/**
* Кэширумеая функция, основной разбор.
*/
public static function parse($src)
{
$args = array($src);
$param = Cache::exec(array($src), 'docx_parse', function ($src, $re) {
$conf = Docx::$conf;
$imgmaxwidth = $conf['imgmaxwidth'];
$previewlen = $conf['previewlen'];
$cachename = Path::tofs(Path::encode($src));
$cacheFolder = Path::mkdir(Docx::$conf['cache'] . $cachename . '/');
//В винде ингда вылетает о шибка что нет прав удалить какой-то файл в папке и как следствие саму папку
//Обновление страницы проходит уже нормально
//Полагаю в линукс такой ошибки не будет хз почему возникает
Cache::fullrmdir($cacheFolder);
$path = Path::theme($src);
if (!$path) {
return array('html' => false);
}
$xmls = docx_getTextFromZippedXML($path, 'word/document.xml', $cacheFolder, $re);
$rIds = array();
$param = array('folder' => $cacheFolder, 'imgmaxwidth' => $imgmaxwidth, 'previewlen' => $previewlen, 'rIds' => $rIds);
if ($xmls[0]) {
$xmlar = docx_dom_to_array($xmls[0]);
$xmlar2 = docx_dom_to_array($xmls[1]);
foreach ($xmlar2['Relationships']['Relationship'] as $v) {
$rIds[$v['Id']] = $v['Target'];
}
$param['rIds'] = $rIds;
$html = docx_each($xmlar, '\\infrajs\\doc\\docx_analyse', $param);
} else {
$param['rIds'] = array();
$html = '';
}
$param['html'] = $html;
return $param;
}, $args, isset($_GET['re']));
unset($param['rIds']);
unset($param['type']);
unset($param['imgmaxwidth']);
unset($param['previewlen']);
unset($param['isli']);
unset($param['isul']);
unset($param['imgnum']);
unset($param['folder']);
return $param;
}
public static function parse($src)
{
$src = Path::theme($src);
if (!$src) {
return;
}
$args = array($src);
return Cache::exec(array($src), 'mhtparse', function ($src) {
$conf = Docx::$conf;
$imgmaxwidth = $conf['imgmaxwidth'];
$previewlen = $conf['previewlen'];
$filename = Path::theme($src);
$fdata = Load::srcInfo($src);
if ($fdata['ext'] == 'php') {
$data = Load::loadTEXT($filename);
} else {
$data = file_get_contents($filename);
}
$ans = array();
if ($fdata['ext'] == 'mht') {
$p = explode('/', $filename);
$fname = array_pop($p);
$fnameext = $fname;
//$fname=basename($filename);
preg_match("/^(\\d*)/", $fname, $match);
$date = $match[0];
$fname = Path::toutf(preg_replace('/^\\d*\\s+/', '', $fname));
$fname = preg_replace('/\\.\\w{0,4}$/', '', $fname);
$ar = preg_split('/------=_NextPart_.*/', $data);
if (sizeof($ar) > 1) {
//На первом месте идёт информация о ворде...
unset($ar[0]);
unset($ar[sizeof($ar) - 1]);
}
$ar = array_values($ar);
$folder = Path::mkdir(Docx::$conf['cache'] . md5($src) . '/');
$html = '';
for ($i = 0, $l = sizeof($ar); $i < $l; ++$i) {
if (!$ar[$i]) {
continue;
}
$d = preg_split("/\n/", $ar[$i], 6);
$j = -1;
do {
++$j;
} while (@$d[$j][0] !== 'C' && $j <= 5);
if ($j >= 5) {
/*
не нашли
Content-Location: file:///C:/0FCF1655/file9909.files/header.htm
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html; charset="us-ascii"
*/
continue;
}
$location = preg_replace('/Content-Location: /', '', $d[$j]);
$location = trim($location);
$encoding = preg_replace('/Content-Transfer-Encoding: /', '', $d[$j + 1]);
$type = preg_replace('/Content-Type: /', '', $d[$j + 2]);
$content = $d[5];
$name = basename($location);
if (preg_match("/text\\/html/", $type) || preg_match('/Subject:/', $type)) {
$html .= $content;
} else {
@file_put_contents($folder . $name, base64_decode($content));
//Сохраняем картинку или тп...
}
}
if (!$html) {
$html = '';
}
$html = preg_replace("/=\r\n/", '', $html);
$html = preg_replace("/\\s+/", ' ', $html);
$html = preg_replace("/^.*<body .*>\\s*/U", '', $html, 1);
$html = preg_replace("/\\s*<\\/body>.*/", '', $html, 1);
$images = array();
preg_match_all('/src=3D".*\\.files\\/(image.+)"/U', $html, $match, PREG_PATTERN_ORDER);
for ($i = 0, $l = sizeof($match[1]); $i < $l; $i = $i + 2) {
$min = $match[1][$i + 1];
if (!$min) {
$min = $match[1][$i];
}
$images[$min] = $match[1][$i];
//Каждая следующая картинка есть уменьшенная копия предыдущей оригинального размера
}
$html = preg_replace("/<\\!--.*-->/U", '', $html);
$html = preg_replace("/<!\\[if !vml\\]>/", '', $html);
$html = preg_replace("/<!\\[endif\\]>/", '', $html);
$html = preg_replace('/=3D/', '=', $html);
$html = preg_replace('/align="right"/', 'align="right" class="right"', $html);
$html = preg_replace('/align="left"/', 'align="left" class="left"', $html);
$html = preg_replace('/align=right/', 'align="right" class="right"', $html);
$html = preg_replace('/align=left/', 'align="left" class="left"', $html);
$html = Path::toutf($html);
//Виндовые файлы хранятся в cp1251
$folder = Path::toutf($folder);
$html = preg_replace('/ src=".*\\/(.*)"/U', ' src="' . $folder . '${1}"', $html);
$html = preg_replace('/<span class=SpellE>(.*)<\\/span>/U', '${1}', $html);
$html = preg_replace('/<span lang=.*>(.*)<\\/span>/U', '${1}', $html);
$html = preg_replace('/<span class=GramE>(.*)<\\/span>/U', '${1}', $html);
$html = preg_replace("/<span style='mso.*>(.*)<\\/span>/U", '${1}', $html);
$html = preg_replace("/<span style='mso.*>(.*)<\\/span>/U", '${1}', $html);
$html = preg_replace("/<span style='mso.*>(.*)<\\/span>/U", '${1}', $html);
$html = preg_replace("/<span style='mso.*>(.*)<\\/span>/U", '${1}', $html);
$html = preg_replace('/ class=MsoNormal/U', '', $html);
$html = preg_replace('/<a name="_.*>(.*)<\\/a>/U', '${1}', $html);
//Приводим к единому виду маркерные списки
$patern = '/<p class=MsoListParagraphCxSp(\\w+) .*>(.*)<\\/p>/U';
$count = 3;
do {
preg_match($patern, $html, $match);
if (sizeof($match) == $count) {
$pos = strtolower($match[1]);
$text = $match[2];
$text = preg_replace('/^.*(<\\/span>)+/U', '', $text, 1);
$text = '<li>' . $text . '</li>';
if ($pos == 'first') {
$text = '<ul>' . $text;
}
if ($pos == 'last') {
$text = $text . '</ul>';
}
$html = preg_replace($patern, $text, $html, 1);
} else {
break;
}
} while (sizeof($match) == $count);
$title = $fname;
$patern = '/<img(.*)>/U';
$count = 2;
do {
preg_match($patern, $html, $match);
if (sizeof($match) == $count) {
$sfind = $match[1];
//$sfind='<img src="/image.asdf">';
preg_match("/width=(\\d*)/", $sfind, $match2);
$w = trim($match2[1]);
preg_match("/height=(\\d*)/", $sfind, $match2);
$h = trim($match2[1]);
if (!$w || $w > $imgmaxwidth) {
$w = $imgmaxwidth;
}
preg_match('/src="(.*\\/)(image.*)"/U', $sfind, $match2);
$path = trim($match2[1]);
$small = $match2[2];
preg_match('/alt="(.*)".*/U', $sfind, $match2);
$alt = trim(@$match2[1]);
$alt = html_entity_decode($alt, ENT_QUOTES, 'utf-8');
preg_match('/align="(.*)".*/U', $sfind, $match2);
$align = trim($match2[1]);
$align = html_entity_decode($align, ENT_QUOTES, 'utf-8');
$big = $images[$small];
if (!$big) {
$big = $small;
}
$isbig = preg_match('/#/', $alt);
if ($isbig) {
$alt = preg_replace('/#/', '', $alt);
}
//$i="<IMG title='$alt' src='/-imager/?w=$w&h=$h&src=".($path.$big)."' align='$align' class='$align' alt='$alt'>";
$i = "<IMG src='/-imager/?w={$w}&h={$h}&src=" . ($path . $big) . "' align='{$align}' class='{$align}'>";
//urlencode решает проблему с ie7 когда иллюстрации с адресом содержащим пробел не показываются
if ($isbig) {
$i = "<a target='about:blank' href='/-imager/?src=" . urlencode($path . $big) . "'>{$i}</a>";
}
//$i.='<textarea style="width:500px; height:300px">'.$i.'</textarea>';
$html = preg_replace($patern, $i, $html, 1);
} else {
break;
}
} while (sizeof($match) == $count);
$patern = "/###\\{(.*)\\}###/U";
//js код
do {
preg_match($patern, $html, $match);
if (sizeof($match) > 0) {
$param = $match[1];
$param = strip_tags($param);
$param = html_entity_decode($param, ENT_QUOTES, 'utf-8');
$param = preg_replace('/(‘|’)/', "'", $param);
$param = preg_replace('/(“|«|»|”)/', '"', $param);
$html = preg_replace($patern, $param, $html, 1);
} else {
break;
}
} while (sizeof($match) > 1);
$patern = "/####.*<table.*>(.*)<\\/table>.*####/U";
do {
preg_match($patern, $html, $match);
if (sizeof($match) > 0) {
$param = $match[1];
$param = preg_replace('/style=".*"/U', '', $param);
$param = preg_replace("/style='.*'/U", '', $param);
$html = preg_replace($patern, '<table class="table table-striped">' . $param . '</table>', $html, 1);
} else {
break;
}
} while (sizeof($match) > 1);
$ans['images'] = array();
foreach ($images as $img) {
$ans['images'][] = array('src' => $folder . $img);
}
} else {
$html = $data;
$images = array();
preg_match_all('/<img.*src="(.*)".*>/U', $html, $match, PREG_PATTERN_ORDER);
for ($i = 0, $l = sizeof($match[1]); $i < $l; $i++) {
$images[] = array('src' => $match[1][$i]);
//Каждая следующая картинка есть уменьшенная копия предыдущей оригинального размера
}
$ans['images'] = $images;
}
$r = preg_match('/<h.*>(.*)<\\/h.>/U', $html, $match);
if ($r) {
$heading = strip_tags($match[1]);
} else {
$heading = false;
}
$ans['heading'] = $heading;
preg_match_all('/<a.*href="(.*)".*>(.*)<\\/a>/U', $html, $match);
$links = array();
foreach ($match[1] as $k => $v) {
$title = strip_tags($match[2][$k]);
if (!$title) {
continue;
}
$links[] = array('title' => $title, 'href' => $match[1][$k]);
}
$ans['links'] = $links;
$html = trim($html);
$html = html_entity_decode($html, ENT_COMPAT, 'UTF-8');
$html = preg_replace('/
/U', '', $html);
//bugfix списки в mht порождаются адский символ. в eval-е скрипта недопустим.
$ans['html'] = $html;
foreach ($ans['images'] as &$item) {
$item['src'] = preg_replace('/^\\//', '', $item['src']);
}
return $ans;
}, $args);
}
<?php namespace infrajs\doc; use infrajs\path\Path; Path::mkdir(Docx::$conf['cache']);
<?php
namespace infrajs\autoedit;
use infrajs\path\Path;
require_once __DIR__ . '/../../../vendor/autoload.php';
Path::mkdir('!admin_takefiles/');
<?php
namespace infrajs\mem;
use infrajs\path\Path;
Path::mkdir(Mem::$conf['cache']);
$mem =& Mem::memcache();
if ($mem) {
$mem->flush();
}
<?php namespace infrajs\imager; use infrajs\path\Path; require_once __DIR__ . '/../../../vendor/autoload.php'; Path::mkdir(Imager::$conf['cache']); Path::mkdir(Imager::$conf['cache'] . 'resize/'); Path::mkdir(Imager::$conf['cache'] . 'remote/');
<?php namespace infrajs\excel; use infrajs\path\Path; Path::mkdir(Xlsx::$conf['cache']);
function &imager_makeInfo($src)
{
$info = imager_readInfo($src);
if ($info) {
return $info;
}
$dir = '~imager/';
Path::mkdir($dir);
$dir .= '.notwater/';
Path::mkdir($dir);
$i = '';
$orig = $dir . Path::encode($src);
while (is_file($orig)) {
$orig = $orig . $i;
$i .= 'i';
}
$r = copy($src, $orig);
//по адресу orig не существует файла было проверено
if (!$r) {
die('Не удалось сохранить оригинал');
}
$info = array();
$info['host'] = $_SERVER['HTTP_HOST'];
$info['size'] = filesize(Path::tofs($orig));
$info['date'] = date('j.m.Y');
$info['orig'] = Path::toutf($orig);
return $info;
}
<?php
namespace infrajs\autoedit;
use infrajs\path\Path;
require_once __DIR__ . '/../../../vendor/autoload.php';
Path::mkdir('~.marks/');
