# Cpdf

A PHP class to provide the basic functionality to create a pdf document without any requirement for additional modules.

## Defined (1)

The class is defined in the following location(s).

/lib/dompdf/lib/class.pdf.php
`class Cpdf {    /**    * @var integer The current number of pdf objects in the document    */   public \$numObj = 0;    /**    * @var array This array contains all of the pdf objects, ready for final assembly    */   public \$objects = array();    /**    * @var integer The objectId (number within the objects array) of the document catalog    */   public \$catalogId;    /**    * @var array Array carrying information about the fonts that the system currently knows about    * Used to ensure that a font is not loaded twice, among other things    */   public \$fonts = array();    /**    * @var string The default font metrics file to use if no other font has been loaded.    * The path to the directory containing the font metrics should be included    */   public \$defaultFont = './fonts/Helvetica.afm';    /**    * @string A record of the current font    */   public \$currentFont = '';    /**    * @var string The current base font    */   public \$currentBaseFont = '';    /**    * @var integer The number of the current font within the font array    */   public \$currentFontNum = 0;    /**    * @var integer    */   public \$currentNode;    /**    * @var integer Object number of the current page    */   public \$currentPage;    /**    * @var integer Object number of the currently active contents block    */   public \$currentContents;    /**    * @var integer Number of fonts within the system    */   public \$numFonts = 0;    /**    * @var integer Number of graphic state resources used    */   private \$numStates = 0;    /**    * @var array Current color for fill operations, defaults to inactive value,     * all three components should be between 0 and 1 inclusive when active    */   public \$currentColor = null;    /**    * @var array Current color for stroke operations (lines etc.)    */   public \$currentStrokeColor = null;    /**    * @var string Current style that lines are drawn in    */   public \$currentLineStyle = '';    /**    * @var array Current line transparency (partial graphics state)    */   public \$currentLineTransparency = array("mode" => "Normal", "opacity" => 1.0);    /**    * array Current fill transparency (partial graphics state)    */   public \$currentFillTransparency = array("mode" => "Normal", "opacity" => 1.0);    /**    * @var array An array which is used to save the state of the document, mainly the colors and styles    * it is used to temporarily change to another state, the change back to what it was before    */   public \$stateStack = array();    /**    * @var integer Number of elements within the state stack    */   public \$nStateStack = 0;    /**    * @var integer Number of page objects within the document    */   public \$numPages = 0;    /**    * @var array Object Id storage stack    */   public \$stack = array();    /**    * @var integer Number of elements within the object Id storage stack    */   public \$nStack = 0;    /**    * an array which contains information about the objects which are not firmly attached to pages    * these have been added with the addObject function    */   public \$looseObjects = array();    /**    * array contains infomation about how the loose objects are to be added to the document    */   public \$addLooseObjects = array();    /**    * @var integer The objectId of the information object for the document    * this contains authorship, title etc.    */   public \$infoObject = 0;    /**    * @var integer Number of images being tracked within the document    */   public \$numImages = 0;    /**    * @var array An array containing options about the document    * it defaults to turning on the compression of the objects    */   public \$options = array('compression' => true);    /**    * @var integer The objectId of the first page of the document    */   public \$firstPageId;    /**    * @var float Used to track the last used value of the inter-word spacing, this is so that it is known    * when the spacing is changed.    */   public \$wordSpaceAdjust = 0;    /**    * @var float Used to track the last used value of the inter-letter spacing, this is so that it is known    * when the spacing is changed.    */   public \$charSpaceAdjust = 0;    /**    * @var integer The object Id of the procset object    */   public \$procsetObjectId;    /**    * @var array Store the information about the relationship between font families    * this used so that the code knows which font is the bold version of another font, etc.    * the value of this array is initialised in the constuctor function.    */   public \$fontFamilies = array();    /**    * @var string Folder for php serialized formats of font metrics files.    * If empty string, use same folder as original metrics files.    * This can be passed in from class creator.    * If this folder does not exist or is not writable, Cpdf will be **much** slower.    * Because of potential trouble with php safe mode, folder cannot be created at runtime.    */   public \$fontcache = '';    /**    * @var integer The version of the font metrics cache file.    * This value must be manually incremented whenever the internal font data structure is modified.    */   public \$fontcacheVersion = 6;    /**    * @var string Temporary folder.    * If empty string, will attempty system tmp folder.    * This can be passed in from class creator.    * Only used for conversion of gd images to jpeg images.    */   public \$tmp = '';    /**    * @var string Track if the current font is bolded or italicised    */   public \$currentTextState = '';    /**    * @var string Messages are stored here during processing, these can be selected afterwards to give some useful debug information    */   public \$messages = '';    /**    * @var string The ancryption array for the document encryption is stored here    */   public \$arc4 = '';    /**    * @var integer The object Id of the encryption information    */   public \$arc4_objnum = 0;    /**    * @var string The file identifier, used to uniquely identify a pdf document    */   public \$fileIdentifier = '';    /**    * @var boolean A flag to say if a document is to be encrypted or not    */   public \$encrypted = false;    /**    * @var string The encryption key for the encryption of all the document content (structure is not encrypted)    */   public \$encryptionKey = '';    /**    * @var array Array which forms a stack to keep track of nested callback functions    */   public \$callback = array();    /**    * @var integer The number of callback functions in the callback array    */   public \$nCallback = 0;    /**    * @var array Store label->id pairs for named destinations, these will be used to replace internal links    * done this way so that destinations can be defined after the location that links to them    */   public \$destinations = array();    /**    * @var array Store the stack for the transaction commands, each item in here is a record of the values of all the    * publiciables within the class, so that the user can rollback at will (from each 'start' command)    * note that this includes the objects array, so these can be large.    */   public \$checkpoint = '';    /**    * @var array Table of Image origin filenames and image labels which were already added with o_image().    * Allows to merge identical images    */   public \$imagelist = array();    /**    * @var boolean Whether the text passed in should be treated as Unicode or just local character set.    */   public \$isUnicode = false;    /**    * @var string the JavaScript code of the document    */   public \$javascript = '';    /**    * @var boolean whether the compression is possible    */   protected \$compressionReady = false;    /**    * @var array Current page size    */   protected \$currentPageSize = array("width" => 0, "height" => 0);    /**    * @var array All the chars that will be required in the font subsets    */   protected \$stringSubsets = array();    /**    * @var string The target internal encoding    */   static protected \$targetEncoding = 'iso-8859-1';    /**    * @var array The list of the core fonts    */   static protected \$coreFonts = array(     'courier', 'courier-bold', 'courier-oblique', 'courier-boldoblique',      'helvetica', 'helvetica-bold', 'helvetica-oblique', 'helvetica-boldoblique',      'times-roman', 'times-bold', 'times-italic', 'times-bolditalic',      'symbol', 'zapfdingbats'  );    /**    * Class constructor    * This will start a new document    *    * @param array   \$pageSize  Array of 4 numbers, defining the bottom left and upper right corner of the page. first two are normally zero.    * @param boolean \$isUnicode Whether text will be treated as Unicode or not.    * @param string  \$fontcache The font cache folder    * @param string  \$tmp       The temporary folder    */   function __construct(\$pageSize = array(0, 0, 612, 792), \$isUnicode = false, \$fontcache = '', \$tmp = '') {     \$this->isUnicode = \$isUnicode;     \$this->fontcache = \$fontcache;     \$this->tmp = \$tmp;     \$this->newDocument(\$pageSize);      \$this->compressionReady = function_exists('gzcompress');          if ( in_array('Windows-1252', mb_list_encodings()) ) {       self::\$targetEncoding = 'Windows-1252';     }      // also initialize the font families that are known about already     \$this->setFontFamily('init');     //  \$this->fileIdentifier = md5('xxxxxxxx'.time());   }    /**    * Document object methods (internal use only)    *    * There is about one object method for each type of object in the pdf document    * Each function has the same call list (\$id, \$action, \$options).    * \$id = the object ID of the object, or what it is to be if it is being created    * \$action = a string specifying the action to be performed, though ALL must support:    *           'new' - create the object with the id \$id    *           'out' - produce the output for the pdf object    * \$options = optional, a string or array containing the various parameters for the object    *    * These, in conjunction with the output function are the ONLY way for output to be produced    * within the pdf 'file'.    */    /**    * Destination object, used to specify the location for the user to jump to, presently on opening    */   protected function o_destination(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = &\$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'destination', 'info' => array());       \$tmp = '';       switch (\$options['type']) {         case 'XYZ':         case 'FitR':           \$tmp = ' '.\$options['p3'].\$tmp;         case 'FitH':         case 'FitV':         case 'FitBH':         case 'FitBV':           \$tmp = ' '.\$options['p1'].' '.\$options['p2'].\$tmp;         case 'Fit':         case 'FitB':           \$tmp = \$options['type'].\$tmp;           \$this->objects[\$id]['info']['string'] = \$tmp;           \$this->objects[\$id]['info']['page'] = \$options['page'];       }       break;      case 'out':       \$tmp = \$o['info'];       \$res = "\n\$id 0 obj\n".'['.\$tmp['page'].' 0 R /'.\$tmp['string']."]\nendobj";       return \$res;     }   }    /**    * set the viewer preferences    */   protected function o_viewerPreferences(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'viewerPreferences', 'info' => array());       break;      case 'add':       foreach (\$options as \$k => \$v) {         switch (\$k) {           case 'HideToolbar':           case 'HideMenubar':           case 'HideWindowUI':           case 'FitWindow':           case 'CenterWindow':           case 'NonFullScreenPageMode':           case 'Direction':             \$o['info'][\$k] = \$v;             break;         }       }       break;      case 'out':       \$res = "\n\$id 0 obj\n<< ";       foreach (\$o['info'] as \$k => \$v) {         \$res.= "\n/\$k \$v";       }       \$res.= "\n>>\n";       return \$res;     }   }    /**    * define the document catalog, the overall controller for the document    */   protected function o_catalog(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         \$this->objects[\$id] = array('t' => 'catalog', 'info' => array());         \$this->catalogId = \$id;         break;        case 'outlines':       case 'pages':       case 'openHere':       case 'javascript':         \$o['info'][\$action] = \$options;         break;        case 'viewerPreferences':         if (!isset(\$o['info']['viewerPreferences'])) {           \$this->numObj++;           \$this->o_viewerPreferences(\$this->numObj, 'new');           \$o['info']['viewerPreferences'] = \$this->numObj;         }          \$vp = \$o['info']['viewerPreferences'];         \$this->o_viewerPreferences(\$vp, 'add', \$options);          break;        case 'out':         \$res = "\n\$id 0 obj\n<< /Type /Catalog";          foreach (\$o['info'] as \$k => \$v) {           switch (\$k) {             case 'outlines':               \$res.= "\n/Outlines \$v 0 R";               break;              case 'pages':               \$res.= "\n/Pages \$v 0 R";               break;              case 'viewerPreferences':               \$res.= "\n/ViewerPreferences \$v 0 R";               break;              case 'openHere':               \$res.= "\n/OpenAction \$v 0 R";               break;              case 'javascript':               \$res.= "\n/Names <</JavaScript \$v 0 R>>";               break;           }         }          \$res.= " >>\nendobj";         return \$res;     }   }    /**    * object which is a parent to the pages in the document    */   protected function o_pages(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'pages', 'info' => array());       \$this->o_catalog(\$this->catalogId, 'pages', \$id);       break;      case 'page':       if (!is_array(\$options)) {         // then it will just be the id of the new page         \$o['info']['pages'][] = \$options;       }       else {         // then it should be an array having 'id', 'rid', 'pos', where rid=the page to which this one will be placed relative         // and pos is either 'before' or 'after', saying where this page will fit.         if (isset(\$options['id']) && isset(\$options['rid']) && isset(\$options['pos'])) {           \$i = array_search(\$options['rid'], \$o['info']['pages']);           if (isset(\$o['info']['pages'][\$i]) && \$o['info']['pages'][\$i] == \$options['rid']) {              // then there is a match             // make a space             switch (\$options['pos']) {               case 'before':                 \$k = \$i;                 break;                case 'after':                 \$k = \$i+1;                 break;                default:                 \$k = -1;                 break;             }              if (\$k >= 0) {               for (\$j = count(\$o['info']['pages'])-1; \$j >= \$k; \$j--) {                 \$o['info']['pages'][\$j+1] = \$o['info']['pages'][\$j];               }                \$o['info']['pages'][\$k] = \$options['id'];             }           }         }       }       break;      case 'procset':       \$o['info']['procset'] = \$options;       break;      case 'mediaBox':       \$o['info']['mediaBox'] = \$options;       // which should be an array of 4 numbers       \$this->currentPageSize = array('width' => \$options[2], 'height' => \$options[3]);       break;      case 'font':       \$o['info']['fonts'][] = array('objNum' => \$options['objNum'], 'fontNum' => \$options['fontNum']);       break;      case 'extGState':       \$o['info']['extGStates'][] = array('objNum' => \$options['objNum'], 'stateNum' => \$options['stateNum']);       break;      case 'xObject':       \$o['info']['xObjects'][] = array('objNum' => \$options['objNum'], 'label' => \$options['label']);       break;      case 'out':       if (count(\$o['info']['pages'])) {         \$res = "\n\$id 0 obj\n<< /Type /Pages\n/Kids [";         foreach (\$o['info']['pages'] as \$v) {           \$res.= "\$v 0 R\n";         }          \$res.= "]\n/Count ".count(\$this->objects[\$id]['info']['pages']);          if ( (isset(\$o['info']['fonts']) && count(\$o['info']['fonts'])) ||               isset(\$o['info']['procset']) ||               (isset(\$o['info']['extGStates']) && count(\$o['info']['extGStates']))) {           \$res.= "\n/Resources <<";            if (isset(\$o['info']['procset'])) {             \$res.= "\n/ProcSet ".\$o['info']['procset']." 0 R";           }            if (isset(\$o['info']['fonts']) && count(\$o['info']['fonts'])) {             \$res.= "\n/Font << ";             foreach (\$o['info']['fonts'] as \$finfo) {               \$res.= "\n/F".\$finfo['fontNum']." ".\$finfo['objNum']." 0 R";             }             \$res.= "\n>>";           }            if (isset(\$o['info']['xObjects']) && count(\$o['info']['xObjects'])) {             \$res.= "\n/XObject << ";             foreach (\$o['info']['xObjects'] as \$finfo) {               \$res.= "\n/".\$finfo['label']." ".\$finfo['objNum']." 0 R";             }             \$res.= "\n>>";           }            if ( isset(\$o['info']['extGStates']) && count(\$o['info']['extGStates'])) {             \$res.= "\n/ExtGState << ";             foreach (\$o['info']['extGStates'] as \$gstate) {               \$res.= "\n/GS" . \$gstate['stateNum'] . " " . \$gstate['objNum'] . " 0 R";             }             \$res.= "\n>>";           }            \$res.= "\n>>";           if (isset(\$o['info']['mediaBox'])) {             \$tmp = \$o['info']['mediaBox'];             \$res.= "\n/MediaBox [".sprintf('%.3F %.3F %.3F %.3F', \$tmp[0], \$tmp[1], \$tmp[2], \$tmp[3]) .']';           }         }          \$res.= "\n >>\nendobj";       }       else {         \$res = "\n\$id 0 obj\n<< /Type /Pages\n/Count 0\n>>\nendobj";       }        return \$res;     }   }    /**    * define the outlines in the doc, empty for now    */   protected function o_outlines(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = &\$this->objects[\$id];     }      switch (\$action) {       case 'new':         \$this->objects[\$id] = array('t' => 'outlines', 'info' => array('outlines' => array()));         \$this->o_catalog(\$this->catalogId, 'outlines', \$id);         break;        case 'outline':         \$o['info']['outlines'][] = \$options;         break;        case 'out':         if (count(\$o['info']['outlines'])) {           \$res = "\n\$id 0 obj\n<< /Type /Outlines /Kids [";           foreach (\$o['info']['outlines'] as \$v) {             \$res.= "\$v 0 R ";           }            \$res.= "] /Count ".count(\$o['info']['outlines']) ." >>\nendobj";         } else {           \$res = "\n\$id 0 obj\n<< /Type /Outlines /Count 0 >>\nendobj";         }          return \$res;     }   }    /**    * an object to hold the font description    */   protected function o_font(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = &\$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'font', 'info' => array('name' => \$options['name'], 'fontFileName' => \$options['fontFileName'], 'SubType' => 'Type1'));       \$fontNum = \$this->numFonts;       \$this->objects[\$id]['info']['fontNum'] = \$fontNum;        // deal with the encoding and the differences       if (isset(\$options['differences'])) {         // then we'll need an encoding dictionary         \$this->numObj++;         \$this->o_fontEncoding(\$this->numObj, 'new', \$options);         \$this->objects[\$id]['info']['encodingDictionary'] = \$this->numObj;       }       else if (isset(\$options['encoding'])) {         // we can specify encoding here         switch (\$options['encoding']) {           case 'WinAnsiEncoding':           case 'MacRomanEncoding':           case 'MacExpertEncoding':             \$this->objects[\$id]['info']['encoding'] = \$options['encoding'];             break;            case 'none':             break;            default:             \$this->objects[\$id]['info']['encoding'] = 'WinAnsiEncoding';             break;         }       }       else {         \$this->objects[\$id]['info']['encoding'] = 'WinAnsiEncoding';       }        if (\$this->fonts[\$options['fontFileName']]['isUnicode']) {         // For Unicode fonts, we need to incorporate font data into         // sub-sections that are linked from the primary font section.         // Look at o_fontGIDtoCID and o_fontDescendentCID functions         // for more informaiton.         //         // All of this code is adapted from the excellent changes made to         // transform FPDF to TCPDF (http://tcpdf.sourceforge.net/)          \$toUnicodeId = ++\$this->numObj;         \$this->o_contents(\$toUnicodeId, 'new', 'raw');         \$this->objects[\$id]['info']['toUnicode'] = \$toUnicodeId;          \$stream = <<<EOT /CIDInit /ProcSet findresource begin 12 dict begin begincmap /CIDSystemInfo <</Registry (Adobe) /Ordering (UCS) /Supplement 0 >> def /CMapName /Adobe-Identity-UCS def /CMapType 2 def 1 begincodespacerange <0000> <FFFF> endcodespacerange 1 beginbfrange <0000> <FFFF> <0000> endbfrange endcmap CMapName currentdict /CMap defineresource pop end end EOT;          \$res = "<</Length " . mb_strlen(\$stream, '8bit') . " >>\n";         \$res .= "stream\n" . \$stream . "endstream";          \$this->objects[\$toUnicodeId]['c'] = \$res;          \$cidFontId = ++\$this->numObj;         \$this->o_fontDescendentCID(\$cidFontId, 'new', \$options);         \$this->objects[\$id]['info']['cidFont'] = \$cidFontId;       }        // also tell the pages node about the new font       \$this->o_pages(\$this->currentNode, 'font', array('fontNum' => \$fontNum, 'objNum' => \$id));       break;      case 'add':       foreach (\$options as \$k => \$v) {         switch (\$k) {           case 'BaseFont':             \$o['info']['name'] = \$v;             break;           case 'FirstChar':           case 'LastChar':           case 'Widths':           case 'FontDescriptor':           case 'SubType':             \$this->addMessage('o_font '.\$k." : ".\$v);             \$o['info'][\$k] = \$v;             break;         }       }        // pass values down to descendent font       if (isset(\$o['info']['cidFont'])) {         \$this->o_fontDescendentCID(\$o['info']['cidFont'], 'add', \$options);       }       break;      case 'out':       if (\$this->fonts[\$this->objects[\$id]['info']['fontFileName']]['isUnicode']) {         // For Unicode fonts, we need to incorporate font data into         // sub-sections that are linked from the primary font section.         // Look at o_fontGIDtoCID and o_fontDescendentCID functions         // for more informaiton.         //         // All of this code is adapted from the excellent changes made to         // transform FPDF to TCPDF (http://tcpdf.sourceforge.net/)          \$res = "\n\$id 0 obj\n<</Type /Font\n/Subtype /Type0\n";         \$res.= "/BaseFont /".\$o['info']['name']."\n";          // The horizontal identity mapping for 2-byte CIDs; may be used         // with CIDFonts using any Registry, Ordering, and Supplement values.         \$res.= "/Encoding /Identity-H\n";         \$res.= "/DescendantFonts [".\$o['info']['cidFont']." 0 R]\n";         \$res.= "/ToUnicode ".\$o['info']['toUnicode']." 0 R\n";         \$res.= ">>\n";         \$res.= "endobj";       } else {         \$res = "\n\$id 0 obj\n<< /Type /Font\n/Subtype /".\$o['info']['SubType']."\n";         \$res.= "/Name /F".\$o['info']['fontNum']."\n";         \$res.= "/BaseFont /".\$o['info']['name']."\n";          if (isset(\$o['info']['encodingDictionary'])) {           // then place a reference to the dictionary           \$res.= "/Encoding ".\$o['info']['encodingDictionary']." 0 R\n";         } else if (isset(\$o['info']['encoding'])) {           // use the specified encoding           \$res.= "/Encoding /".\$o['info']['encoding']."\n";         }          if (isset(\$o['info']['FirstChar'])) {           \$res.= "/FirstChar ".\$o['info']['FirstChar']."\n";         }          if (isset(\$o['info']['LastChar'])) {           \$res.= "/LastChar ".\$o['info']['LastChar']."\n";         }          if (isset(\$o['info']['Widths'])) {           \$res.= "/Widths ".\$o['info']['Widths']." 0 R\n";         }          if (isset(\$o['info']['FontDescriptor'])) {           \$res.= "/FontDescriptor ".\$o['info']['FontDescriptor']." 0 R\n";         }          \$res.= ">>\n";         \$res.= "endobj";       }        return \$res;     }   }    /**    * a font descriptor, needed for including additional fonts    */   protected function o_fontDescriptor(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         \$this->objects[\$id] = array('t' => 'fontDescriptor', 'info' => \$options);         break;        case 'out':         \$res = "\n\$id 0 obj\n<< /Type /FontDescriptor\n";         foreach (\$o['info'] as \$label => \$value) {           switch (\$label) {           case 'Ascent':           case 'CapHeight':           case 'Descent':           case 'Flags':           case 'ItalicAngle':           case 'StemV':           case 'AvgWidth':           case 'Leading':           case 'MaxWidth':           case 'MissingWidth':           case 'StemH':           case 'XHeight':           case 'CharSet':             if (mb_strlen(\$value, '8bit')) {               \$res.= "/\$label \$value\n";             }              break;           case 'FontFile':           case 'FontFile2':           case 'FontFile3':             \$res.= "/\$label \$value 0 R\n";             break;            case 'FontBBox':             \$res.= "/\$label [\$value[0] \$value[1] \$value[2] \$value[3]]\n";             break;            case 'FontName':             \$res.= "/\$label /\$value\n";             break;           }         }          \$res.= ">>\nendobj";          return \$res;     }   }    /**    * the font encoding    */   protected function o_fontEncoding(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       // the options array should contain 'differences' and maybe 'encoding'       \$this->objects[\$id] = array('t' => 'fontEncoding', 'info' => \$options);       break;      case 'out':       \$res = "\n\$id 0 obj\n<< /Type /Encoding\n";       if (!isset(\$o['info']['encoding'])) {         \$o['info']['encoding'] = 'WinAnsiEncoding';       }        if (\$o['info']['encoding'] !== 'none') {         \$res.= "/BaseEncoding /".\$o['info']['encoding']."\n";       }        \$res.= "/Differences \n[";        \$onum = -100;        foreach (\$o['info']['differences'] as \$num => \$label) {         if (\$num != \$onum+1) {           // we cannot make use of consecutive numbering           \$res.= "\n\$num /\$label";         } else {           \$res.= " /\$label";         }          \$onum = \$num;       }        \$res.= "\n]\n>>\nendobj";       return \$res;     }   }    /**    * a descendent cid font, needed for unicode fonts    */   protected function o_fontDescendentCID(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'fontDescendentCID', 'info' => \$options);        // we need a CID system info section       \$cidSystemInfoId = ++\$this->numObj;       \$this->o_contents(\$cidSystemInfoId, 'new', 'raw');       \$this->objects[\$id]['info']['cidSystemInfo'] = \$cidSystemInfoId;       \$res = "<</Registry (Adobe)\n"; // A string identifying an issuer of character collections       \$res.= "/Ordering (UCS)\n"; // A string that uniquely names a character collection issued by a specific registry       \$res.= "/Supplement 0\n"; // The supplement number of the character collection.       \$res.= ">>";       \$this->objects[\$cidSystemInfoId]['c'] = \$res;        // and a CID to GID map       \$cidToGidMapId = ++\$this->numObj;       \$this->o_fontGIDtoCIDMap(\$cidToGidMapId, 'new', \$options);       \$this->objects[\$id]['info']['cidToGidMap'] = \$cidToGidMapId;       break;      case 'add':       foreach (\$options as \$k => \$v) {         switch (\$k) {         case 'BaseFont':           \$o['info']['name'] = \$v;           break;          case 'FirstChar':         case 'LastChar':         case 'MissingWidth':         case 'FontDescriptor':         case 'SubType':           \$this->addMessage("o_fontDescendentCID \$k : \$v");           \$o['info'][\$k] = \$v;           break;         }       }        // pass values down to cid to gid map       \$this->o_fontGIDtoCIDMap(\$o['info']['cidToGidMap'], 'add', \$options);       break;      case 'out':       \$res = "\n\$id 0 obj\n";       \$res.= "<</Type /Font\n";       \$res.= "/Subtype /CIDFontType2\n";       \$res.= "/BaseFont /".\$o['info']['name']."\n";       \$res.= "/CIDSystemInfo ".\$o['info']['cidSystemInfo']." 0 R\n"; //      if (isset(\$o['info']['FirstChar'])) { //        \$res.= "/FirstChar ".\$o['info']['FirstChar']."\n"; //      }  //      if (isset(\$o['info']['LastChar'])) { //        \$res.= "/LastChar ".\$o['info']['LastChar']."\n"; //      }       if (isset(\$o['info']['FontDescriptor'])) {         \$res.= "/FontDescriptor ".\$o['info']['FontDescriptor']." 0 R\n";       }        if (isset(\$o['info']['MissingWidth'])) {         \$res.= "/DW ".\$o['info']['MissingWidth']."\n";       }        if (isset(\$o['info']['fontFileName']) && isset(\$this->fonts[\$o['info']['fontFileName']]['CIDWidths'])) {         \$cid_widths = &\$this->fonts[\$o['info']['fontFileName']]['CIDWidths'];         \$w = '';         foreach (\$cid_widths as \$cid => \$width) {           \$w .= "\$cid [\$width] ";         }         \$res.= "/W [\$w]\n";       }        \$res.= "/CIDToGIDMap ".\$o['info']['cidToGidMap']." 0 R\n";       \$res.= ">>\n";       \$res.= "endobj";        return \$res;     }   }    /**    * a font glyph to character map, needed for unicode fonts    */   protected function o_fontGIDtoCIDMap(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'fontGIDtoCIDMap', 'info' => \$options);       break;      case 'out':       \$res = "\n\$id 0 obj\n";       \$fontFileName = \$o['info']['fontFileName'];       \$tmp = \$this->fonts[\$fontFileName]['CIDtoGID'] = base64_decode(\$this->fonts[\$fontFileName]['CIDtoGID']);        \$compressed = isset(\$this->fonts[\$fontFileName]['CIDtoGID_Compressed']) &&                     \$this->fonts[\$fontFileName]['CIDtoGID_Compressed'];        if (!\$compressed && isset(\$o['raw'])) {         \$res.= \$tmp;       } else {         \$res.= "<<";          if (!\$compressed && \$this->compressionReady && \$this->options['compression']) {           // then implement ZLIB based compression on this content stream           \$compressed = true;           \$tmp = gzcompress(\$tmp, 6);         }         if (\$compressed) {           \$res.= "\n/Filter /FlateDecode";         }          \$res.= "\n/Length ".mb_strlen(\$tmp, '8bit') .">>\nstream\n\$tmp\nendstream";       }        \$res.= "\nendobj";       return \$res;     }   }    /**    * the document procset, solves some problems with printing to old PS printers    */   protected function o_procset(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         \$this->objects[\$id] = array('t' => 'procset', 'info' => array('PDF' => 1, 'Text' => 1));         \$this->o_pages(\$this->currentNode, 'procset', \$id);         \$this->procsetObjectId = \$id;         break;        case 'add':         // this is to add new items to the procset list, despite the fact that this is considered         // obselete, the items are required for printing to some postscript printers         switch (\$options) {         case 'ImageB':         case 'ImageC':         case 'ImageI':           \$o['info'][\$options] = 1;           break;         }         break;        case 'out':         \$res = "\n\$id 0 obj\n[";         foreach (\$o['info'] as \$label => \$val) {           \$res.= "/\$label ";         }         \$res.= "]\nendobj";         return \$res;     }   }    /**    * define the document information    */   protected function o_info(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         \$this->infoObject = \$id;         \$date = 'D:'.@date('Ymd');         \$this->objects[\$id] = array('t' => 'info', 'info' => array('Creator' => 'R and OS php pdf writer, http://www.ros.co.nz', 'CreationDate' => \$date));         break;       case 'Title':       case 'Author':       case 'Subject':       case 'Keywords':       case 'Creator':       case 'Producer':       case 'CreationDate':       case 'ModDate':       case 'Trapped':         \$o['info'][\$action] = \$options;         break;        case 'out':         if (\$this->encrypted) {           \$this->encryptInit(\$id);         }          \$res = "\n\$id 0 obj\n<<\n";         foreach (\$o['info'] as \$k => \$v) {           \$res.= "/\$k (";            if (\$this->encrypted) {             \$v = \$this->ARC4(\$v);           }            // dates must be outputted as-is, without Unicode transformations           elseif (!in_array(\$k, array('CreationDate', 'ModDate'))) {             \$v = \$this->filterText(\$v);           }            \$res.= \$v;           \$res.= ")\n";         }          \$res.= ">>\nendobj";         return \$res;     }   }    /**    * an action object, used to link to URLS initially    */   protected function o_action(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         if (is_array(\$options)) {           \$this->objects[\$id] = array('t' => 'action', 'info' => \$options, 'type' => \$options['type']);         } else {           // then assume a URI action           \$this->objects[\$id] = array('t' => 'action', 'info' => \$options, 'type' => 'URI');         }         break;        case 'out':         if (\$this->encrypted) {           \$this->encryptInit(\$id);         }          \$res = "\n\$id 0 obj\n<< /Type /Action";         switch (\$o['type']) {           case 'ilink':             if (!isset(\$this->destinations[(string)\$o['info']['label']])) break;              // there will be an 'label' setting, this is the name of the destination             \$res.= "\n/S /GoTo\n/D ".\$this->destinations[(string)\$o['info']['label']]." 0 R";             break;            case 'URI':             \$res.= "\n/S /URI\n/URI (";             if (\$this->encrypted) {               \$res.= \$this->filterText(\$this->ARC4(\$o['info']), true, false);             } else {               \$res.= \$this->filterText(\$o['info'], true, false);             }              \$res.= ")";             break;         }          \$res.= "\n>>\nendobj";         return \$res;     }   }    /**    * an annotation object, this will add an annotation to the current page.    * initially will support just link annotations    */   protected function o_annotation(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {       case 'new':         // add the annotation to the current page         \$pageId = \$this->currentPage;         \$this->o_page(\$pageId, 'annot', \$id);          // and add the action object which is going to be required         switch (\$options['type']) {           case 'link':             \$this->objects[\$id] = array('t' => 'annotation', 'info' => \$options);             \$this->numObj++;             \$this->o_action(\$this->numObj, 'new', \$options['url']);             \$this->objects[\$id]['info']['actionId'] = \$this->numObj;             break;            case 'ilink':             // this is to a named internal link             \$label = \$options['label'];             \$this->objects[\$id] = array('t' => 'annotation', 'info' => \$options);             \$this->numObj++;             \$this->o_action(\$this->numObj, 'new', array('type' => 'ilink', 'label' => \$label));             \$this->objects[\$id]['info']['actionId'] = \$this->numObj;             break;         }         break;        case 'out':         \$res = "\n\$id 0 obj\n<< /Type /Annot";         switch (\$o['info']['type']) {           case 'link':           case 'ilink':             \$res.= "\n/Subtype /Link";             break;         }         \$res.= "\n/A ".\$o['info']['actionId']." 0 R";         \$res.= "\n/Border [0 0 0]";         \$res.= "\n/H /I";         \$res.= "\n/Rect [ ";          foreach (\$o['info']['rect'] as \$v) {           \$res.= sprintf("%.4F ", \$v);         }          \$res.= "]";         \$res.= "\n>>\nendobj";         return \$res;     }   }    /**    * a page object, it also creates a contents object to hold its contents    */   protected function o_page(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->numPages++;       \$this->objects[\$id] = array('t' => 'page', 'info' => array('parent' => \$this->currentNode, 'pageNum' => \$this->numPages));        if (is_array(\$options)) {         // then this must be a page insertion, array should contain 'rid', 'pos'=[before|after]         \$options['id'] = \$id;         \$this->o_pages(\$this->currentNode, 'page', \$options);       } else {         \$this->o_pages(\$this->currentNode, 'page', \$id);       }        \$this->currentPage = \$id;       //make a contents object to go with this page       \$this->numObj++;       \$this->o_contents(\$this->numObj, 'new', \$id);       \$this->currentContents = \$this->numObj;       \$this->objects[\$id]['info']['contents'] = array();       \$this->objects[\$id]['info']['contents'][] = \$this->numObj;        \$match = (\$this->numPages%2 ? 'odd' : 'even');       foreach (\$this->addLooseObjects as \$oId => \$target) {         if (\$target === 'all' || \$match === \$target) {           \$this->objects[\$id]['info']['contents'][] = \$oId;         }       }       break;      case 'content':       \$o['info']['contents'][] = \$options;       break;      case 'annot':       // add an annotation to this page       if (!isset(\$o['info']['annot'])) {         \$o['info']['annot'] = array();       }        // \$options should contain the id of the annotation dictionary       \$o['info']['annot'][] = \$options;       break;      case 'out':       \$res = "\n\$id 0 obj\n<< /Type /Page";       \$res.= "\n/Parent ".\$o['info']['parent']." 0 R";        if (isset(\$o['info']['annot'])) {         \$res.= "\n/Annots [";         foreach (\$o['info']['annot'] as \$aId) {           \$res.= " \$aId 0 R";         }         \$res.= " ]";       }        \$count = count(\$o['info']['contents']);       if (\$count == 1) {         \$res.= "\n/Contents ".\$o['info']['contents'][0]." 0 R";       } else if (\$count > 1) {         \$res.= "\n/Contents [\n";          // reverse the page contents so added objects are below normal content         //foreach (array_reverse(\$o['info']['contents']) as \$cId) {         // Back to normal now that I've got transparency working --Benj         foreach (\$o['info']['contents'] as \$cId) {           \$res.= "\$cId 0 R\n";         }         \$res.= "]";       }        \$res.= "\n>>\nendobj";       return \$res;     }   }    /**    * the contents objects hold all of the content which appears on pages    */   protected function o_contents(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'contents', 'c' => '', 'info' => array());       if (mb_strlen(\$options, '8bit') && intval(\$options)) {         // then this contents is the primary for a page         \$this->objects[\$id]['onPage'] = \$options;       } else if (\$options === 'raw') {         // then this page contains some other type of system object         \$this->objects[\$id]['raw'] = 1;       }       break;      case 'add':       // add more options to the decleration       foreach (\$options as \$k => \$v) {         \$o['info'][\$k] = \$v;       }      case 'out':       \$tmp = \$o['c'];       \$res = "\n\$id 0 obj\n";        if (isset(\$this->objects[\$id]['raw'])) {         \$res.= \$tmp;       } else {         \$res.= "<<";         if (\$this->compressionReady && \$this->options['compression']) {           // then implement ZLIB based compression on this content stream           \$res.= " /Filter /FlateDecode";           \$tmp = gzcompress(\$tmp, 6);         }          if (\$this->encrypted) {           \$this->encryptInit(\$id);           \$tmp = \$this->ARC4(\$tmp);         }          foreach (\$o['info'] as \$k => \$v) {           \$res.= "\n/\$k \$v";         }          \$res.= "\n/Length ".mb_strlen(\$tmp, '8bit') ." >>\nstream\n\$tmp\nendstream";       }        \$res.= "\nendobj";       return \$res;     }   }    protected function o_embedjs(\$id, \$action) {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'embedjs', 'info' => array(         'Names' => '[(EmbeddedJS) '.(\$id+1).' 0 R]'  ));       break;      case 'out':       \$res = "\n\$id 0 obj\n<< ";       foreach (\$o['info'] as \$k => \$v) {         \$res.= "\n/\$k \$v";       }       \$res.= "\n>>\nendobj";       return \$res;     }   }    protected function o_javascript(\$id, \$action, \$code = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       \$this->objects[\$id] = array('t' => 'javascript', 'info' => array(         'S' => '/JavaScript',          'JS' => '('.\$this->filterText(\$code).')',   ));       break;      case 'out':       \$res = "\n\$id 0 obj\n<< ";       foreach (\$o['info'] as \$k => \$v) {         \$res.= "\n/\$k \$v";       }       \$res.= "\n>>\nendobj";       return \$res;     }   }    /**    * an image object, will be an XObject in the document, includes description and data    */   protected function o_image(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       // make the new object       \$this->objects[\$id] = array('t' => 'image', 'data' => &\$options['data'], 'info' => array());        \$info =& \$this->objects[\$id]['info'];        \$info['Type'] = '/XObject';       \$info['Subtype'] = '/Image';       \$info['Width'] = \$options['iw'];       \$info['Height'] = \$options['ih'];        if (isset(\$options['masked']) && \$options['masked']) {         \$info['SMask'] = (\$this->numObj-1).' 0 R';       }        if (!isset(\$options['type']) || \$options['type'] === 'jpg') {         if (!isset(\$options['channels'])) {           \$options['channels'] = 3;         }          switch (\$options['channels']) {           case  1: \$info['ColorSpace'] = '/DeviceGray'; break;           case  4: \$info['ColorSpace'] = '/DeviceCMYK'; break;           default: \$info['ColorSpace'] = '/DeviceRGB'; break;         }          if (\$info['ColorSpace'] === '/DeviceCMYK') {           \$info['Decode'] = '[1 0 1 0 1 0 1 0]';         }          \$info['Filter'] = '/DCTDecode';         \$info['BitsPerComponent'] = 8;       }         else if (\$options['type'] === 'png') {         \$info['Filter'] = '/FlateDecode';         \$info['DecodeParms'] = '<< /Predictor 15 /Colors '.\$options['ncolor'].' /Columns '.\$options['iw'].' /BitsPerComponent '.\$options['bitsPerComponent'].'>>';          if (\$options['isMask']) {           \$info['ColorSpace'] = '/DeviceGray';         }         else {           if (mb_strlen(\$options['pdata'], '8bit')) {             \$tmp = ' [ /Indexed /DeviceRGB '.(mb_strlen(\$options['pdata'], '8bit') /3-1) .' ';             \$this->numObj++;             \$this->o_contents(\$this->numObj, 'new');             \$this->objects[\$this->numObj]['c'] = \$options['pdata'];             \$tmp.= \$this->numObj.' 0 R';             \$tmp.= ' ]';             \$info['ColorSpace'] = \$tmp;              if (isset(\$options['transparency'])) {               \$transparency = \$options['transparency'];               switch (\$transparency['type']) {               case 'indexed':                 \$tmp = ' [ '.\$transparency['data'].' '.\$transparency['data'].'] ';                 \$info['Mask'] = \$tmp;                 break;                case 'color-key':                 \$tmp = ' [ '.                   \$transparency['r'] . ' ' . \$transparency['r'] .                   \$transparency['g'] . ' ' . \$transparency['g'] .                   \$transparency['b'] . ' ' . \$transparency['b'] .                   ' ] ';                 \$info['Mask'] = \$tmp;                 break;               }             }           } else {             if (isset(\$options['transparency'])) {               \$transparency = \$options['transparency'];                switch (\$transparency['type']) {               case 'indexed':                 \$tmp = ' [ '.\$transparency['data'].' '.\$transparency['data'].'] ';                 \$info['Mask'] = \$tmp;                 break;                case 'color-key':                 \$tmp = ' [ '.                   \$transparency['r'] . ' ' . \$transparency['r'] . ' ' .                   \$transparency['g'] . ' ' . \$transparency['g'] . ' ' .                   \$transparency['b'] . ' ' . \$transparency['b'] .                   ' ] ';                 \$info['Mask'] = \$tmp;                 break;               }             }             \$info['ColorSpace'] = '/'.\$options['color'];           }         }          \$info['BitsPerComponent'] = \$options['bitsPerComponent'];       }        // assign it a place in the named resource dictionary as an external object, according to       // the label passed in with it.       \$this->o_pages(\$this->currentNode, 'xObject', array('label' => \$options['label'], 'objNum' => \$id));        // also make sure that we have the right procset object for it.       \$this->o_procset(\$this->procsetObjectId, 'add', 'ImageC');       break;      case 'out':       \$tmp = &\$o['data'];       \$res = "\n\$id 0 obj\n<<";        foreach (\$o['info'] as \$k => \$v) {         \$res.= "\n/\$k \$v";       }        if (\$this->encrypted) {         \$this->encryptInit(\$id);         \$tmp = \$this->ARC4(\$tmp);       }        \$res.= "\n/Length ".mb_strlen(\$tmp, '8bit') .">>\nstream\n\$tmp\nendstream\nendobj";        return \$res;     }   }    /**    * graphics state object    */   protected function o_extGState(\$id, \$action, \$options = "") {     static \$valid_params = array("LW", "LC", "LC", "LJ", "ML",                                     "D", "RI", "OP", "op", "OPM",                                     "Font", "BG", "BG2", "UCR",                                     "TR", "TR2", "HT", "FL",                                     "SM", "SA", "BM", "SMask",                                     "CA", "ca", "AIS", "TK");      if (\$action !== "new") {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case "new":       \$this->objects[\$id] = array('t' => 'extGState', 'info' => \$options);        // Tell the pages about the new resource       \$this->numStates++;       \$this->o_pages(\$this->currentNode, 'extGState', array("objNum" => \$id, "stateNum" => \$this->numStates));       break;      case "out":       \$res = "\n\$id 0 obj\n<< /Type /ExtGState\n";        foreach (\$o["info"] as \$k => \$v) {         if ( !in_array(\$k, \$valid_params))           continue;         \$res.= "/\$k \$v\n";       }        \$res.= ">>\nendobj";       return \$res;     }   }    /**    * encryption object.    */   protected function o_encryption(\$id, \$action, \$options = '') {     if (\$action !== 'new') {       \$o = & \$this->objects[\$id];     }      switch (\$action) {     case 'new':       // make the new object       \$this->objects[\$id] = array('t' => 'encryption', 'info' => \$options);       \$this->arc4_objnum = \$id;        // figure out the additional paramaters required       \$pad = chr(0x28) .chr(0xBF) .chr(0x4E) .chr(0x5E) .chr(0x4E) .chr(0x75) .chr(0x8A) .chr(0x41)              .chr(0x64) .chr(0x00) .chr(0x4E) .chr(0x56) .chr(0xFF) .chr(0xFA) .chr(0x01) .chr(0x08)              .chr(0x2E) .chr(0x2E) .chr(0x00) .chr(0xB6) .chr(0xD0) .chr(0x68) .chr(0x3E) .chr(0x80)              .chr(0x2F) .chr(0x0C) .chr(0xA9) .chr(0xFE) .chr(0x64) .chr(0x53) .chr(0x69) .chr(0x7A);        \$len = mb_strlen(\$options['owner'], '8bit');        if (\$len > 32) {         \$owner = substr(\$options['owner'], 0, 32);       } else if (\$len < 32) {         \$owner = \$options['owner'].substr(\$pad, 0, 32-\$len);       } else {         \$owner = \$options['owner'];       }        \$len = mb_strlen(\$options['user'], '8bit');       if (\$len > 32) {         \$user = substr(\$options['user'], 0, 32);       } else if (\$len < 32) {         \$user = \$options['user'].substr(\$pad, 0, 32-\$len);       } else {         \$user = \$options['user'];       }        \$tmp = \$this->md5_16(\$owner);       \$okey = substr(\$tmp, 0, 5);       \$this->ARC4_init(\$okey);       \$ovalue = \$this->ARC4(\$user);       \$this->objects[\$id]['info']['O'] = \$ovalue;        // now make the u value, phew.       \$tmp = \$this->md5_16(\$user.\$ovalue.chr(\$options['p']) .chr(255) .chr(255) .chr(255) .\$this->fileIdentifier);        \$ukey = substr(\$tmp, 0, 5);       \$this->ARC4_init(\$ukey);       \$this->encryptionKey = \$ukey;       \$this->encrypted = true;       \$uvalue = \$this->ARC4(\$pad);       \$this->objects[\$id]['info']['U'] = \$uvalue;       \$this->encryptionKey = \$ukey;       // initialize the arc4 array       break;      case 'out':       \$res = "\n\$id 0 obj\n<<";       \$res.= "\n/Filter /Standard";       \$res.= "\n/V 1";       \$res.= "\n/R 2";       \$res.= "\n/O (".\$this->filterText(\$o['info']['O'], true, false) .')';       \$res.= "\n/U (".\$this->filterText(\$o['info']['U'], true, false) .')';       // and the p-value needs to be converted to account for the twos-complement approach       \$o['info']['p'] = ((\$o['info']['p']^255) +1) *-1;       \$res.= "\n/P ".(\$o['info']['p']);       \$res.= "\n>>\nendobj";       return \$res;     }   }    /**    * ARC4 functions    * A series of function to implement ARC4 encoding in PHP    */    /**    * calculate the 16 byte version of the 128 bit md5 digest of the string    */   function md5_16(\$string) {     \$tmp = md5(\$string);     \$out = '';     for (\$i = 0; \$i <= 30; \$i = \$i+2) {       \$out.= chr(hexdec(substr(\$tmp, \$i, 2)));     }     return \$out;   }    /**    * initialize the encryption for processing a particular object    */   function encryptInit(\$id) {     \$tmp = \$this->encryptionKey;     \$hex = dechex(\$id);     if (mb_strlen(\$hex, '8bit') < 6) {       \$hex = substr('000000', 0, 6-mb_strlen(\$hex, '8bit')) .\$hex;     }     \$tmp.= chr(hexdec(substr(\$hex, 4, 2))) .chr(hexdec(substr(\$hex, 2, 2))) .chr(hexdec(substr(\$hex, 0, 2))) .chr(0) .chr(0);     \$key = \$this->md5_16(\$tmp);     \$this->ARC4_init(substr(\$key, 0, 10));   }    /**    * initialize the ARC4 encryption    */   function ARC4_init(\$key = '') {     \$this->arc4 = '';      // setup the control array     if (mb_strlen(\$key, '8bit') == 0) {       return;     }      \$k = '';     while (mb_strlen(\$k, '8bit') < 256) {       \$k.= \$key;     }      \$k = substr(\$k, 0, 256);     for (\$i = 0; \$i < 256; \$i++) {       \$this->arc4.= chr(\$i);     }      \$j = 0;      for (\$i = 0; \$i < 256; \$i++) {       \$t = \$this->arc4[\$i];       \$j = (\$j + ord(\$t) + ord(\$k[\$i])) %256;       \$this->arc4[\$i] = \$this->arc4[\$j];       \$this->arc4[\$j] = \$t;     }   }    /**    * ARC4 encrypt a text string    */   function ARC4(\$text) {     \$len = mb_strlen(\$text, '8bit');     \$a = 0;     \$b = 0;     \$c = \$this->arc4;     \$out = '';     for (\$i = 0; \$i < \$len; \$i++) {       \$a = (\$a+1) %256;       \$t = \$c[\$a];       \$b = (\$b+ord(\$t)) %256;       \$c[\$a] = \$c[\$b];       \$c[\$b] = \$t;       \$k = ord(\$c[(ord(\$c[\$a]) + ord(\$c[\$b])) %256]);       \$out.= chr(ord(\$text[\$i]) ^ \$k);     }     return \$out;   }    /**    * functions which can be called to adjust or add to the document    */    /**    * add a link in the document to an external URL    */   function addLink(\$url, \$x0, \$y0, \$x1, \$y1) {     \$this->numObj++;     \$info = array('type' => 'link', 'url' => \$url, 'rect' => array(\$x0, \$y0, \$x1, \$y1));     \$this->o_annotation(\$this->numObj, 'new', \$info);   }    /**    * add a link in the document to an internal destination (ie. within the document)    */   function addInternalLink(\$label, \$x0, \$y0, \$x1, \$y1) {     \$this->numObj++;     \$info = array('type' => 'ilink', 'label' => \$label, 'rect' => array(\$x0, \$y0, \$x1, \$y1));     \$this->o_annotation(\$this->numObj, 'new', \$info);   }    /**    * set the encryption of the document    * can be used to turn it on and/or set the passwords which it will have.    * also the functions that the user will have are set here, such as print, modify, add    */   function setEncryption(\$userPass = '', \$ownerPass = '', \$pc = array()) {     \$p = bindec("11000000");      \$options = array('print' => 4, 'modify' => 8, 'copy' => 16, 'add' => 32);      foreach (\$pc as \$k => \$v) {       if (\$v && isset(\$options[\$k])) {         \$p+= \$options[\$k];       } else if (isset(\$options[\$v])) {         \$p+= \$options[\$v];       }     }      // implement encryption on the document     if (\$this->arc4_objnum == 0) {       // then the block does not exist already, add it.       \$this->numObj++;       if (mb_strlen(\$ownerPass) == 0) {         \$ownerPass = \$userPass;       }        \$this->o_encryption(\$this->numObj, 'new', array('user' => \$userPass, 'owner' => \$ownerPass, 'p' => \$p));     }   }    /**    * should be used for internal checks, not implemented as yet    */   function checkAllHere() {   }    /**    * return the pdf stream as a string returned from the function    */   function output(\$debug = false) {     if (\$debug) {       // turn compression off       \$this->options['compression'] = false;     }      if (\$this->javascript) {       \$this->numObj++;        \$js_id = \$this->numObj;       \$this->o_embedjs(\$js_id, 'new');       \$this->o_javascript(++\$this->numObj, 'new', \$this->javascript);        \$id = \$this->catalogId;        \$this->o_catalog(\$id, 'javascript', \$js_id);     }      if (\$this->arc4_objnum) {       \$this->ARC4_init(\$this->encryptionKey);     }      \$this->checkAllHere();      \$xref = array();     \$content = '%PDF-1.3';     \$pos = mb_strlen(\$content, '8bit');      foreach (\$this->objects as \$k => \$v) {       \$tmp = 'o_'.\$v['t'];       \$cont = \$this->\$tmp(\$k, 'out');       \$content.= \$cont;       \$xref[] = \$pos;       \$pos+= mb_strlen(\$cont, '8bit');     }      \$content.= "\nxref\n0 ".(count(\$xref) +1) ."\n0000000000 65535 f \n";      foreach (\$xref as \$p) {       \$content.= str_pad(\$p, 10, "0", STR_PAD_LEFT) . " 00000 n \n";     }      \$content.= "trailer\n<<\n/Size ".(count(\$xref) +1) ."\n/Root 1 0 R\n/Info \$this->infoObject 0 R\n";      // if encryption has been applied to this document then add the marker for this dictionary     if (\$this->arc4_objnum > 0) {       \$content.= "/Encrypt \$this->arc4_objnum 0 R\n";     }      if (mb_strlen(\$this->fileIdentifier, '8bit')) {       \$content.= "/ID[<\$this->fileIdentifier><\$this->fileIdentifier>]\n";     }      // account for \n added at start of xref table     \$pos++;      \$content.= ">>\nstartxref\n\$pos\n%%EOF\n";      return \$content;   }    /**    * intialize a new document    * if this is called on an existing document results may be unpredictable, but the existing document would be lost at minimum    * this function is called automatically by the constructor function    */   private function newDocument(\$pageSize = array(0, 0, 612, 792)) {     \$this->numObj = 0;     \$this->objects = array();      \$this->numObj++;     \$this->o_catalog(\$this->numObj, 'new');      \$this->numObj++;     \$this->o_outlines(\$this->numObj, 'new');      \$this->numObj++;     \$this->o_pages(\$this->numObj, 'new');      \$this->o_pages(\$this->numObj, 'mediaBox', \$pageSize);     \$this->currentNode = 3;      \$this->numObj++;     \$this->o_procset(\$this->numObj, 'new');      \$this->numObj++;     \$this->o_info(\$this->numObj, 'new');      \$this->numObj++;     \$this->o_page(\$this->numObj, 'new');      // need to store the first page id as there is no way to get it to the user during     // startup     \$this->firstPageId = \$this->currentContents;   }    /**    * open the font file and return a php structure containing it.    * first check if this one has been done before and saved in a form more suited to php    * note that if a php serialized version does not exist it will try and make one, but will    * require write access to the directory to do it... it is MUCH faster to have these serialized    * files.    */   private function openFont(\$font) {     // assume that \$font contains the path and file but not the extension     \$pos = strrpos(\$font, '/');      if (\$pos === false) {       \$dir = './';       \$name = \$font;     } else {       \$dir = substr(\$font, 0, \$pos+1);       \$name = substr(\$font, \$pos+1);     }      \$fontcache = \$this->fontcache;     if (\$fontcache == '') {       \$fontcache = \$dir;     }      //\$name       filename without folder and extension of font metrics     //\$dir      folder of font metrics     //\$fontcache  folder of runtime created php serialized version of font metrics.     //            If this is not given, the same folder as the font metrics will be used.     //            Storing and reusing serialized versions improves speed much      \$this->addMessage("openFont: \$font - \$name");      if ( !\$this->isUnicode || in_array(mb_strtolower(basename(\$name)), self::\$coreFonts) ) {       \$metrics_name = "\$name.afm";     }     else {       \$metrics_name = "\$name.ufm";     }      \$cache_name = "\$metrics_name.php";     \$this->addMessage("metrics: \$metrics_name, cache: \$cache_name");      if (file_exists(\$fontcache . \$cache_name)) {       \$this->addMessage("openFont: php file exists \$fontcache\$cache_name");       \$this->fonts[\$font] = require(\$fontcache . \$cache_name);        if (!isset(\$this->fonts[\$font]['_version_']) || \$this->fonts[\$font]['_version_'] != \$this->fontcacheVersion) {         // if the font file is old, then clear it out and prepare for re-creation         \$this->addMessage('openFont: clear out, make way for new version.');         \$this->fonts[\$font] = null;         unset(\$this->fonts[\$font]);       }     }     else {       \$old_cache_name = "php_\$metrics_name";       if (file_exists(\$fontcache . \$old_cache_name)) {         \$this->addMessage("openFont: php file doesn't exist \$fontcache\$cache_name, creating it from the old format");         \$old_cache = file_get_contents(\$fontcache . \$old_cache_name);         file_put_contents(\$fontcache . \$cache_name, '<?php return ' . \$old_cache . ';');         return \$this->openFont(\$font);       }     }      if (!isset(\$this->fonts[\$font]) && file_exists(\$dir . \$metrics_name)) {       // then rebuild the php_<font>.afm file from the <font>.afm file       \$this->addMessage("openFont: build php file from \$dir\$metrics_name");       \$data = array();        // 20 => 'space'       \$data['codeToName'] = array();         // Since we're not going to enable Unicode for the core fonts we need to use a font-based       // setting for Unicode support rather than a global setting.       \$data['isUnicode'] = (strtolower(substr(\$metrics_name, -3)) !== 'afm');        \$cidtogid = '';       if (\$data['isUnicode']) {         \$cidtogid = str_pad('', 256*256*2, "\x00");       }        \$file = file(\$dir . \$metrics_name);        foreach (\$file as \$rowA) {         \$row = trim(\$rowA);         \$pos = strpos(\$row, ' ');          if (\$pos) {           // then there must be some keyword           \$key = substr(\$row, 0, \$pos);           switch (\$key) {           case 'FontName':           case 'FullName':           case 'FamilyName':           case 'PostScriptName':           case 'Weight':           case 'ItalicAngle':           case 'IsFixedPitch':           case 'CharacterSet':           case 'UnderlinePosition':           case 'UnderlineThickness':           case 'Version':           case 'EncodingScheme':           case 'CapHeight':           case 'XHeight':           case 'Ascender':           case 'Descender':           case 'StdHW':           case 'StdVW':           case 'StartCharMetrics':           case 'FontHeightOffset': // OAR - Added so we can offset the height calculation of a Windows font.  Otherwise it's too big.             \$data[\$key] = trim(substr(\$row, \$pos));             break;            case 'FontBBox':             \$data[\$key] = explode(' ', trim(substr(\$row, \$pos)));             break;            //C 39 ; WX 222 ; N quoteright ; B 53 463 157 718 ;           case 'C': // Found in AFM files             \$bits = explode(';', trim(\$row));             \$dtmp = array();              foreach (\$bits as \$bit) {               \$bits2 = explode(' ', trim(\$bit));               if (mb_strlen(\$bits2[0], '8bit') == 0) continue;                if (count(\$bits2) > 2) {                 \$dtmp[\$bits2[0]] = array();                 for (\$i = 1; \$i < count(\$bits2); \$i++) {                   \$dtmp[\$bits2[0]][] = \$bits2[\$i];                 }               } else if (count(\$bits2) == 2) {                 \$dtmp[\$bits2[0]] = \$bits2[1];               }             }              \$c = (int)\$dtmp['C'];             \$n = \$dtmp['N'];             \$width = floatval(\$dtmp['WX']);              if (\$c >= 0) {               if (\$c != hexdec(\$n)) {                 \$data['codeToName'][\$c] = \$n;               }               \$data['C'][\$c] = \$width;             } else {               \$data['C'][\$n] = \$width;             }              if (!isset(\$data['MissingWidth']) && \$c == -1 && \$n === '.notdef') {               \$data['MissingWidth'] = \$width;             }              break;            // U 827 ; WX 0 ; N squaresubnosp ; G 675 ;           case 'U': // Found in UFM files             if (!\$data['isUnicode']) break;              \$bits = explode(';', trim(\$row));             \$dtmp = array();              foreach (\$bits as \$bit) {               \$bits2 = explode(' ', trim(\$bit));               if (mb_strlen(\$bits2[0], '8bit') === 0) continue;                if (count(\$bits2) > 2) {                 \$dtmp[\$bits2[0]] = array();                 for (\$i = 1; \$i < count(\$bits2); \$i++) {                   \$dtmp[\$bits2[0]][] = \$bits2[\$i];                 }               } else if (count(\$bits2) == 2) {                 \$dtmp[\$bits2[0]] = \$bits2[1];               }             }              \$c = (int)\$dtmp['U'];             \$n = \$dtmp['N'];             \$glyph = \$dtmp['G'];             \$width = floatval(\$dtmp['WX']);              if (\$c >= 0) {               // Set values in CID to GID map               if (\$c >= 0 && \$c < 0xFFFF && \$glyph) {                 \$cidtogid[\$c*2] = chr(\$glyph >> 8);                 \$cidtogid[\$c*2 + 1] = chr(\$glyph & 0xFF);               }                if (\$c != hexdec(\$n)) {                 \$data['codeToName'][\$c] = \$n;               }               \$data['C'][\$c] = \$width;             } else {               \$data['C'][\$n] = \$width;             }              if (!isset(\$data['MissingWidth']) && \$c == -1 && \$n === '.notdef') {               \$data['MissingWidth'] = \$width;             }              break;            case 'KPX':             break; // don't include them as they are not used yet             //KPX Adieresis yacute -40             \$bits = explode(' ', trim(\$row));             \$data['KPX'][\$bits[1]][\$bits[2]] = \$bits[3];             break;           }         }       }        if (\$this->compressionReady && \$this->options['compression']) {         // then implement ZLIB based compression on CIDtoGID string         \$data['CIDtoGID_Compressed'] = true;         \$cidtogid = gzcompress(\$cidtogid, 6);       }       \$data['CIDtoGID'] = base64_encode(\$cidtogid);       \$data['_version_'] = \$this->fontcacheVersion;       \$this->fonts[\$font] = \$data;        //Because of potential trouble with php safe mode, expect that the folder already exists.       //If not existing, this will hit performance because of missing cached results.       if ( is_dir(substr(\$fontcache, 0, -1)) && is_writable(substr(\$fontcache, 0, -1)) ) {         file_put_contents(\$fontcache . \$cache_name, '<?php return ' . var_export(\$data, true) . ';');       }       \$data = null;     }      if (!isset(\$this->fonts[\$font])) {       \$this->addMessage("openFont: no font file found for \$font. Do you need to run load_font.php?");     }      //pre_r(\$this->messages);   }    /**    * if the font is not loaded then load it and make the required object    * else just make it the current font    * the encoding array can contain 'encoding'=> 'none', 'WinAnsiEncoding', 'MacRomanEncoding' or 'MacExpertEncoding'    * note that encoding='none' will need to be used for symbolic fonts    * and 'differences' => an array of mappings between numbers 0->255 and character names.    *    */   function selectFont(\$fontName, \$encoding = '', \$set = true) {     \$ext = substr(\$fontName, -4);     if (\$ext === '.afm' || \$ext === '.ufm') {       \$fontName = substr(\$fontName, 0, mb_strlen(\$fontName)-4);     }      if (!isset(\$this->fonts[\$fontName])) {       \$this->addMessage("selectFont: selecting - \$fontName - \$encoding, \$set");        // load the file       \$this->openFont(\$fontName);        if (isset(\$this->fonts[\$fontName])) {         \$this->numObj++;         \$this->numFonts++;          \$font = &\$this->fonts[\$fontName];          //\$this->numFonts = md5(\$fontName);         \$pos = strrpos(\$fontName, '/');         //      \$dir = substr(\$fontName, 0, \$pos+1);         \$name = substr(\$fontName, \$pos+1);         \$options = array('name' => \$name, 'fontFileName' => \$fontName);          if (is_array(\$encoding)) {           // then encoding and differences might be set           if (isset(\$encoding['encoding'])) {             \$options['encoding'] = \$encoding['encoding'];           }            if (isset(\$encoding['differences'])) {             \$options['differences'] = \$encoding['differences'];           }         } else if (mb_strlen(\$encoding, '8bit')) {           // then perhaps only the encoding has been set           \$options['encoding'] = \$encoding;         }          \$fontObj = \$this->numObj;         \$this->o_font(\$this->numObj, 'new', \$options);         \$font['fontNum'] = \$this->numFonts;          // if this is a '.afm' font, and there is a '.pfa' file to go with it ( as there         // should be for all non-basic fonts), then load it into an object and put the         // references into the font object         \$basefile = \$fontName;          \$fbtype = '';         if (file_exists("\$basefile.pfb")) {           \$fbtype = 'pfb';         }          else if (file_exists("\$basefile.ttf")) {           \$fbtype = 'ttf';         }          \$fbfile = "\$basefile.\$fbtype";          //      \$pfbfile = substr(\$fontName, 0, strlen(\$fontName)-4).'.pfb';         //      \$ttffile = substr(\$fontName, 0, strlen(\$fontName)-4).'.ttf';         \$this->addMessage('selectFont: checking for - '.\$fbfile);          // OAR - I don't understand this old check         // if (substr(\$fontName, -4) ===  '.afm' &&  strlen(\$fbtype)) {         if (\$fbtype) {           \$adobeFontName = isset(\$font['PostScriptName']) ? \$font['PostScriptName'] : \$font['FontName'];           //        \$fontObj = \$this->numObj;           \$this->addMessage("selectFont: adding font file - \$fbfile - \$adobeFontName");            // find the array of font widths, and put that into an object.           \$firstChar = -1;           \$lastChar = 0;           \$widths = array();           \$cid_widths = array();            foreach (\$font['C'] as \$num => \$d) {             if (intval(\$num) > 0 || \$num == '0') {               if (!\$font['isUnicode']) {                 // With Unicode, widths array isn't used                 if (\$lastChar>0 && \$num>\$lastChar+1) {                   for (\$i = \$lastChar+1; \$i<\$num; \$i++) {                     \$widths[] = 0;                   }                 }               }                \$widths[] = \$d;                if (\$font['isUnicode']) {                 \$cid_widths[\$num] = \$d;               }                if (\$firstChar == -1) {                 \$firstChar = \$num;               }                \$lastChar = \$num;             }           }            // also need to adjust the widths for the differences array           if (isset(\$options['differences'])) {             foreach (\$options['differences'] as \$charNum => \$charName) {               if (\$charNum > \$lastChar) {                 if (!\$font['isUnicode']) {                   // With Unicode, widths array isn't used                   for (\$i = \$lastChar + 1; \$i <= \$charNum; \$i++) {                     \$widths[] = 0;                   }                 }                  \$lastChar = \$charNum;               }                if (isset(\$font['C'][\$charName])) {                 \$widths[\$charNum-\$firstChar] = \$font['C'][\$charName];                 if (\$font['isUnicode']) {                   \$cid_widths[\$charName] = \$font['C'][\$charName];                 }               }             }           }            if (\$font['isUnicode']) {             \$font['CIDWidths'] = \$cid_widths;           }            \$this->addMessage('selectFont: FirstChar = '.\$firstChar);           \$this->addMessage('selectFont: LastChar = '.\$lastChar);            \$widthid = -1;            if (!\$font['isUnicode']) {             // With Unicode, widths array isn't used              \$this->numObj++;             \$this->o_contents(\$this->numObj, 'new', 'raw');             \$this->objects[\$this->numObj]['c'].= '['.implode(' ', \$widths).']';             \$widthid = \$this->numObj;           }            \$missing_width = 500;           \$stemV = 70;            if (isset(\$font['MissingWidth'])) {             \$missing_width = \$font['MissingWidth'];           }           if (isset(\$font['StdVW'])) {             \$stemV = \$font['StdVW'];           } else if (isset(\$font['Weight']) && preg_match('!(bold|black)!i', \$font['Weight'])) {             \$stemV = 120;           }            // load the pfb file, and put that into an object too.           // note that pdf supports only binary format type 1 font files, though there is a           // simple utility to convert them from pfa to pfb.           // FIXME: should we move font subset creation to CPDF::output? See notes in issue #750.           if (!\$this->isUnicode || \$fbtype !== 'ttf' || empty(\$this->stringSubsets)) {             \$data = file_get_contents(\$fbfile);           }           else {             \$this->stringSubsets[\$fontName][] = 32; // Force space if not in yet              \$subset = \$this->stringSubsets[\$fontName];             sort(\$subset);              // Load font             \$font_obj = Font::load(\$fbfile);             \$font_obj->parse();              // Define subset             \$font_obj->setSubset(\$subset);             \$font_obj->reduce();              // Write new font             \$tmp_name = "\$fbfile.tmp.".uniqid();             \$font_obj->open(\$tmp_name, Font_Binary_Stream::modeWrite);             \$font_obj->encode(array("OS/2"));             \$font_obj->close();              // Parse the new font to get cid2gid and widths             \$font_obj = Font::load(\$tmp_name);              // Find Unicode char map table             \$subtable = null;             foreach (\$font_obj->getData("cmap", "subtables") as \$_subtable) {               if (\$_subtable["platformID"] == 0 || \$_subtable["platformID"] == 3 && \$_subtable["platformSpecificID"] == 1) {                 \$subtable = \$_subtable;                 break;               }             }              if (\$subtable) {               \$glyphIndexArray = \$subtable["glyphIndexArray"];               \$hmtx = \$font_obj->getData("hmtx");                unset(\$glyphIndexArray[0xFFFF]);                \$cidtogid = str_pad('', max(array_keys(\$glyphIndexArray))*2+1, "\x00");               \$font['CIDWidths'] = array();               foreach (\$glyphIndexArray as \$cid => \$gid) {                 if (\$cid >= 0 && \$cid < 0xFFFF && \$gid) {                   \$cidtogid[\$cid*2] = chr(\$gid >> 8);                   \$cidtogid[\$cid*2 + 1] = chr(\$gid & 0xFF);                 }                  \$width = \$font_obj->normalizeFUnit(isset(\$hmtx[\$gid]) ? \$hmtx[\$gid][0] : \$hmtx[0][0]);                 \$font['CIDWidths'][\$cid] = \$width;               }                \$font['CIDtoGID'] = base64_encode(gzcompress(\$cidtogid));               \$font['CIDtoGID_Compressed'] = true;                \$data = file_get_contents(\$tmp_name);             }             else {               \$data = file_get_contents(\$fbfile);             }              \$font_obj->close();             unlink(\$tmp_name);           }            // create the font descriptor           \$this->numObj++;           \$fontDescriptorId = \$this->numObj;            \$this->numObj++;           \$pfbid = \$this->numObj;            // determine flags (more than a little flakey, hopefully will not matter much)           \$flags = 0;            if (\$font['ItalicAngle'] != 0) {             \$flags+= pow(2, 6);           }            if (\$font['IsFixedPitch'] === 'true') {             \$flags+= 1;           }            \$flags+= pow(2, 5); // assume non-sybolic           \$list = array(             'Ascent' => 'Ascender',              'CapHeight' => 'CapHeight',              'MissingWidth' => 'MissingWidth',              'Descent' => 'Descender',              'FontBBox' => 'FontBBox',              'ItalicAngle' => 'ItalicAngle'  );           \$fdopt = array(             'Flags' => \$flags,              'FontName' => \$adobeFontName,              'StemV' => \$stemV  );            foreach (\$list as \$k => \$v) {             if (isset(\$font[\$v])) {               \$fdopt[\$k] = \$font[\$v];             }           }            if (\$fbtype === 'pfb') {             \$fdopt['FontFile'] = \$pfbid;           } else if (\$fbtype === 'ttf') {             \$fdopt['FontFile2'] = \$pfbid;           }            \$this->o_fontDescriptor(\$fontDescriptorId, 'new', \$fdopt);            // embed the font program           \$this->o_contents(\$this->numObj, 'new');           \$this->objects[\$pfbid]['c'].= \$data;            // determine the cruicial lengths within this file           if (\$fbtype === 'pfb') {             \$l1 = strpos(\$data, 'eexec') +6;             \$l2 = strpos(\$data, '00000000') -\$l1;             \$l3 = mb_strlen(\$data, '8bit') -\$l2-\$l1;             \$this->o_contents(\$this->numObj, 'add', array('Length1' => \$l1, 'Length2' => \$l2, 'Length3' => \$l3));           } else if (\$fbtype == 'ttf') {             \$l1 = mb_strlen(\$data, '8bit');             \$this->o_contents(\$this->numObj, 'add', array('Length1' => \$l1));           }            // tell the font object about all this new stuff           \$tmp = array(             'BaseFont' => \$adobeFontName,              'MissingWidth' => \$missing_width,              'Widths' => \$widthid,              'FirstChar' => \$firstChar,              'LastChar' => \$lastChar,              'FontDescriptor' => \$fontDescriptorId  );            if (\$fbtype === 'ttf') {             \$tmp['SubType'] = 'TrueType';           }            \$this->addMessage("adding extra info to font.(\$fontObj)");            foreach (\$tmp as \$fk => \$fv) {             \$this->addMessage("\$fk : \$fv");           }            \$this->o_font(\$fontObj, 'add', \$tmp);         } else {           \$this->addMessage('selectFont: pfb or ttf file not found, ok if this is one of the 14 standard fonts');         }          // also set the differences here, note that this means that these will take effect only the         //first time that a font is selected, else they are ignored         if (isset(\$options['differences'])) {           \$font['differences'] = \$options['differences'];         }       }     }      if (\$set && isset(\$this->fonts[\$fontName])) {       // so if for some reason the font was not set in the last one then it will not be selected       \$this->currentBaseFont = \$fontName;        // the next lines mean that if a new font is selected, then the current text state will be       // applied to it as well.       \$this->currentFont = \$this->currentBaseFont;       \$this->currentFontNum = \$this->fonts[\$this->currentFont]['fontNum'];        //\$this->setCurrentFont();     }      return \$this->currentFontNum;     //return \$this->numObj;   }    /**    * sets up the current font, based on the font families, and the current text state    * note that this system is quite flexible, a bold-italic font can be completely different to a    * italic-bold font, and even bold-bold will have to be defined within the family to have meaning    * This function is to be called whenever the currentTextState is changed, it will update    * the currentFont setting to whatever the appropriatte family one is.    * If the user calls selectFont themselves then that will reset the currentBaseFont, and the currentFont    * This function will change the currentFont to whatever it should be, but will not change the    * currentBaseFont.    */   private function setCurrentFont() {     //   if (strlen(\$this->currentBaseFont) == 0) {     //     // then assume an initial font     //     \$this->selectFont(\$this->defaultFont);     //   }     //   \$cf = substr(\$this->currentBaseFont, strrpos(\$this->currentBaseFont, '/')+1);     //   if (strlen(\$this->currentTextState)     //     && isset(\$this->fontFamilies[\$cf])     //       && isset(\$this->fontFamilies[\$cf][\$this->currentTextState])) {     //     // then we are in some state or another     //     // and this font has a family, and the current setting exists within it     //     // select the font, then return it     //     \$nf = substr(\$this->currentBaseFont, 0, strrpos(\$this->currentBaseFont, '/')+1).\$this->fontFamilies[\$cf][\$this->currentTextState];     //     \$this->selectFont(\$nf, '', 0);     //     \$this->currentFont = \$nf;     //     \$this->currentFontNum = \$this->fonts[\$nf]['fontNum'];     //   } else {     //     // the this font must not have the right family member for the current state     //     // simply assume the base font     \$this->currentFont = \$this->currentBaseFont;     \$this->currentFontNum = \$this->fonts[\$this->currentFont]['fontNum'];     //  }   }    /**    * function for the user to find out what the ID is of the first page that was created during    * startup - useful if they wish to add something to it later.    */   function getFirstPageId() {     return \$this->firstPageId;   }    /**    * add content to the currently active object    */   private function addContent(\$content) {     \$this->objects[\$this->currentContents]['c'] .=  \$content;   }    /**    * sets the color for fill operations    */   function setColor(\$color, \$force = false) {     \$new_color = array(\$color[0], \$color[1], \$color[2], isset(\$color[3]) ? \$color[3] : null);      if (!\$force && \$this->currentColor == \$new_color) {       return;     }      if (isset(\$new_color[3])) {       \$this->currentColor = \$new_color;       \$this->addContent(vsprintf("\n%.3F %.3F %.3F %.3F k", \$this->currentColor));     }      else if (isset(\$new_color[2])) {       \$this->currentColor = \$new_color;       \$this->addContent(vsprintf("\n%.3F %.3F %.3F rg", \$this->currentColor));     }   }    /**    * sets the color for stroke operations    */   function setStrokeColor(\$color, \$force = false) {     \$new_color = array(\$color[0], \$color[1], \$color[2], isset(\$color[3]) ? \$color[3] : null);      if (!\$force && \$this->currentStrokeColor == \$new_color) return;      if (isset(\$new_color[3])) {       \$this->currentStrokeColor = \$new_color;       \$this->addContent(vsprintf("\n%.3F %.3F %.3F %.3F K", \$this->currentStrokeColor));     }      else if (isset(\$new_color[2])) {       \$this->currentStrokeColor = \$new_color;       \$this->addContent(vsprintf("\n%.3F %.3F %.3F RG", \$this->currentStrokeColor));     }   }    /**    * Set the graphics state for compositions    */   function setGraphicsState(\$parameters) {     // Create a new graphics state object     // FIXME: should actually keep track of states that have already been created...     \$this->numObj++;     \$this->o_extGState(\$this->numObj, 'new', \$parameters);     \$this->addContent("\n/GS\$this->numStates gs");   }    /**    * Set current blend mode & opacity for lines.    *    * Valid blend modes are:    *    * Normal, Multiply, Screen, Overlay, Darken, Lighten,     * ColorDogde, ColorBurn, HardLight, SoftLight, Difference,     * Exclusion    *    * @param string \$mode the blend mode to use    * @param float \$opacity 0.0 fully transparent, 1.0 fully opaque    */   function setLineTransparency(\$mode, \$opacity) {     static \$blend_modes = array("Normal", "Multiply", "Screen",                                  "Overlay", "Darken", "Lighten",                                  "ColorDogde", "ColorBurn", "HardLight",                                  "SoftLight", "Difference", "Exclusion");      if ( !in_array(\$mode, \$blend_modes) )       \$mode = "Normal";      // Only create a new graphics state if required     if ( \$mode === \$this->currentLineTransparency["mode"] &&          \$opacity == \$this->currentLineTransparency["opacity"] )       return;      \$this->currentLineTransparency["mode"] = \$mode;     \$this->currentLineTransparency["opacity"] = \$opacity;      \$options = array("BM" => "/\$mode",                       "CA" => (float)\$opacity);      \$this->setGraphicsState(\$options);   }    /**    * Set current blend mode & opacity for filled objects.    *    * Valid blend modes are:    *    * Normal, Multiply, Screen, Overlay, Darken, Lighten,     * ColorDogde, ColorBurn, HardLight, SoftLight, Difference,     * Exclusion    *    * @param string \$mode the blend mode to use    * @param float \$opacity 0.0 fully transparent, 1.0 fully opaque    */   function setFillTransparency(\$mode, \$opacity) {     static \$blend_modes = array("Normal", "Multiply", "Screen",                                  "Overlay", "Darken", "Lighten",                                  "ColorDogde", "ColorBurn", "HardLight",                                  "SoftLight", "Difference", "Exclusion");      if ( !in_array(\$mode, \$blend_modes) ) {       \$mode = "Normal";     }      if ( \$mode === \$this->currentFillTransparency["mode"] &&          \$opacity == \$this->currentFillTransparency["opacity"] ) {       return;     }      \$this->currentFillTransparency["mode"] = \$mode;     \$this->currentFillTransparency["opacity"] = \$opacity;      \$options = array(       "BM" => "/\$mode",        "ca" => (float)\$opacity,   );      \$this->setGraphicsState(\$options);   }    /**    * draw a line from one set of coordinates to another    */   function line(\$x1, \$y1, \$x2, \$y2, \$stroke = true) {     \$this->addContent(sprintf("\n%.3F %.3F m %.3F %.3F l", \$x1, \$y1, \$x2, \$y2));      if (\$stroke) {       \$this->addContent(' S');     }   }    /**    * draw a bezier curve based on 4 control points    */   function curve(\$x0, \$y0, \$x1, \$y1, \$x2, \$y2, \$x3, \$y3) {     // in the current line style, draw a bezier curve from (x0, y0) to (x3, y3) using the other two points     // as the control points for the curve.     \$this->addContent(sprintf("\n%.3F %.3F m %.3F %.3F %.3F %.3F %.3F %.3F c S", \$x0, \$y0, \$x1, \$y1, \$x2, \$y2, \$x3, \$y3));   }    /**    * draw a part of an ellipse    */   function partEllipse(\$x0, \$y0, \$astart, \$afinish, \$r1, \$r2 = 0, \$angle = 0, \$nSeg = 8) {     \$this->ellipse(\$x0, \$y0, \$r1, \$r2, \$angle, \$nSeg, \$astart, \$afinish, false);   }    /**    * draw a filled ellipse    */   function filledEllipse(\$x0, \$y0, \$r1, \$r2 = 0, \$angle = 0, \$nSeg = 8, \$astart = 0, \$afinish = 360) {     return \$this->ellipse(\$x0, \$y0, \$r1, \$r2, \$angle, \$nSeg, \$astart, \$afinish, true, true);   }    /**    * draw an ellipse    * note that the part and filled ellipse are just special cases of this function    *    * draws an ellipse in the current line style    * centered at \$x0, \$y0, radii \$r1, \$r2    * if \$r2 is not set, then a circle is drawn    * from \$astart to \$afinish, measured in degrees, running anti-clockwise from the right hand side of the ellipse.    * nSeg is not allowed to be less than 2, as this will simply draw a line (and will even draw a    * pretty crappy shape at 2, as we are approximating with bezier curves.    */   function ellipse(\$x0, \$y0, \$r1, \$r2 = 0, \$angle = 0, \$nSeg = 8, \$astart = 0, \$afinish = 360, \$close = true, \$fill = false, \$stroke = true, \$incomplete = false) {     if (\$r1 == 0) {       return;     }      if (\$r2 == 0) {       \$r2 = \$r1;     }      if (\$nSeg < 2) {       \$nSeg = 2;     }      \$astart = deg2rad((float)\$astart);     \$afinish = deg2rad((float)\$afinish);     \$totalAngle = \$afinish-\$astart;      \$dt = \$totalAngle/\$nSeg;     \$dtm = \$dt/3;      if (\$angle != 0) {       \$a = -1*deg2rad((float)\$angle);        \$this->addContent(sprintf("\n q %.3F %.3F %.3F %.3F %.3F %.3F cm", cos(\$a), -sin(\$a), sin(\$a), cos(\$a), \$x0, \$y0));        \$x0 = 0;       \$y0 = 0;     }      \$t1 = \$astart;     \$a0 = \$x0 + \$r1*cos(\$t1);     \$b0 = \$y0 + \$r2*sin(\$t1);     \$c0 = -\$r1 * sin(\$t1);     \$d0 = \$r2 * cos(\$t1);      if (!\$incomplete) {       \$this->addContent(sprintf("\n%.3F %.3F m ", \$a0, \$b0));     }      for (\$i = 1; \$i <= \$nSeg; \$i++) {       // draw this bit of the total curve       \$t1 = \$i * \$dt + \$astart;       \$a1 = \$x0 + \$r1 * cos(\$t1);       \$b1 = \$y0 + \$r2 * sin(\$t1);       \$c1 = -\$r1 * sin(\$t1);       \$d1 = \$r2 * cos(\$t1);        \$this->addContent(sprintf("\n%.3F %.3F %.3F %.3F %.3F %.3F c", (\$a0+\$c0*\$dtm), (\$b0+\$d0*\$dtm), (\$a1-\$c1*\$dtm), (\$b1-\$d1*\$dtm), \$a1, \$b1));        \$a0 = \$a1;       \$b0 = \$b1;       \$c0 = \$c1;       \$d0 = \$d1;     }      if (!\$incomplete) {       if (\$fill) {         \$this->addContent(' f');       }       else if (\$close) {         \$this->addContent(' s'); // small 's' signifies closing the path as well       }       else if (\$stroke) {         \$this->addContent(' S');       }     }      if (\$angle != 0) {       \$this->addContent(' Q');     }   }    /**    * this sets the line drawing style.    * width, is the thickness of the line in user units    * cap is the type of cap to put on the line, values can be 'butt', 'round', 'square'    *    where the diffference between 'square' and 'butt' is that 'square' projects a flat end past the    *    end of the line.    * join can be 'miter', 'round', 'bevel'    * dash is an array which sets the dash pattern, is a series of length values, which are the lengths of the    *   on and off dashes.    *   (2) represents 2 on, 2 off, 2 on , 2 off ...    *   (2, 1) is 2 on, 1 off, 2 on, 1 off.. etc    * phase is a modifier on the dash pattern which is used to shift the point at which the pattern starts.    */   function setLineStyle(\$width = 1, \$cap = '', \$join = '', \$dash = '', \$phase = 0) {     // this is quite inefficient in that it sets all the parameters whenever 1 is changed, but will fix another day     \$string = '';      if (\$width > 0) {       \$string.= "\$width w";     }      \$ca = array('butt' => 0, 'round' => 1, 'square' => 2);      if (isset(\$ca[\$cap])) {       \$string.= " \$ca[\$cap] J";     }      \$ja = array('miter' => 0, 'round' => 1, 'bevel' => 2);      if (isset(\$ja[\$join])) {       \$string.= " \$ja[\$join] j";     }      if (is_array(\$dash)) {       \$string.= ' [ ' . implode(' ', \$dash) . " ] \$phase d";     }      \$this->currentLineStyle = \$string;     \$this->addContent("\n\$string");   }    /**    * draw a polygon, the syntax for this is similar to the GD polygon command    */   function polygon(\$p, \$np, \$f = false) {     \$this->addContent(sprintf("\n%.3F %.3F m ", \$p[0], \$p[1]));      for (\$i = 2; \$i < \$np * 2; \$i = \$i + 2) {       \$this->addContent(sprintf("%.3F %.3F l ", \$p[\$i], \$p[\$i+1]));     }      if (\$f) {       \$this->addContent(' f');     } else {       \$this->addContent(' S');     }   }    /**    * a filled rectangle, note that it is the width and height of the rectangle which are the secondary paramaters, not    * the coordinates of the upper-right corner    */   function filledRectangle(\$x1, \$y1, \$width, \$height) {     \$this->addContent(sprintf("\n%.3F %.3F %.3F %.3F re f", \$x1, \$y1, \$width, \$height));   }    /**    * draw a rectangle, note that it is the width and height of the rectangle which are the secondary paramaters, not    * the coordinates of the upper-right corner    */   function rectangle(\$x1, \$y1, \$width, \$height) {     \$this->addContent(sprintf("\n%.3F %.3F %.3F %.3F re S", \$x1, \$y1, \$width, \$height));   }    /**    * save the current graphic state    */   function save() {     // we must reset the color cache or it will keep bad colors after clipping     \$this->currentColor = null;     \$this->currentStrokeColor = null;     \$this->addContent("\nq");   }    /**    * restore the last graphic state    */   function restore() {     // we must reset the color cache or it will keep bad colors after clipping     \$this->currentColor = null;     \$this->currentStrokeColor = null;     \$this->addContent("\nQ");   }    /**    * draw a clipping rectangle, all the elements added after this will be clipped    */   function clippingRectangle(\$x1, \$y1, \$width, \$height) {     \$this->save();     \$this->addContent(sprintf("\n%.3F %.3F %.3F %.3F re W n", \$x1, \$y1, \$width, \$height));   }    /**    * draw a clipping rounded rectangle, all the elements added after this will be clipped    */   function clippingRectangleRounded(\$x1, \$y1, \$w, \$h, \$rTL, \$rTR, \$rBR, \$rBL) {     \$this->save();      // start: top edge, left end     \$this->addContent(sprintf("\n%.3F %.3F m ", \$x1, \$y1 - \$rTL + \$h));      // curve: bottom-left corner     \$this->ellipse(\$x1 + \$rBL, \$y1 + \$rBL, \$rBL, 0, 0, 8, 180, 270, false, false, false, true);      // line: right edge, bottom end     \$this->addContent(sprintf("\n%.3F %.3F l ", \$x1 + \$w - \$rBR, \$y1));      // curve: bottom-right corner     \$this->ellipse(\$x1 + \$w - \$rBR, \$y1 + \$rBR, \$rBR, 0, 0, 8, 270, 360, false, false, false, true);      // line: right edge, top end     \$this->addContent(sprintf("\n%.3F %.3F l ", \$x1 + \$w, \$y1 + \$h - \$rTR));      // curve: bottom-right corner     \$this->ellipse(\$x1 + \$w - \$rTR, \$y1 + \$h - \$rTR, \$rTR, 0, 0, 8, 0, 90, false, false, false, true);      // line: bottom edge, right end     \$this->addContent(sprintf("\n%.3F %.3F l ", \$x1 + \$rTL, \$y1 + \$h));      // curve: top-right corner     \$this->ellipse(\$x1 + \$rTL, \$y1 + \$h - \$rTL, \$rTL, 0, 0, 8, 90, 180, false, false, false, true);      // line: top edge, left end     \$this->addContent(sprintf("\n%.3F %.3F l ", \$x1 + \$rBL, \$y1));      // Close & clip     \$this->addContent(" W n");   }    /**    * ends the last clipping shape    */   function clippingEnd() {     \$this->restore();   }    /**    * scale    * @param float \$s_x scaling factor for width as percent    * @param float \$s_y scaling factor for height as percent    * @param float \$x Origin abscisse    * @param float \$y Origin ordinate    */   function scale(\$s_x, \$s_y, \$x, \$y) {     \$y = \$this->currentPageSize["height"] - \$y;      \$tm = array(       \$s_x, 0,        0, \$s_y,        \$x*(1-\$s_x), \$y*(1-\$s_y)  );      \$this->transform(\$tm);   }    /**    * translate    * @param float \$t_x movement to the right    * @param float \$t_y movement to the bottom    */   function translate(\$t_x, \$t_y) {     \$tm = array(       1, 0,        0, 1,        \$t_x, -\$t_y  );      \$this->transform(\$tm);   }    /**    * rotate    * @param float \$angle angle in degrees for counter-clockwise rotation    * @param float \$x Origin abscisse    * @param float \$y Origin ordinate    */   function rotate(\$angle, \$x, \$y) {     \$y = \$this->currentPageSize["height"] - \$y;      \$a = deg2rad(\$angle);     \$cos_a = cos(\$a);     \$sin_a = sin(\$a);      \$tm = array(       \$cos_a, -\$sin_a,        \$sin_a, \$cos_a,        \$x - \$sin_a*\$y - \$cos_a*\$x, \$y - \$cos_a*\$y + \$sin_a*\$x,   );      \$this->transform(\$tm);   }    /**    * skew    * @param float \$angle_x    * @param float \$angle_y    * @param float \$x Origin abscisse    * @param float \$y Origin ordinate    */   function skew(\$angle_x, \$angle_y, \$x, \$y) {     \$y = \$this->currentPageSize["height"] - \$y;      \$tan_x = tan(deg2rad(\$angle_x));     \$tan_y = tan(deg2rad(\$angle_y));      \$tm = array(       1, -\$tan_y,        -\$tan_x, 1,        \$tan_x*\$y, \$tan_y*\$x,   );      \$this->transform(\$tm);   }    /**    * apply graphic transformations    * @param array \$tm transformation matrix    */   function transform(\$tm) {     \$this->addContent(vsprintf("\n %.3F %.3F %.3F %.3F %.3F %.3F cm", \$tm));   }    /**    * add a new page to the document    * this also makes the new page the current active object    */   function newPage(\$insert = 0, \$id = 0, \$pos = 'after') {     // if there is a state saved, then go up the stack closing them     // then on the new page, re-open them with the right setings      if (\$this->nStateStack) {       for (\$i = \$this->nStateStack; \$i >= 1; \$i--) {         \$this->restoreState(\$i);       }     }      \$this->numObj++;      if (\$insert) {       // the id from the ezPdf class is the id of the contents of the page, not the page object itself       // query that object to find the parent       \$rid = \$this->objects[\$id]['onPage'];       \$opt = array('rid' => \$rid, 'pos' => \$pos);       \$this->o_page(\$this->numObj, 'new', \$opt);     } else {       \$this->o_page(\$this->numObj, 'new');     }      // if there is a stack saved, then put that onto the page     if (\$this->nStateStack) {       for (\$i = 1; \$i <= \$this->nStateStack; \$i++) {         \$this->saveState(\$i);       }     }      // and if there has been a stroke or fill color set, then transfer them     if (isset(\$this->currentColor)) {       \$this->setColor(\$this->currentColor, true);     }      if (isset(\$this->currentStrokeColor)) {       \$this->setStrokeColor(\$this->currentStrokeColor, true);     }      // if there is a line style set, then put this in too     if (mb_strlen(\$this->currentLineStyle, '8bit')) {       \$this->addContent("\n\$this->currentLineStyle");     }      // the call to the o_page object set currentContents to the present page, so this can be returned as the page id     return \$this->currentContents;   }    /**    * output the pdf code, streaming it to the browser    * the relevant headers are set so that hopefully the browser will recognise it    */   function stream(\$options = '') {     // setting the options allows the adjustment of the headers     // values at the moment are:     // 'Content-Disposition' => 'filename'  - sets the filename, though not too sure how well this will     //        work as in my trial the browser seems to use the filename of the php file with .pdf on the end     // 'Accept-Ranges' => 1 or 0 - if this is not set to 1, then this header is not included, off by default     //    this header seems to have caused some problems despite tha fact that it is supposed to solve     //    them, so I am leaving it off by default.     // 'compress' = > 1 or 0 - apply content stream compression, this is on (1) by default     // 'Attachment' => 1 or 0 - if 1, force the browser to open a download dialog     if (!is_array(\$options)) {       \$options = array();     }      if ( headers_sent()) {       die("Unable to stream pdf: headers already sent");     }      \$debug = empty(\$options['compression']);     \$tmp = ltrim(\$this->output(\$debug));      header("Cache-Control: private");     header("Content-type: application/pdf");      //FIXME: I don't know that this is sufficient for determining content length (i.e. what about transport compression?)     header("Content-Length: " . mb_strlen(\$tmp, '8bit'));     \$fileName = (isset(\$options['Content-Disposition']) ?  \$options['Content-Disposition'] :  'file.pdf');      if ( !isset(\$options["Attachment"]))       \$options["Attachment"] = true;      \$attachment = \$options["Attachment"] ? "attachment" : "inline";          // detect the character encoding of the incoming file     \$encoding = mb_detect_encoding(\$fileName);     \$fallbackfilename = mb_convert_encoding(\$fileName, "ISO-8859-1", \$encoding);      \$encodedfallbackfilename = rawurlencode(\$fallbackfilename);     \$encodedfilename = rawurlencode(\$fileName);      header("Content-Disposition: \$attachment; filename=". \$encodedfallbackfilename ."; filename*=UTF-8''\$encodedfilename");      if (isset(\$options['Accept-Ranges']) && \$options['Accept-Ranges'] == 1) {       //FIXME: Is this the correct value ... spec says 1#range-unit       header("Accept-Ranges: " . mb_strlen(\$tmp, '8bit'));     }      echo \$tmp;     flush();   }    /**    * return the height in units of the current font in the given size    */   function getFontHeight(\$size) {     if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      \$font = \$this->fonts[\$this->currentFont];      // for the current font, and the given size, what is the height of the font in user units     if ( isset(\$font['Ascender']) && isset(\$font['Descender']) ) {       \$h = \$font['Ascender']-\$font['Descender'];     }     else {       \$h = \$font['FontBBox'][3]-\$font['FontBBox'][1];     }      // have to adjust by a font offset for Windows fonts.  unfortunately it looks like     // the bounding box calculations are wrong and I don't know why.     if (isset(\$font['FontHeightOffset'])) {       // For CourierNew from Windows this needs to be -646 to match the       // Adobe native Courier font.       //       // For FreeMono from GNU this needs to be -337 to match the       // Courier font.       //       // Both have been added manually to the .afm and .ufm files.       \$h += (int)\$font['FontHeightOffset'];     }      return \$size*\$h/1000;   }    function getFontXHeight(\$size) {     if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      \$font = \$this->fonts[\$this->currentFont];      // for the current font, and the given size, what is the height of the font in user units     if ( isset(\$font['XHeight']) ) {       \$xh = \$font['Ascender']-\$font['Descender'];     }     else {       \$xh = \$this->getFontHeight(\$size) / 2;     }      return \$size*\$xh/1000;   }    /**    * return the font descender, this will normally return a negative number    * if you add this number to the baseline, you get the level of the bottom of the font    * it is in the pdf user units    */   function getFontDescender(\$size) {     // note that this will most likely return a negative value     if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      //\$h = \$this->fonts[\$this->currentFont]['FontBBox'][1];     \$h = \$this->fonts[\$this->currentFont]['Descender'];      return \$size*\$h/1000;   }    /**    * filter the text, this is applied to all text just before being inserted into the pdf document    * it escapes the various things that need to be escaped, and so on    *    * @access private    */   function filterText(\$text, \$bom = true, \$convert_encoding = true) {     if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      if (\$convert_encoding) {       \$cf = \$this->currentFont;       if (isset(\$this->fonts[\$cf]) && \$this->fonts[\$cf]['isUnicode']) {         //\$text = html_entity_decode(\$text, ENT_QUOTES, 'UTF-8');         \$text = \$this->utf8toUtf16BE(\$text, \$bom);       } else {         //\$text = html_entity_decode(\$text, ENT_QUOTES);         \$text = mb_convert_encoding(\$text, self::\$targetEncoding, 'UTF-8');       }     }      // the chr(13) substitution fixes a bug seen in TCPDF (bug #1421290)     return strtr(\$text, array(')' => '\\)', '(' => '\\(', '\\' => '\\\\', chr(13) => '\r'));   }    /**    * return array containing codepoints (UTF-8 character values) for the    * string passed in.    *    * based on the excellent TCPDF code by Nicola Asuni and the    * RFC for UTF-8 at http://www.faqs.org/rfcs/rfc3629.html    *    * @access private    * @author Orion Richardson    * @since January 5, 2008    * @param string \$text UTF-8 string to process    * @return array UTF-8 codepoints array for the string    */   function utf8toCodePointsArray(&\$text) {     \$length = mb_strlen(\$text, '8bit'); // http://www.php.net/manual/en/function.mb-strlen.php#77040     \$unicode = array(); // array containing unicode values     \$bytes = array(); // array containing single character byte sequences     \$numbytes = 1; // number of octetc needed to represent the UTF-8 character      for (\$i = 0; \$i < \$length; \$i++) {       \$c = ord(\$text[\$i]); // get one string character at time       if (count(\$bytes) === 0) { // get starting octect         if (\$c <= 0x7F) {           \$unicode[] = \$c; // use the character "as is" because is ASCII           \$numbytes = 1;         } elseif ((\$c >> 0x05) === 0x06) { // 2 bytes character (0x06 = 110 BIN)           \$bytes[] = (\$c - 0xC0) << 0x06;           \$numbytes = 2;         } elseif ((\$c >> 0x04) === 0x0E) { // 3 bytes character (0x0E = 1110 BIN)           \$bytes[] = (\$c - 0xE0) << 0x0C;           \$numbytes = 3;         } elseif ((\$c >> 0x03) === 0x1E) { // 4 bytes character (0x1E = 11110 BIN)           \$bytes[] = (\$c - 0xF0) << 0x12;           \$numbytes = 4;         } else {           // use replacement character for other invalid sequences           \$unicode[] = 0xFFFD;           \$bytes = array();           \$numbytes = 1;         }       } elseif ((\$c >> 0x06) === 0x02) { // bytes 2, 3 and 4 must start with 0x02 = 10 BIN         \$bytes[] = \$c - 0x80;         if (count(\$bytes) === \$numbytes) {           // compose UTF-8 bytes to a single unicode value           \$c = \$bytes[0];           for (\$j = 1; \$j < \$numbytes; \$j++) {             \$c += (\$bytes[\$j] << ((\$numbytes - \$j - 1) * 0x06));           }           if (((\$c >= 0xD800) AND (\$c <= 0xDFFF)) OR (\$c >= 0x10FFFF)) {             // The definition of UTF-8 prohibits encoding character numbers between             // U+D800 and U+DFFF, which are reserved for use with the UTF-16             // encoding form (as surrogate pairs) and do not directly represent             // characters.             \$unicode[] = 0xFFFD; // use replacement character           } else {             \$unicode[] = \$c; // add char to array           }           // reset data for next char           \$bytes = array();           \$numbytes = 1;         }       } else {         // use replacement character for other invalid sequences         \$unicode[] = 0xFFFD;         \$bytes = array();         \$numbytes = 1;       }     }     return \$unicode;   }    /**    * convert UTF-8 to UTF-16 with an additional byte order marker    * at the front if required.    *    * based on the excellent TCPDF code by Nicola Asuni and the    * RFC for UTF-8 at http://www.faqs.org/rfcs/rfc3629.html    *    * @access private    * @author Orion Richardson    * @since January 5, 2008    * @param string \$text UTF-8 string to process    * @param boolean \$bom whether to add the byte order marker    * @return string UTF-16 result string    */   function utf8toUtf16BE(&\$text, \$bom = true) {     \$cf = \$this->currentFont;     if (!\$this->fonts[\$cf]['isUnicode']) return \$text;     \$out = \$bom ? "\xFE\xFF" : '';      \$unicode = \$this->utf8toCodePointsArray(\$text);     foreach (\$unicode as \$c) {       if (\$c === 0xFFFD) {         \$out .= "\xFF\xFD"; // replacement character       } elseif (\$c < 0x10000) {         \$out .= chr(\$c >> 0x08) . chr(\$c & 0xFF);        } else {         \$c -= 0x10000;         \$w1 = 0xD800 | (\$c >> 0x10);         \$w2 = 0xDC00 | (\$c & 0x3FF);         \$out .= chr(\$w1 >> 0x08) . chr(\$w1 & 0xFF) . chr(\$w2 >> 0x08) . chr(\$w2 & 0xFF);       }     }     return \$out;   }    /**    * given a start position and information about how text is to be laid out, calculate where    * on the page the text will end    */   private function getTextPosition(\$x, \$y, \$angle, \$size, \$wa, \$text) {     // given this information return an array containing x and y for the end position as elements 0 and 1     \$w = \$this->getTextWidth(\$size, \$text);      // need to adjust for the number of spaces in this text     \$words = explode(' ', \$text);     \$nspaces = count(\$words) -1;     \$w+= \$wa*\$nspaces;     \$a = deg2rad((float)\$angle);      return array(cos(\$a) *\$w+\$x, -sin(\$a) *\$w+\$y);   }    /**    * Callback method used by smallCaps    *     * @param array \$matches    * @return string    */   function toUpper(\$matches) {     return mb_strtoupper(\$matches[0]);   }    function concatMatches(\$matches) {     \$str = "";     foreach (\$matches as \$match) {       \$str .= \$match[0];     }     return \$str;   }    /**    * add text to the document, at a specified location, size and angle on the page    */   function registerText(\$font, \$text) {     if ( !\$this->isUnicode || in_array(mb_strtolower(basename(\$font)), self::\$coreFonts) ) {       return;     }      if ( !isset(\$this->stringSubsets[\$font]) ) {       \$this->stringSubsets[\$font] = array();     }      \$this->stringSubsets[\$font] = array_unique(array_merge(\$this->stringSubsets[\$font], \$this->utf8toCodePointsArray(\$text)));   }    /**    * add text to the document, at a specified location, size and angle on the page    */   function addText(\$x, \$y, \$size, \$text, \$angle = 0, \$wordSpaceAdjust = 0, \$charSpaceAdjust = 0, \$smallCaps = false) {     if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      \$text = str_replace(array("\r", "\n"), "", \$text);      if ( \$smallCaps ) {       preg_match_all("/(\P{Ll}+)/u", \$text, \$matches, PREG_SET_ORDER);       \$lower = \$this->concatMatches(\$matches);       d(\$lower);        preg_match_all("/(\p{Ll}+)/u", \$text, \$matches, PREG_SET_ORDER);       \$other = \$this->concatMatches(\$matches);       d(\$other);        //\$text = preg_replace_callback("/\p{Ll}/u", array(\$this, "toUpper"), \$text);     }      // if there are any open callbacks, then they should be called, to show the start of the line     if (\$this->nCallback > 0) {       for (\$i = \$this->nCallback; \$i > 0; \$i--) {         // call each function         \$info = array(           'x' => \$x,            'y' => \$y,            'angle' => \$angle,            'status' => 'sol',            'p' => \$this->callback[\$i]['p'],            'nCallback' => \$this->callback[\$i]['nCallback'],            'height' => \$this->callback[\$i]['height'],            'descender' => \$this->callback[\$i]['descender']  );          \$func = \$this->callback[\$i]['f'];         \$this->\$func(\$info);       }     }      if (\$angle == 0) {       \$this->addContent(sprintf("\nBT %.3F %.3F Td", \$x, \$y));     } else {       \$a = deg2rad((float)\$angle);       \$this->addContent(sprintf("\nBT %.3F %.3F %.3F %.3F %.3F %.3F Tm", cos(\$a), -sin(\$a), sin(\$a), cos(\$a), \$x, \$y));     }      if (\$wordSpaceAdjust != 0 || \$wordSpaceAdjust != \$this->wordSpaceAdjust) {       \$this->wordSpaceAdjust = \$wordSpaceAdjust;       \$this->addContent(sprintf(" %.3F Tw", \$wordSpaceAdjust));     }      if (\$charSpaceAdjust != 0 || \$charSpaceAdjust != \$this->charSpaceAdjust) {       \$this->charSpaceAdjust = \$charSpaceAdjust;       \$this->addContent(sprintf(" %.3F Tc", \$charSpaceAdjust));     }      \$len = mb_strlen(\$text);     \$start = 0;      if (\$start < \$len) {       \$part = \$text; // OAR - Don't need this anymore, given that \$start always equals zero.  substr(\$text, \$start);       \$place_text = \$this->filterText(\$part, false);       // modify unicode text so that extra word spacing is manually implemented (bug #)       \$cf = \$this->currentFont;       if (\$this->fonts[\$cf]['isUnicode'] && \$wordSpaceAdjust != 0) {         \$space_scale = 1000 / \$size;         //\$place_text = str_replace(' ', ') ( ) '.(\$this->getTextWidth(\$size, chr(32), \$wordSpaceAdjust)*-75).' (', \$place_text);         \$place_text = str_replace(' ', ' ) '.(-round(\$space_scale*\$wordSpaceAdjust)).' (', \$place_text);       }       \$this->addContent(" /F\$this->currentFontNum ".sprintf('%.1F Tf ', \$size));       \$this->addContent(" [(\$place_text)] TJ");     }      \$this->addContent(' ET');      // if there are any open callbacks, then they should be called, to show the end of the line     if (\$this->nCallback > 0) {       for (\$i = \$this->nCallback; \$i > 0; \$i--) {         // call each function         \$tmp = \$this->getTextPosition(\$x, \$y, \$angle, \$size, \$wordSpaceAdjust, \$text);         \$info = array(           'x' => \$tmp[0],            'y' => \$tmp[1],            'angle' => \$angle,            'status' => 'eol',            'p' => \$this->callback[\$i]['p'],            'nCallback' => \$this->callback[\$i]['nCallback'],            'height' => \$this->callback[\$i]['height'],            'descender' => \$this->callback[\$i]['descender']  );         \$func = \$this->callback[\$i]['f'];         \$this->\$func(\$info);       }     }   }    /**    * calculate how wide a given text string will be on a page, at a given size.    * this can be called externally, but is also used by the other class functions    */   function getTextWidth(\$size, \$text, \$word_spacing = 0, \$char_spacing = 0) {     static \$ord_cache = array();      // this function should not change any of the settings, though it will need to     // track any directives which change during calculation, so copy them at the start     // and put them back at the end.     \$store_currentTextState = \$this->currentTextState;      if (!\$this->numFonts) {       \$this->selectFont(\$this->defaultFont);     }      \$text = str_replace(array("\r", "\n"), "", \$text);      // converts a number or a float to a string so it can get the width     \$text = "\$text";      // hmm, this is where it all starts to get tricky - use the font information to     // calculate the width of each character, add them up and convert to user units     \$w = 0;     \$cf = \$this->currentFont;     \$current_font = \$this->fonts[\$cf];     \$space_scale = 1000 / ( \$size > 0 ? \$size : 1 );     \$n_spaces = 0;      if ( \$current_font['isUnicode']) {       // for Unicode, use the code points array to calculate width rather       // than just the string itself       \$unicode = \$this->utf8toCodePointsArray(\$text);        foreach (\$unicode as \$char) {         // check if we have to replace character         if ( isset(\$current_font['differences'][\$char])) {           \$char = \$current_font['differences'][\$char];         }          if ( isset(\$current_font['C'][\$char]) ) {           \$char_width = \$current_font['C'][\$char];            // add the character width           \$w += \$char_width;            // add additional padding for space           if ( isset(\$current_font['codeToName'][\$char]) && \$current_font['codeToName'][\$char] === 'space' ) {  // Space             \$w += \$word_spacing * \$space_scale;             \$n_spaces++;           }         }       }        // add additionnal char spacing       if ( \$char_spacing != 0 ) {         \$w += \$char_spacing * \$space_scale * (count(\$unicode) + \$n_spaces);       }      } else {       // If CPDF is in Unicode mode but the current font does not support Unicode we need to convert the character set to Windows-1252       if ( \$this->isUnicode ) {          \$text = mb_convert_encoding(\$text, 'Windows-1252', 'UTF-8');       }        \$len = mb_strlen(\$text, 'Windows-1252');        for (\$i = 0; \$i < \$len; \$i++) {         \$c = \$text[\$i];         \$char = isset(\$ord_cache[\$c]) ? \$ord_cache[\$c] : (\$ord_cache[\$c] = ord(\$c));          // check if we have to replace character         if ( isset(\$current_font['differences'][\$char])) {           \$char = \$current_font['differences'][\$char];         }          if ( isset(\$current_font['C'][\$char]) ) {           \$char_width = \$current_font['C'][\$char];            // add the character width           \$w += \$char_width;            // add additional padding for space           if ( isset(\$current_font['codeToName'][\$char]) && \$current_font['codeToName'][\$char] === 'space' ) {  // Space             \$w += \$word_spacing * \$space_scale;             \$n_spaces++;           }         }       }        // add additionnal char spacing       if ( \$char_spacing != 0 )  {         \$w += \$char_spacing * \$space_scale * (\$len + \$n_spaces);       }     }      \$this->currentTextState = \$store_currentTextState;     \$this->setCurrentFont();      return \$w*\$size/1000;   }    /**    * this will be called at a new page to return the state to what it was on the    * end of the previous page, before the stack was closed down    * This is to get around not being able to have open 'q' across pages    *    */   function saveState(\$pageEnd = 0) {     if (\$pageEnd) {       // this will be called at a new page to return the state to what it was on the       // end of the previous page, before the stack was closed down       // This is to get around not being able to have open 'q' across pages       \$opt = \$this->stateStack[\$pageEnd];       // ok to use this as stack starts numbering at 1       \$this->setColor(\$opt['col'], true);       \$this->setStrokeColor(\$opt['str'], true);       \$this->addContent("\n".\$opt['lin']);       //    \$this->currentLineStyle = \$opt['lin'];     } else {       \$this->nStateStack++;       \$this->stateStack[\$this->nStateStack] = array(         'col' => \$this->currentColor,          'str' => \$this->currentStrokeColor,          'lin' => \$this->currentLineStyle  );     }      \$this->save();   }    /**    * restore a previously saved state    */   function restoreState(\$pageEnd = 0) {     if (!\$pageEnd) {       \$n = \$this->nStateStack;       \$this->currentColor =       \$this->stateStack[\$n]['col'];       \$this->currentStrokeColor = \$this->stateStack[\$n]['str'];       \$this->addContent("\n".\$this->stateStack[\$n]['lin']);       \$this->currentLineStyle =    \$this->stateStack[\$n]['lin'];       \$this->stateStack[\$n] = null;       unset(\$this->stateStack[\$n]);       \$this->nStateStack--;     }      \$this->restore();   }    /**    * make a loose object, the output will go into this object, until it is closed, then will revert to    * the current one.    * this object will not appear until it is included within a page.    * the function will return the object number    */   function openObject() {     \$this->nStack++;     \$this->stack[\$this->nStack] = array('c' => \$this->currentContents, 'p' => \$this->currentPage);     // add a new object of the content type, to hold the data flow     \$this->numObj++;     \$this->o_contents(\$this->numObj, 'new');     \$this->currentContents = \$this->numObj;     \$this->looseObjects[\$this->numObj] = 1;      return \$this->numObj;   }    /**    * open an existing object for editing    */   function reopenObject(\$id) {     \$this->nStack++;     \$this->stack[\$this->nStack] = array('c' => \$this->currentContents, 'p' => \$this->currentPage);     \$this->currentContents = \$id;      // also if this object is the primary contents for a page, then set the current page to its parent     if (isset(\$this->objects[\$id]['onPage'])) {       \$this->currentPage = \$this->objects[\$id]['onPage'];     }   }    /**    * close an object    */   function closeObject() {     // close the object, as long as there was one open in the first place, which will be indicated by     // an objectId on the stack.     if (\$this->nStack > 0) {       \$this->currentContents = \$this->stack[\$this->nStack]['c'];       \$this->currentPage = \$this->stack[\$this->nStack]['p'];       \$this->nStack--;       // easier to probably not worry about removing the old entries, they will be overwritten       // if there are new ones.     }   }    /**    * stop an object from appearing on pages from this point on    */   function stopObject(\$id) {     // if an object has been appearing on pages up to now, then stop it, this page will     // be the last one that could contian it.     if (isset(\$this->addLooseObjects[\$id])) {       \$this->addLooseObjects[\$id] = '';     }   }    /**    * after an object has been created, it wil only show if it has been added, using this function.    */   function addObject(\$id, \$options = 'add') {     // add the specified object to the page     if (isset(\$this->looseObjects[\$id]) && \$this->currentContents != \$id) {       // then it is a valid object, and it is not being added to itself       switch (\$options) {       case 'all':         // then this object is to be added to this page (done in the next block) and         // all future new pages.         \$this->addLooseObjects[\$id] = 'all';        case 'add':         if (isset(\$this->objects[\$this->currentContents]['onPage'])) {           // then the destination contents is the primary for the page           // (though this object is actually added to that page)           \$this->o_page(\$this->objects[\$this->currentContents]['onPage'], 'content', \$id);         }         break;        case 'even':         \$this->addLooseObjects[\$id] = 'even';         \$pageObjectId = \$this->objects[\$this->currentContents]['onPage'];         if (\$this->objects[\$pageObjectId]['info']['pageNum']%2 == 0) {           \$this->addObject(\$id);           // hacky huh :)         }         break;        case 'odd':         \$this->addLooseObjects[\$id] = 'odd';         \$pageObjectId = \$this->objects[\$this->currentContents]['onPage'];         if (\$this->objects[\$pageObjectId]['info']['pageNum']%2 == 1) {           \$this->addObject(\$id);           // hacky huh :)         }         break;        case 'next':         \$this->addLooseObjects[\$id] = 'all';         break;        case 'nexteven':         \$this->addLooseObjects[\$id] = 'even';         break;        case 'nextodd':         \$this->addLooseObjects[\$id] = 'odd';         break;       }     }   }    /**    * return a storable representation of a specific object    */   function serializeObject(\$id) {     if ( array_key_exists(\$id, \$this->objects)) {       return serialize(\$this->objects[\$id]);     }   }    /**    * restore an object from its stored representation.  returns its new object id.    */   function restoreSerializedObject(\$obj) {     \$obj_id = \$this->openObject();     \$this->objects[\$obj_id] = unserialize(\$obj);     \$this->closeObject();     return \$obj_id;   }    /**    * add content to the documents info object    */   function addInfo(\$label, \$value = 0) {     // this will only work if the label is one of the valid ones.     // modify this so that arrays can be passed as well.     // if \$label is an array then assume that it is key => value pairs     // else assume that they are both scalar, anything else will probably error     if (is_array(\$label)) {       foreach (\$label as \$l => \$v) {         \$this->o_info(\$this->infoObject, \$l, \$v);       }     } else {       \$this->o_info(\$this->infoObject, \$label, \$value);     }   }    /**    * set the viewer preferences of the document, it is up to the browser to obey these.    */   function setPreferences(\$label, \$value = 0) {     // this will only work if the label is one of the valid ones.     if (is_array(\$label)) {       foreach (\$label as \$l => \$v) {         \$this->o_catalog(\$this->catalogId, 'viewerPreferences', array(\$l => \$v));       }     } else {       \$this->o_catalog(\$this->catalogId, 'viewerPreferences', array(\$label => \$value));     }   }    /**    * extract an integer from a position in a byte stream    */   private function getBytes(&\$data, \$pos, \$num) {     // return the integer represented by \$num bytes from \$pos within \$data     \$ret = 0;     for (\$i = 0; \$i < \$num; \$i++) {       \$ret *= 256;       \$ret += ord(\$data[\$pos+\$i]);     }      return \$ret;   }    /**    * Check if image already added to pdf image directory.    * If yes, need not to create again (pass empty data)    */   function image_iscached(\$imgname) {     return isset(\$this->imagelist[\$imgname]);   }    /**    * add a PNG image into the document, from a GD object    * this should work with remote files    *    * @param string   \$file    The PNG file    * @param float    \$x       X position    * @param float    \$y       Y position    * @param float    \$w       Width    * @param float    \$h       Height    * @param resource \$img     A GD resource    * @param bool     \$is_mask true if the image is a mask    * @param bool     \$mask    true if the image is masked    */   function addImagePng(\$file, \$x, \$y, \$w = 0.0, \$h = 0.0, &\$img, \$is_mask = false, \$mask = null) {     if (!function_exists("imagepng")) {       throw new Exception("The PHP GD extension is required, but is not installed.");     }      //if already cached, need not to read again     if ( isset(\$this->imagelist[\$file]) ) {       \$data = null;     }     else {       // Example for transparency handling on new image. Retain for current image       // \$tIndex = imagecolortransparent(\$img);       // if (\$tIndex > 0) {       //   \$tColor = imagecolorsforindex(\$img, \$tIndex);       //   \$new_tIndex = imagecolorallocate(\$new_img, \$tColor['red'], \$tColor['green'], \$tColor['blue']);       //   imagefill(\$new_img, 0, 0, \$new_tIndex);       //   imagecolortransparent(\$new_img, \$new_tIndex);       // }       // blending mode (literal/blending) on drawing into current image. not relevant when not saved or not drawn       //imagealphablending(\$img, true);        //default, but explicitely set to ensure pdf compatibility       imagesavealpha(\$img, false/**!\$is_mask && !\$mask*/);        \$error = 0;       //DEBUG_IMG_TEMP       //debugpng       if (DEBUGPNG) print '[addImagePng '.\$file.']';        ob_start();       @imagepng(\$img);       \$data = ob_get_clean();        if (\$data == '') {         \$error = 1;         \$errormsg = 'trouble writing file from GD';         //DEBUG_IMG_TEMP         //debugpng         if (DEBUGPNG) print 'trouble writing file from GD';       }        if (\$error) {         \$this->addMessage('PNG error - ('.\$file.') '.\$errormsg);         return;       }     }  //End isset(\$this->imagelist[\$file]) (png Duplicate removal)      \$this->addPngFromBuf(\$file, \$x, \$y, \$w, \$h, \$data, \$is_mask, \$mask);   }    protected function addImagePngAlpha(\$file, \$x, \$y, \$w, \$h, \$byte) {     // generate images     \$img = imagecreatefrompng(\$file);      if (\$img === false) {       return;     }      // FIXME The pixel transformation doesn't work well with 8bit PNGs     \$eight_bit = (\$byte & 4) !== 4;      \$wpx = imagesx(\$img);     \$hpx = imagesy(\$img);      imagesavealpha(\$img, false);      // create temp alpha file     \$tempfile_alpha = tempnam(\$this->tmp, "cpdf_img_");     @unlink(\$tempfile_alpha);     \$tempfile_alpha = "\$tempfile_alpha.png";      // create temp plain file     \$tempfile_plain = tempnam(\$this->tmp, "cpdf_img_");     @unlink(\$tempfile_plain);     \$tempfile_plain = "\$tempfile_plain.png";      \$imgalpha = imagecreate(\$wpx, \$hpx);     imagesavealpha(\$imgalpha, false);      // generate gray scale palette (0 -> 255)     for (\$c = 0; \$c < 256; ++\$c) {       imagecolorallocate(\$imgalpha, \$c, \$c, \$c);     }      // Use PECL gmagick + Graphics Magic to process transparent PNG images     if (extension_loaded("gmagick")) {       \$gmagick = new Gmagick(\$file);       \$gmagick->setimageformat('png');        // Get opacity channel (negative of alpha channel)       \$alpha_channel_neg = clone \$gmagick;       \$alpha_channel_neg->separateimagechannel(Gmagick::CHANNEL_OPACITY);        // Negate opacity channel       \$alpha_channel = new Gmagick();       \$alpha_channel->newimage(\$wpx, \$hpx, "#FFFFFF", "png");       \$alpha_channel->compositeimage(\$alpha_channel_neg, Gmagick::COMPOSITE_DIFFERENCE, 0, 0);       \$alpha_channel->separateimagechannel(Gmagick::CHANNEL_RED);       \$alpha_channel->writeimage(\$tempfile_alpha);        // Cast to 8bit+palette       \$imgalpha_ = imagecreatefrompng(\$tempfile_alpha);       imagecopy(\$imgalpha, \$imgalpha_, 0, 0, 0, 0, \$wpx, \$hpx);       imagedestroy(\$imgalpha_);       imagepng(\$imgalpha, \$tempfile_alpha);        // Make opaque image       \$color_channels = new Gmagick();       \$color_channels->newimage(\$wpx, \$hpx, "#FFFFFF", "png");       \$color_channels->compositeimage(\$gmagick, Gmagick::COMPOSITE_COPYRED, 0, 0);       \$color_channels->compositeimage(\$gmagick, Gmagick::COMPOSITE_COPYGREEN, 0, 0);       \$color_channels->compositeimage(\$gmagick, Gmagick::COMPOSITE_COPYBLUE, 0, 0);       \$color_channels->writeimage(\$tempfile_plain);        \$imgplain = imagecreatefrompng(\$tempfile_plain);     }      // Use PECL imagick + ImageMagic to process transparent PNG images     elseif (extension_loaded("imagick")) {       // Native cloning was added to pecl-imagick in svn commit 263814       // the first version containing it was 3.0.1RC1       static \$imagickClonable = null;       if(\$imagickClonable === null) {         \$imagickClonable = version_compare(phpversion('imagick'), '3.0.1rc1') > 0;       }        \$imagick = new Imagick(\$file);       \$imagick->setFormat('png');        // Get opacity channel (negative of alpha channel)       \$alpha_channel = \$imagickClonable ? clone \$imagick : \$imagick->clone();       \$alpha_channel->separateImageChannel(Imagick::CHANNEL_ALPHA);       \$alpha_channel->negateImage(true);       \$alpha_channel->writeImage(\$tempfile_alpha);        // Cast to 8bit+palette       \$imgalpha_ = imagecreatefrompng(\$tempfile_alpha);       imagecopy(\$imgalpha, \$imgalpha_, 0, 0, 0, 0, \$wpx, \$hpx);       imagedestroy(\$imgalpha_);       imagepng(\$imgalpha, \$tempfile_alpha);        // Make opaque image       \$color_channels = new Imagick();       \$color_channels->newImage(\$wpx, \$hpx, "#FFFFFF", "png");       \$color_channels->compositeImage(\$imagick, Imagick::COMPOSITE_COPYRED, 0, 0);       \$color_channels->compositeImage(\$imagick, Imagick::COMPOSITE_COPYGREEN, 0, 0);       \$color_channels->compositeImage(\$imagick, Imagick::COMPOSITE_COPYBLUE, 0, 0);       \$color_channels->writeImage(\$tempfile_plain);        \$imgplain = imagecreatefrompng(\$tempfile_plain);     }     else {       // allocated colors cache       \$allocated_colors = array();        // extract alpha channel       for (\$xpx = 0; \$xpx < \$wpx; ++\$xpx) {         for (\$ypx = 0; \$ypx < \$hpx; ++\$ypx) {           \$color = imagecolorat(\$img, \$xpx, \$ypx);           \$col = imagecolorsforindex(\$img, \$color);           \$alpha = \$col['alpha'];            if (\$eight_bit) {             // with gamma correction             \$gammacorr = 2.2;             \$pixel = pow((((127 - \$alpha) * 255 / 127) / 255), \$gammacorr) * 255;           }            else {             // without gamma correction             \$pixel = (127 - \$alpha) * 2;              \$key = \$col['red'].\$col['green'].\$col['blue'];              if (!isset(\$allocated_colors[\$key])) {               \$pixel_img = imagecolorallocate(\$img, \$col['red'], \$col['green'], \$col['blue']);               \$allocated_colors[\$key] = \$pixel_img;             }             else {               \$pixel_img = \$allocated_colors[\$key];              }              imagesetpixel(\$img, \$xpx, \$ypx, \$pixel_img);           }            imagesetpixel(\$imgalpha, \$xpx, \$ypx, \$pixel);         }       }        // extract image without alpha channel       \$imgplain = imagecreatetruecolor(\$wpx, \$hpx);       imagecopy(\$imgplain, \$img, 0, 0, 0, 0, \$wpx, \$hpx);       imagedestroy(\$img);        imagepng(\$imgalpha, \$tempfile_alpha);       imagepng(\$imgplain, \$tempfile_plain);     }      // embed mask image     \$this->addImagePng(\$tempfile_alpha, \$x, \$y, \$w, \$h, \$imgalpha, true);     imagedestroy(\$imgalpha);      // embed image, masked with previously embedded mask     \$this->addImagePng(\$tempfile_plain, \$x, \$y, \$w, \$h, \$imgplain, false, true);     imagedestroy(\$imgplain);      // remove temp files     unlink(\$tempfile_alpha);     unlink(\$tempfile_plain);   }    /**    * add a PNG image into the document, from a file    * this should work with remote files    */   function addPngFromFile(\$file, \$x, \$y, \$w = 0, \$h = 0) {     if (!function_exists("imagecreatefrompng")) {       throw new Exception("The PHP GD extension is required, but is not installed.");     }      //if already cached, need not to read again     if ( isset(\$this->imagelist[\$file]) ) {       \$img = null;     }       else {       \$info = file_get_contents (\$file, false, null, 24, 5);       \$meta = unpack("CbitDepth/CcolorType/CcompressionMethod/CfilterMethod/CinterlaceMethod", \$info);       \$bit_depth = \$meta["bitDepth"];       \$color_type = \$meta["colorType"];        // http://www.w3.org/TR/PNG/#11IHDR       // 3 => indexed       // 4 => greyscale with alpha       // 6 => fullcolor with alpha       \$is_alpha = in_array(\$color_type, array(4, 6)) || (\$color_type == 3 && \$bit_depth != 4);        if (\$is_alpha) { // exclude grayscale alpha         return \$this->addImagePngAlpha(\$file, \$x, \$y, \$w, \$h, \$color_type);       }        //png files typically contain an alpha channel.       //pdf file format or class.pdf does not support alpha blending.       //on alpha blended images, more transparent areas have a color near black.       //This appears in the result on not storing the alpha channel.       //Correct would be the box background image or its parent when transparent.       //But this would make the image dependent on the background.       //Therefore create an image with white background and copy in       //A more natural background than black is white.       //Therefore create an empty image with white background and merge the       //image in with alpha blending.       \$imgtmp = @imagecreatefrompng(\$file);       if (!\$imgtmp) {         return;       }       \$sx = imagesx(\$imgtmp);       \$sy = imagesy(\$imgtmp);       \$img = imagecreatetruecolor(\$sx, \$sy);       imagealphablending(\$img, true);        // @todo is it still needed ??       \$ti = imagecolortransparent(\$imgtmp);       if (\$ti >= 0) {         \$tc = imagecolorsforindex(\$imgtmp, \$ti);         \$ti = imagecolorallocate(\$img, \$tc['red'], \$tc['green'], \$tc['blue']);         imagefill(\$img, 0, 0, \$ti);         imagecolortransparent(\$img, \$ti);       } else {         imagefill(\$img, 1, 1, imagecolorallocate(\$img, 255, 255, 255));       }        imagecopy(\$img, \$imgtmp, 0, 0, 0, 0, \$sx, \$sy);       imagedestroy(\$imgtmp);     }     \$this->addImagePng(\$file, \$x, \$y, \$w, \$h, \$img);      if ( \$img ) {       imagedestroy(\$img);     }   }    /**    * add a PNG image into the document, from a memory buffer of the file    */   function addPngFromBuf(\$file, \$x, \$y, \$w = 0.0, \$h = 0.0, &\$data, \$is_mask = false, \$mask = null) {     if ( isset(\$this->imagelist[\$file]) ) {       \$data = null;       \$info['width'] = \$this->imagelist[\$file]['w'];       \$info['height'] = \$this->imagelist[\$file]['h'];       \$label = \$this->imagelist[\$file]['label'];     }      else {       if (\$data == null) {         \$this->addMessage('addPngFromBuf error - data not present!');         return;       }        \$error = 0;        if (!\$error) {         \$header = chr(137) .chr(80) .chr(78) .chr(71) .chr(13) .chr(10) .chr(26) .chr(10);          if (mb_substr(\$data, 0, 8, '8bit') != \$header) {           \$error = 1;            if (DEBUGPNG) print '[addPngFromFile this file does not have a valid header '.\$file.']';            \$errormsg = 'this file does not have a valid header';         }       }        if (!\$error) {         // set pointer         \$p = 8;         \$len = mb_strlen(\$data, '8bit');          // cycle through the file, identifying chunks         \$haveHeader = 0;         \$info = array();         \$idata = '';         \$pdata = '';          while (\$p < \$len) {           \$chunkLen = \$this->getBytes(\$data, \$p, 4);           \$chunkType = mb_substr(\$data, \$p+4, 4, '8bit');            switch (\$chunkType) {           case 'IHDR':             // this is where all the file information comes from             \$info['width'] = \$this->getBytes(\$data, \$p+8, 4);             \$info['height'] = \$this->getBytes(\$data, \$p+12, 4);             \$info['bitDepth'] = ord(\$data[\$p+16]);             \$info['colorType'] = ord(\$data[\$p+17]);             \$info['compressionMethod'] = ord(\$data[\$p+18]);             \$info['filterMethod'] = ord(\$data[\$p+19]);             \$info['interlaceMethod'] = ord(\$data[\$p+20]);              //print_r(\$info);             \$haveHeader = 1;             if (\$info['compressionMethod'] != 0) {               \$error = 1;                //debugpng               if (DEBUGPNG) print '[addPngFromFile unsupported compression method '.\$file.']';                \$errormsg = 'unsupported compression method';             }              if (\$info['filterMethod'] != 0) {               \$error = 1;                //debugpng               if (DEBUGPNG) print '[addPngFromFile unsupported filter method '.\$file.']';                \$errormsg = 'unsupported filter method';             }             break;            case 'PLTE':             \$pdata.= mb_substr(\$data, \$p+8, \$chunkLen, '8bit');             break;            case 'IDAT':             \$idata.= mb_substr(\$data, \$p+8, \$chunkLen, '8bit');             break;            case 'tRNS':             //this chunk can only occur once and it must occur after the PLTE chunk and before IDAT chunk             //print "tRNS found, color type = ".\$info['colorType']."\n";             \$transparency = array();              switch (\$info['colorType']) {               // indexed color, rbg               case 3:                 /** corresponding to entries in the plte chunk                  Alpha for palette index 0: 1 byte                  Alpha for palette index 1: 1 byte                  ...etc...                 */                 // there will be one entry for each palette entry. up until the last non-opaque entry.                 // set up an array, stretching over all palette entries which will be o (opaque) or 1 (transparent)                 \$transparency['type'] = 'indexed';                 \$trans = 0;                  for (\$i = \$chunkLen; \$i >= 0; \$i--) {                   if (ord(\$data[\$p+8+\$i]) == 0) {                     \$trans = \$i;                   }                 }                  \$transparency['data'] = \$trans;               break;                // grayscale               case 0:                 /** corresponding to entries in the plte chunk                  Gray: 2 bytes, range 0 .. (2^bitdepth)-1                 */                 //            \$transparency['grayscale'] = \$this->PRVT_getBytes(\$data, \$p+8, 2); // g = grayscale                 \$transparency['type'] = 'indexed';                 \$transparency['data'] = ord(\$data[\$p+8+1]);               break;                // truecolor               case 2:                       /** corresponding to entries in the plte chunk                  Red: 2 bytes, range 0 .. (2^bitdepth)-1                  Green: 2 bytes, range 0 .. (2^bitdepth)-1                  Blue: 2 bytes, range 0 .. (2^bitdepth)-1                 */                 \$transparency['r'] = \$this->getBytes(\$data, \$p+8, 2);                 // r from truecolor                 \$transparency['g'] = \$this->getBytes(\$data, \$p+10, 2);                 // g from truecolor                 \$transparency['b'] = \$this->getBytes(\$data, \$p+12, 2);                 // b from truecolor                  \$transparency['type'] = 'color-key';               break;                //unsupported transparency type               default:                 if (DEBUGPNG) print '[addPngFromFile unsupported transparency type '.\$file.']';                 break;             }              // KS End new code             break;            default:             break;           }            \$p += \$chunkLen+12;         }          if (!\$haveHeader) {           \$error = 1;            //debugpng           if (DEBUGPNG) print '[addPngFromFile information header is missing '.\$file.']';            \$errormsg = 'information header is missing';         }          if (isset(\$info['interlaceMethod']) && \$info['interlaceMethod']) {           \$error = 1;            //debugpng           if (DEBUGPNG) print '[addPngFromFile no support for interlaced images in pdf '.\$file.']';            \$errormsg = 'There appears to be no support for interlaced images in pdf.';         }       }        if (!\$error && \$info['bitDepth'] > 8) {         \$error = 1;          //debugpng         if (DEBUGPNG) print '[addPngFromFile bit depth of 8 or less is supported '.\$file.']';          \$errormsg = 'only bit depth of 8 or less is supported';       }          if (!\$error) {         switch (\$info['colorType']) {         case 3:           \$color = 'DeviceRGB';           \$ncolor = 1;           break;          case 2:           \$color = 'DeviceRGB';           \$ncolor = 3;           break;          case 0:           \$color = 'DeviceGray';           \$ncolor = 1;           break;          default:            \$error = 1;            //debugpng           if (DEBUGPNG) print '[addPngFromFile alpha channel not supported: '.\$info['colorType'].' '.\$file.']';            \$errormsg = 'transparancey alpha channel not supported, transparency only supported for palette images.';         }       }        if (\$error) {         \$this->addMessage('PNG error - ('.\$file.') '.\$errormsg);         return;       }        //print_r(\$info);       // so this image is ok... add it in.       \$this->numImages++;       \$im = \$this->numImages;       \$label = "I\$im";       \$this->numObj++;        //  \$this->o_image(\$this->numObj, 'new', array('label' => \$label, 'data' => \$idata, 'iw' => \$w, 'ih' => \$h, 'type' => 'png', 'ic' => \$info['width']));       \$options = array(         'label' => \$label,          'data' => \$idata,          'bitsPerComponent' => \$info['bitDepth'],          'pdata' => \$pdata,          'iw' => \$info['width'],          'ih' => \$info['height'],          'type' => 'png',          'color' => \$color,          'ncolor' => \$ncolor,          'masked' => \$mask,          'isMask' => \$is_mask  );        if (isset(\$transparency)) {         \$options['transparency'] = \$transparency;       }        \$this->o_image(\$this->numObj, 'new', \$options);       \$this->imagelist[\$file] = array('label' =>\$label, 'w' => \$info['width'], 'h' => \$info['height']);     }      if (\$is_mask) {       return;     }      if (\$w <= 0 && \$h <= 0) {       \$w = \$info['width'];       \$h = \$info['height'];     }      if (\$w <= 0) {       \$w = \$h/\$info['height']*\$info['width'];     }      if (\$h <= 0) {       \$h = \$w*\$info['height']/\$info['width'];     }      \$this->addContent(sprintf("\nq\n%.3F 0 0 %.3F %.3F %.3F cm /%s Do\nQ", \$w, \$h, \$x, \$y, \$label));   }    /**    * add a JPEG image into the document, from a file    */   function addJpegFromFile(\$img, \$x, \$y, \$w = 0, \$h = 0) {     // attempt to add a jpeg image straight from a file, using no GD commands     // note that this function is unable to operate on a remote file.      if (!file_exists(\$img)) {       return;     }      if ( \$this->image_iscached(\$img) ) {       \$data = null;       \$imageWidth = \$this->imagelist[\$img]['w'];       \$imageHeight = \$this->imagelist[\$img]['h'];       \$channels = \$this->imagelist[\$img]['c'];     }      else {       \$tmp = getimagesize(\$img);       \$imageWidth = \$tmp[0];       \$imageHeight = \$tmp[1];        if ( isset(\$tmp['channels']) ) {         \$channels = \$tmp['channels'];       } else {         \$channels = 3;       }        \$data = file_get_contents(\$img);     }      if (\$w <= 0 && \$h <= 0) {       \$w = \$imageWidth;     }      if (\$w == 0) {       \$w = \$h/\$imageHeight*\$imageWidth;     }      if (\$h == 0) {       \$h = \$w*\$imageHeight/\$imageWidth;     }      \$this->addJpegImage_common(\$data, \$x, \$y, \$w, \$h, \$imageWidth, \$imageHeight, \$channels, \$img);   }    /**    * common code used by the two JPEG adding functions    */   private function addJpegImage_common(&\$data, \$x, \$y, \$w = 0, \$h = 0, \$imageWidth, \$imageHeight, \$channels = 3, \$imgname) {     if ( \$this->image_iscached(\$imgname) ) {       \$label = \$this->imagelist[\$imgname]['label'];       //debugpng       //if (DEBUGPNG) print '[addJpegImage_common Duplicate '.\$imgname.']';      } else {       if (\$data == null) {         \$this->addMessage('addJpegImage_common error - ('.\$imgname.') data not present!');         return;       }        // note that this function is not to be called externally       // it is just the common code between the GD and the file options       \$this->numImages++;       \$im = \$this->numImages;       \$label = "I\$im";       \$this->numObj++;        \$this->o_image(\$this->numObj, 'new', array(         'label' => \$label,          'data' => &\$data,          'iw' => \$imageWidth,          'ih' => \$imageHeight,          'channels' => \$channels  ));        \$this->imagelist[\$imgname] = array('label' =>\$label, 'w' => \$imageWidth, 'h' => \$imageHeight, 'c'=> \$channels);     }      \$this->addContent(sprintf("\nq\n%.3F 0 0 %.3F %.3F %.3F cm /%s Do\nQ ", \$w, \$h, \$x, \$y, \$label));   }    /**    * specify where the document should open when it first starts    */   function openHere(\$style, \$a = 0, \$b = 0, \$c = 0) {     // this function will open the document at a specified page, in a specified style     // the values for style, and the required paramters are:     // 'XYZ'  left, top, zoom     // 'Fit'     // 'FitH' top     // 'FitV' left     // 'FitR' left, bottom, right     // 'FitB'     // 'FitBH' top     // 'FitBV' left     \$this->numObj++;     \$this->o_destination(\$this->numObj, 'new', array('page' => \$this->currentPage, 'type' => \$style, 'p1' => \$a, 'p2' => \$b, 'p3' => \$c));     \$id = \$this->catalogId;     \$this->o_catalog(\$id, 'openHere', \$this->numObj);   }    /**    * Add JavaScript code to the PDF document    *     * @param string \$code    * @return void    */   function addJavascript(\$code) {     \$this->javascript .= \$code;   }    /**    * create a labelled destination within the document    */   function addDestination(\$label, \$style, \$a = 0, \$b = 0, \$c = 0) {     // associates the given label with the destination, it is done this way so that a destination can be specified after     // it has been linked to     // styles are the same as the 'openHere' function     \$this->numObj++;     \$this->o_destination(\$this->numObj, 'new', array('page' => \$this->currentPage, 'type' => \$style, 'p1' => \$a, 'p2' => \$b, 'p3' => \$c));     \$id = \$this->numObj;      // store the label->idf relationship, note that this means that labels can be used only once     \$this->destinations["\$label"] = \$id;   }    /**    * define font families, this is used to initialize the font families for the default fonts    * and for the user to add new ones for their fonts. The default bahavious can be overridden should    * that be desired.    */   function setFontFamily(\$family, \$options = '') {     if (!is_array(\$options)) {       if (\$family === 'init') {         // set the known family groups         // these font families will be used to enable bold and italic markers to be included         // within text streams. html forms will be used... <b></b> <i></i>         \$this->fontFamilies['Helvetica.afm'] =           array(             'b' => 'Helvetica-Bold.afm',              'i' => 'Helvetica-Oblique.afm',              'bi' => 'Helvetica-BoldOblique.afm',              'ib' => 'Helvetica-BoldOblique.afm'  );          \$this->fontFamilies['Courier.afm'] =           array(             'b' => 'Courier-Bold.afm',              'i' => 'Courier-Oblique.afm',              'bi' => 'Courier-BoldOblique.afm',              'ib' => 'Courier-BoldOblique.afm'  );          \$this->fontFamilies['Times-Roman.afm'] =           array(             'b' => 'Times-Bold.afm',              'i' => 'Times-Italic.afm',              'bi' => 'Times-BoldItalic.afm',              'ib' => 'Times-BoldItalic.afm'  );       }     } else {        // the user is trying to set a font family       // note that this can also be used to set the base ones to something else       if (mb_strlen(\$family)) {         \$this->fontFamilies[\$family] = \$options;       }     }   }    /**    * used to add messages for use in debugging    */   function addMessage(\$message) {     \$this->messages.=  \$message."\n";   }    /**    * a few functions which should allow the document to be treated transactionally.    */   function transaction(\$action) {     switch (\$action) {     case 'start':       // store all the data away into the checkpoint variable       \$data = get_object_vars(\$this);       \$this->checkpoint = \$data;       unset(\$data);       break;      case 'commit':       if (is_array(\$this->checkpoint) && isset(\$this->checkpoint['checkpoint'])) {         \$tmp = \$this->checkpoint['checkpoint'];         \$this->checkpoint = \$tmp;         unset(\$tmp);       } else {         \$this->checkpoint = '';       }       break;      case 'rewind':       // do not destroy the current checkpoint, but move us back to the state then, so that we can try again       if (is_array(\$this->checkpoint)) {         // can only abort if were inside a checkpoint         \$tmp = \$this->checkpoint;          foreach (\$tmp as \$k => \$v) {           if (\$k !== 'checkpoint') {             \$this->\$k = \$v;           }         }         unset(\$tmp);       }       break;      case 'abort':       if (is_array(\$this->checkpoint)) {         // can only abort if were inside a checkpoint         \$tmp = \$this->checkpoint;         foreach (\$tmp as \$k => \$v) {           \$this->\$k = \$v;         }         unset(\$tmp);       }       break;     }   } } `