public function Visualize($table_data, $font_size)
{
$ww_table = $this->wordWrapTable($table_data, $font_size, true);
$rows = count($ww_table);
$cols = array_keys($ww_table[0]);
$col_size = array();
if ($this->encoding->useMB()) {
$plus = mb_convert_encoding("+", $this->encoding->getEncodingType());
$newline = mb_convert_encoding("\n", $this->encoding->getEncodingType());
$dash = mb_convert_encoding("-", $this->encoding->getEncodingType());
$space = mb_convert_encoding(" ", $this->encoding->getEncodingType());
$vert = mb_convert_encoding('|', $this->encoding->getEncodingType());
$out = mb_convert_encoding("", $this->encoding->getEncodingType());
} else {
$plus = "+";
$newline = "\n";
$dash = "-";
$space = " ";
$vert = '|';
$out = "";
}
$lengths = array();
for ($r = 0; $r < $rows; $r++) {
$lengths[$r] = array();
}
foreach ($cols as $col) {
$max = 0;
for ($r = 0; $r < $rows; $r++) {
$z = count($ww_table[$r][$col]);
for ($i = 0; $i < $z; $i++) {
$line = $ww_table[$r][$col][$i];
if ($this->encoding->useMB()) {
$w = mb_strlen($line, $this->encoding->getEncodingType());
} else {
$w = strlen($line);
}
$lengths[$r][$col][$i] = $w;
if ($w > $max) {
$max = $w;
}
}
}
$col_size[$col] = $max;
}
//now do the display
for ($r = 0; $r < $rows; $r++) {
foreach ($cols as $col) {
$out .= $plus;
for ($j = 0; $j < $col_size[$col]; $j++) {
$out .= $dash;
}
}
$out .= $plus . $newline;
$z = count($ww_table[$r][$cols[0]]);
for ($i = 0; $i < $z; $i++) {
foreach ($cols as $col) {
$out .= $vert . $ww_table[$r][$col][$i];
for ($j = $lengths[$r][$col][$i]; $j < $col_size[$col]; $j++) {
$out .= $space;
}
}
$out .= $vert . $newline;
}
}
foreach ($cols as $col) {
$out .= $plus;
for ($j = 0; $j < $col_size[$col]; $j++) {
$out .= $dash;
}
}
$out .= $plus . $newline;
return $out;
}
/**
* Get the parts of a word which breaks along hyphenation points or any non-letter.
* @param string $word the word we wish to break up
* @param bool $supress true (default)to suppress hyphenation points at the beginning/end of a word.
* @returns an the associative array has
* a string 'Subword' which tells what the subword is, the int 'Offset' tells where the subword started,
* the int 'Length' the length of the subword, and the boolean 'IsLetter' which tells us if the
* subword is a composed of letters (by the Unicode convention) or not.
*/
public function getWordParts($word, $supress = true)
{
$word_parts = array();
$mb_encoding = $this->enc->getEncodingType();
$use_mb = $this->enc->useMB();
mb_regex_encoding($mb_encoding);
if ($use_mb) {
$word_len = mb_strlen($word, $mb_encoding);
} else {
$word_len = strlen($word);
}
if ($mb_encoding != 'UTF-8') {
if ($mb_encoding) {
$utf8_word = mb_convert_encoding($word, 'UTF-8', $mb_encoding);
} else {
$utf8_word = mb_convert_encoding($word, 'UTF-8');
}
} else {
$utf8_word = $word;
}
$i = 0;
$prev_i = 0;
$c = '';
do {
$sub_word_len = 0;
$is_letter = true;
while ($i < $word_len && $is_letter) {
$c = mb_substr($utf8_word, 0, 1, 'UTF-8');
//get the first character
$utf8_word = mb_substr($utf8_word, 1);
//delete the first character
if (!preg_match('/^\\p{L}/', $c)) {
//the character is not a letter
$is_letter = false;
}
$sub_word_len++;
$i++;
}
if (!$is_letter) {
//the last character we read was a non-letter
$sub_word_len--;
}
if ($use_mb) {
$subword = mb_substr($word, $prev_i, $sub_word_len, $mb_encoding);
} else {
$subword = substr($word, $prev_i, $sub_word_len);
}
if ($sub_word_len > 0) {
//get the hyphenation points for the word.
$hp = $this->HyphenateWord($subword, $mb_encoding, $supress);
$num_hp = count($hp);
if ($use_mb) {
for ($k = 0; $k < $num_hp - 1; $k++) {
$word_parts[] = array('Offset' => $prev_i + $hp[$k], 'Length' => $hp[$k + 1] - $hp[$k], 'Subword' => mb_substr($subword, $hp[$k], $hp[$k + 1] - $hp[$k], $mb_encoding), 'IsLetter' => True);
}
$word_parts[] = array('Offset' => $prev_i + $hp[$num_hp - 1], 'Length' => $sub_word_len - $hp[$num_hp - 1], 'Subword' => mb_substr($subword, $hp[$num_hp - 1], $sub_word_len - $hp[$num_hp - 1], $mb_encoding), 'IsLetter' => True);
} else {
for ($k = 0; $k < $num_hp - 1; $k++) {
$word_parts[] = array('Offset' => $prev_i + $hp[$k], 'Length' => $hp[$k + 1] - $hp[$k], 'Subword' => substr($subword, $hp[$k], $hp[$k + 1] - $hp[$k]), 'IsLetter' => True);
}
$word_parts[] = array('Offset' => $i + $hp[$num_hp - 1], 'Length' => $sub_word_len - $hp[$num_hp - 1], 'Subword' => substr($subword, $hp[$num_hp - 1]), 'IsLetter' => True);
}
}
if (!$is_letter) {
$word_parts[] = array('Offset' => $i, 'Length' => 1, 'Subword' => $c, 'IsLetter' => False);
}
$prev_i = $i;
} while ($i < $word_len);
return $word_parts;
}
/**
* Get the I2CE_Encoding according to one of the standard adobe encodings
*/
protected function getAdobeStandardEncoding($encoding)
{
if (isset($this->adobe_standard_encondings[$encoding])) {
return;
}
$enc = new I2CE_Encoding($encoding);
if (!isset($this->glyph_list)) {
$this->loadGlyphList();
}
$a = $this->load_file('PDF_CORE', $encoding . '.list');
foreach ($a as $l) {
$l = trim($l);
$e = explode(" ", rtrim($l));
$cc = (int) $e[0];
$gn = $e[1];
$cps = $this->glyph_list[$gn];
if (count($cps) == 1) {
//a adobe glyphname may be associated to several unicode codepoints (e.g. lamedholam)
//if this is the case, we ignore it.
$enc->setGlyphname($cps[0], $gn);
$enc->setCharacterCode($cps[0], $cc);
}
}
$this->adobe_standard_encodings[$encoding] = $enc;
return $enc;
}
/**
* Load the font metrics from an afm file
* (Some of this code was stolen from makefont.php)
* @param I2CE_Encoding $encoding -- the character encoding used in the file
* @param string $afmfile
* Caution: Units for these files are 1/1000 of a point where a point is 1/72 of an inch
*/
protected function loadFontMetricFromAFM($encoding, $afmfile)
{
$prev_dir = $this->getDirection();
$sub = mb_substitute_character();
mb_substitute_character("none");
//Read a font metric file
$found_afmfile = I2CE::getFileSearch()->search('AFM_PATH', $afmfile);
if (!$found_afmfile) {
die('Error: AFM file not found: ' . $afmfile);
} else {
$afmfile = $found_afmfile;
}
$a = file($found_afmfile);
if (empty($a)) {
die('File empty' . $afmfile);
}
$this->setGlobal();
//global information is the default
$this->setLinegap(0);
//no line gap information is present in a AFM file
$mode = 0;
/**
* Modes are 0: header or wirting direction
* 10: Character metrics
* 20: Kerning
* 21: Kerning Tracking
* 21: Kerning Pairs
* 30: Composites
*/
foreach ($a as $l) {
$e = explode(' ', rtrim($l));
$code = $e[0];
switch ($mode) {
case 10:
//character metrics
$vals = array();
unset($cc);
unset($gn);
switch ($code) {
case 'EndCharacterMetrics':
case 'EndCharMetrics':
$mode = 0;
continue 2;
//break out of the switch($mode)
//break out of the switch($mode)
case 'C':
$cc = (int) $e[1];
break;
case 'CH':
$cc = hexdec($e[1]);
break;
}
if (!isset($cc)) {
break;
}
$i = 3;
while ($i < count($e)) {
$subcode = $e[$i];
switch ($subcode) {
case 'WX':
case 'W0X':
case 'W1X':
case 'WY':
case 'W0Y':
case 'W1Y':
//widths and heights
$vals[$subcode] = (double) $e[$i + 1];
$i = $i + 2;
break;
case 'W':
case 'W0':
case 'W1':
case 'VV':
//widths and heights
$vals[$subcode] = array((double) $e[$i + 1], (double) $e[$i + 2]);
$i = $i + 3;
break;
case 'N':
//postscript glyph name
$gn = $e[$i + 1];
$i = $i + 2;
break;
case 'B':
//bounding box
$vals[$subcode] = array((double) $e[$i + 1], (double) $e[$i + 2], (double) $e[$i + 3], (double) $e[$i + 4]);
$i = $i + 5;
break;
case 'L':
//Ligature sequence -- may have more than one
if (!isset($vals['L'])) {
$vals['L'] = array();
}
$vals['L'][] = array($e[$i + 1], $e[$i + 2]);
$i = $i + 3;
$i++;
break;
default:
$i++;
break;
}
}
unset($uc);
if ($cc < 0) {
//not a valid character code
if ($gn !== null) {
//try to get the unicode code point of the glyphname based on our encoding
$uc = $encoding->UnicodeFromGlyphname($gn);
if ($uc === null) {
//we failed
//try to get a unicode code point from the glyphname
if (preg_match('/^uni([0-9A-F]{4})$/', $gn, $ucs)) {
//we have a unicode codepoint
$uc = $ucs[1];
}
}
}
} else {
//we have a valid character code
$uc = $encoding->UnicodeFromCharactercode($cc);
if ($gn !== null) {
$this->gn2cc[$gn] = $cc;
}
}
if ($uc !== null && $uc <= 0xffff && $uc >= 0 && $uc !== 0xfffd) {
//replacement character
$uc = I2CE_UTF8::cp_to_code($uc);
//convert to UTF8
$cc = mb_convert_encoding($uc, $this->getEncoding()->getEncodingType(), 'UTF-8');
if ($this->getEncoding()->useMB()) {
if (mb_strlen($cc, $this->getEncoding()->getEncodingType()) === 0) {
$cc = -1;
}
} else {
if (strlen($cc) === 0) {
$cc = -1;
}
}
} else {
$cc = -1;
}
foreach (array('WX', 'W0X') as $key) {
if (array_key_exists($key, $vals) && $vals[$key] !== null) {
$this->setDirection('H');
if ($cc !== -1) {
$this->setCharacterWidth($cc, $vals[$key]);
}
if ($gn !== null) {
$this->setCharacterWidth($gn, $vals[$key]);
}
}
}
if (array_key_exists('W1X', $vals) && $vals['W1X'] !== null) {
$this->setDirection('V');
if ($cc !== -1) {
$this->setCharacterWidth($cc, $vals['W1X']);
}
if (null !== $gn) {
$this->setCharacterWidth($gn, $vals['W1X']);
}
}
foreach (array('WY', 'W0Y') as $key) {
if (array_key_exists($key, $vals) && null !== $vals[$key]) {
$this->setDirection('H');
if ($cc !== -1) {
$this->setCharacterHeight($cc, $vals[$key]);
}
if (null !== $gn) {
$this->setCharacterHeight($gn, $vals[$key]);
}
}
}
if (array_key_exists('W1Y', $vals) && null !== $vals['W1Y']) {
$this->setDirection('V');
if ($cc !== -1) {
$this->setCharacterHeight($cc, $vals['W1Y']);
}
if (null !== $gn) {
$this->setCharacterHeight($gn, $vals['W1Y']);
}
}
foreach (array('W', 'W0') as $key) {
$this->setDirection('H');
if (array_key_exists($key, $vals) && null !== $vals[$key]) {
if ($cc !== -1) {
$this->setCharacterWidth($cc, $vals[$key]);
$this->setCharacterHeight($cc, $vals[$key]);
}
if (null !== $gn) {
$this->setCharacterWidth($gn, $vals[$key]);
$this->setCharacterHeight($gn, $vals[$key]);
}
}
}
if (array_key_exists('W1', $vals) && null !== $vals['W1']) {
$this->setDirection('V');
if ($cc !== -1) {
$this->setCharacterWidth($cc, $vals[$key]);
$this->setCharacterHeight($cc, $vals[$key]);
}
if (null !== $gn) {
$this->setCharacterWidth($gn, $vals[$key]);
$this->setCharacterHeight($gn, $vals[$key]);
}
}
foreach (array('VVector' => 'VV', 'BoundingBox' => 'B', 'Ligature' => 'L') as $name => $key) {
$this->setGlobal();
if (array_key_exists($key, $vals) && null !== $vals[$key]) {
if ($cc !== -1) {
$this->setCharacterInfo($cc, $name, $vals[$key]);
}
if (null !== $gn) {
$this->setCharacterInfo($gn, $name, $vals[$key]);
}
}
}
break;
case 20:
//kerning
switch ($code) {
case 'EndKernData':
$mode = 0;
break;
case 'StartTrackKern':
$mode = 21;
break;
case 'StartKernPairs':
case 'StartKernPairs0':
$this->setDirection('H');
$mode = 22;
break;
case 'StartKernPairs1':
$this->setDirection('V');
$mode = 22;
break;
}
break;
case 21:
//kerning tracking
switch ($code) {
case 'EndTrackKern':
$mode = 20;
break;
case '':
break;
}
break;
case 22:
//kerning pairs
switch ($code) {
case 'EndKernPairs':
$mode = 21;
break;
case 'KP':
//kerning pairs are given by glyph name
$this->setDirection('H');
$this->setKerningByPair($e[1], $e[2], (double) $e[3]);
$this->setDirection('V');
$this->setKerningByPair($e[1], $e[2], (double) $e[4]);
//get the corresponding character codes and insert into the table
$this->setGlobal();
$cc1 = $this->getEncoding()->getCodeFromGlyphname($e[1]);
$cc2 = $this->getEncoding()->getCodeFromGlyphname($e[2]);
if ($cc1 !== -1 && $cc2 !== -1) {
$this->setDirection('H');
$this->setKerningByPair($cc1, $cc2, (double) $e[3]);
$this->setDirection('V');
$this->setKerningByPair($cc1, $cc2, (double) $e[4]);
}
break;
case 'KPH':
//not sure what is best to do here.
$ch1 = ltrim(rtrim($e[1], '>'), '<');
$ch2 = ltrim(rtrim($e[1], '>'), '<');
$this->setDirection('H');
$this->setKerningByPair($ch1, $ch2, (double) $e[3]);
$this->setDirection('V');
$this->setKerningByPair($ch1, $ch2, (double) $e[4]);
break;
case 'KPX':
$this->setDirection('H');
$this->setKerningByPair($e[1], $e[2], (double) $e[3]);
if (!array_key_exists($e[1], $this->gn2cc) || !array_key_exists($e[2], $this->gn2cc)) {
break;
}
$cc1 = $this->gn2cc[$e[1]];
$cc2 = $this->gn2cc[$e[2]];
if ($cc1 !== -1 && $cc2 !== -1) {
$this->setKerningByPair($cc1, $cc2, (double) $e[3]);
}
break;
case 'KPY':
$this->setDirection('V');
$this->setKerningByPair($e[1], $e[2], (double) $e[3]);
if (!array_key_exists($e[1], $this->gn2cc) || !array_key_exists($e[2], $this->gn2cc)) {
break;
}
$cc1 = $this->gn2cc[$e[1]];
$cc2 = $this->gn2cc[$e[2]];
if ($cc1 !== -1 && $cc2 !== -1) {
$this->setKerningByPair($cc1, $cc2, (double) $e[3]);
}
break;
}
break;
case 30:
//composites
switch ($code) {
case 'EndComposites':
$mode = 0;
break;
}
break;
default:
//header/global information or writing direction
switch ($code) {
case 'BeginCharacterMetrics':
case 'StartCharMetrics':
$mode = 10;
break;
case 'BeginKernData':
case 'StartKernData':
$mode = 20;
break;
case 'BeginComposites':
case 'StartComposites':
$mode = 03;
break;
case 'StartDirection':
//does not have to exist in which case we are in direction 0
if ($e[1] == '1') {
$this->setDirection('V');
} else {
$this->setDirection('H');
}
break;
case 'EndDirection':
$this->setGlobal();
break;
case 'CharWidth':
$dir = $this->getDirection();
//this is not global information
if ($dir === -1) {
// however the keyword StartDirection is optional
$this->setDirection('H');
}
$this->setFixedWidth(true);
$this->setFixedWidthSize((double) $e[1]);
$this->setFixedHeightSize((double) $e[2]);
$this->setDirection($dir);
break;
case 'UnderlinePosition':
case 'UnderlineThickness':
case 'ItalicAngle':
$dir = $this->getDirection();
//this is not global information
if ($dir === -1) {
// however the keyword StartDirection is optional
$this->setDirection('H');
}
$this->setFontCharacteristic($code, $e[1]);
$this->setDirection($dir);
break;
case 'IsFixedPitch':
$isfixed = strpos(strtolower($e[1]), 'true') === 0;
$dir = $this->getDirection();
//this is not global information
if ($dir === -1) {
// however the keyword StartDirection is optional
$this->setDirection('H');
}
$this->setFixedWidth($isfixed);
$this->setDirection($dir);
break;
case 'isFixedV':
case 'isBaseFont':
$this->setGlobal();
$this->setFontCharacteristic($code, strpos(strtolower($e[1]), 'true'));
break;
case 'MappingScheme':
case 'EscCar':
case 'Characters':
$this->setGlobal();
$this->setFontCharacteristic($code, (int) $e[1]);
break;
case 'Notice':
case 'Comment':
$this->setGlobal();
$val = $this->getFontCharacteristic($code);
if ($val === null) {
$val = "";
}
$this->setFontCharacteristic($code, $val . substr($l, strlen($code) + 1));
break;
case 'CapHeight':
case 'XHeight':
$this->setGlobal();
$this->setFontCharacteristic($code, (double) $e[1]);
break;
case 'Ascender':
$this->setGlobal();
$this->setAscender((double) $e[1]);
break;
case 'Descender':
$this->setGlobal();
$this->setDescender((double) $e[1]);
break;
case 'VVector':
$this->setGlobal();
$this->setFontCharacteristic($code, array((double) $e[1], (double) $e[2]));
case 'FontBBox':
$this->setGlobal();
$this->setBoundingBox(array((double) $e[1], (double) $e[2], (double) $e[3], (double) $e[4]));
break;
default:
//the rest are strings for global font information
$this->setGlobal();
$this->setFontCharacteristic($code, $e[1]);
break;
}
break;
}
}
//normalize a few values.
$this->setGlobal();
$asc = $this->getAscender();
if ($asc == 0) {
$d = mb_convert_encoding('d', $this->getEncoding()->getEncodingType(), 'ASCII');
$ht = $this->getCharacterHeight($d);
if ($ht != 0) {
$this->setAscender($ht);
} else {
$this->setDirection('H');
$ht = $this->getCharacterHeight($d);
if ($ht != 0) {
$this->setAscender($ht);
} else {
$this->setGlobal();
$bbox = $this->getBoundingBox();
$this->setAscender($bbox[3]);
}
}
}
$this->setGlobal();
$dsc = $this->getAscender();
if ($dsc == 0) {
$p = mb_convert_encoding('p', $this->getEncoding()->getEncodingType(), 'ASCII');
$bbox = $this->getCharacterInfo($p, 'BoundingBox');
if (!$bbox == null) {
$this->setDescender($bbox[1]);
} else {
$this->setDirection('H');
$bbox = $this->getCharacterInfo($p, 'BoundingBox');
if (!$bbox == null) {
$this->setDescender($bbox[1]);
} else {
$this->setGlobal();
$bbox = $this->getBoundingBox();
$this->setDescender($bbox[1]);
}
}
}
$this->setDirection($prev_dir);
mb_substitute_character($sub);
}
