    /**#@-*/
    
    /**
     * Uses {@link token_get_all()} to tokenize the source code.
     * {@internal
     * Also, it divides the source tokens into separate lines for use by
     * the @filesource tag.
     *
     * {@source}}}
     * @var string source code
     */
    function setup(&$input)
    {
        $input = rtrim(ltrim($input, "\r\n"));
        $this->data = &$input;
        // fix php warnings on invalid source code
        $this->_all = @token_get_all($input);
        $this->_file_source = array();
        $this->addFileSource($this->_all);
        $this->_sourceline = 0;
        $this->pos = 0;
        $this->linenum = 0;
    }
    
    /**
     * @return array
     */
    function getSource()
    {
        $source = $this->source;
        $this->source = array();
        $this->getsource = false;
        return $source;
    }
    
    /**
     * @return array source code tokens split up by line number
     */
    function getFileSource()
    {
        return $this->_file_source;
    }
    
    /**
     * Begin retrieving source code
     * @access private
     * @param string word to add the beginning of source code
     */
    function retrievesource($word = '')
    {
        $this->source = array(array($word));
        $this->_sourceline = 0;
        $this->getsource = true;
    }

    /**
     * Utility function to determine whether two tokens from the tokenizer are equal
     * @static
     */
    function tokenEquals($a, $b)
    {
        if (is_array($a)) $a = $a[1];
        if (is_array($b)) $b = $b[1];
        return $a == $b;
    }
    
    /**
     * Utility function to convert a series of tokens into a string
     * @static
     */
    function concatTokens($a)
    {
        $b = '';
        foreach($a as $c)
        {
            if (is_array($c)) $c = $c[1];
            $b .= $c;
        }
        return $b;
    }
    
    /**
     * Retrieve a token for the phpDocumentorTParser
     * {@internal
     * This method adds source code to the array for a function to be returned
     * to a {@}source} tag, and will return the token unless it is T_WHITESPACE
     * and {@link $returnWhiteSpace} is false.
     *
     * The global variable search is more complicated than it is in the
     * WordParser, as we have to compare an array of tokens to each other, and
     * that is what this code does}}
     * @return string|array token from tokenizer
     */
    function getWord()
    {
        if (!isset($this->_all[$this->pos])) return false;
        $oldlinenum = $this->linenum;
        $word = $this->_all[$this->pos++];
        // if we're looking for a global variable declaration, then this section
        // will search the upcoming tokens to see if they match the tokens
        // that define the global variable
        if (isset($this->_global_search))
        {
            $pos = $this->pos;
            $gpos = 0;
            $found = false;
            if ($this->tokenEquals($word,$this->_global_search[$gpos++]))
            {
                $found = true;
                for(;$gpos<count($this->_global_search);$gpos++,$pos++)
                {
                    if (!$this->tokenEquals($this->_global_search[$gpos],$this->_all[$pos])) $found = false;
                }
            }
            if ($found)
            {
                $a = $this->concatTokens($this->_global_search);
                $this->pos += count($this->_global_search) - 1;
                unset($this->_global_search);
                return $a;
            }
        }
        if ($this->getsource)
        {
            $this->addSource($word);
        }
        if (is_array($word))
        {
            if (in_array($word[0],$this->_nl_check))
            {
                $this->linenum += substr_count($word[1],"\n");
            }
            if ($word[0] == T_WHITESPACE && !$this->returnWhiteSpace) return $this->getWord();
            // seeing if we can get line numbers out of the beast
        }
        if (is_array($word) && $word[0] == T_COMMENT) $this->_docblock_linenum = $oldlinenum;
        return $word;
    }
    
    /**
     * Wrapper for {@link addSource()} used to retrieve the entire source code
     * organized by line number in setup()
     * @param array full file source code
     */
    function addFileSource($word)
    {
        $this->_sourceline = 0;
        foreach($word as $token)
        {
            $this->addSource($token, true);
        }
//        var_dump($this->_file_source);
    }
    
    /**
     * Generate source token arrays organized by line number
     *
     * This code will split up tokens that contain "\n" and add them to the
     * source code as separate tokens on different lines.
     * @param array|string token to add
     * @param boolean true if this should be added to {@link $_file_source}
     * @param array|string next token, for lookahead splitting
     * @uses _set_sars()
     */
    function addSource($word, $file = false)
    {
        if (is_array($word))
        {
            $lines = str_replace("\r", '', explode("\n",$word[1]));
            foreach($lines as $i => $line)
            {
                $this->_set_sars($file, array($word[0],$line));
                if ($i < count($lines) - 1)
                {
                    // increment sourceline
                    $this->_sourceline++;
                }
            }
        } else $this->_set_sars($file, $word);
    }
    
    /**
     * Add tokens to source code
     *
     * {@source}
     * @access private
     * @param boolean true if this is file source, otherwise it is function source
     * @param string|array token to add
     */
    function _set_sars($type,$word)
    {
        if ($type)
        {
            $this->_file_source[$this->_sourceline][] = $word;
        } else
        {
            $this->source[$this->_sourceline][] = $word;
        }
    }
    
    /**
     * Tell the phpDocumentorTWordParser to return the entire global variable
     * if it is found.
     * @uses $_global_search
     * @param array tokens that represent the global variable definition
     */
    function findGlobal($tokens)
    {
        if (!$tokens)
        {
            unset($this->_global_search);
        } else
        $this->_global_search = $tokens;
    }
    
    function backupPos()
    {
        $this->pos--;
        $word = $this->_all[$this->pos];
        if ($this->getsource)
        {
            unset($this->source[$this->_sourceline][count($this->source[$this->_sourceline]) - 1]);
            if (empty($this->source[$this->_sourceline])) unset($this->source[$this->_sourceline]);
            else $this->source[$this->_sourceline] = array_values($this->source[$this->_sourceline]);
        }
        if (is_array($word))
        {
            if ($word[0] == T_WHITESPACE && !$this->returnWhiteSpace) return $this->getWord();
            // seeing if we can get line numbers out of the beast
            if (in_array($word[0],$this->_nl_check))
            {
                $this->linenum -= substr_count($word[1],"\n");
            }
        }
    }
}
?>