Unable to extract zip file which generated using php - php

I have written the PHP script to generate the zip file. it's working fine when I use rar software to extract it but not getting extract with rar software. I can't ask to users to install rar software to extract downloaded zip file.
I don't know where i am commiting mistakes.
Here i attached error screen shot which i get when try to open zip file.
// Here is code snippet
$obj->create_zip($files_to_zip, $dir . '/download.zip');
// Code for create_zip function
//create the archive
$zip = new ZipArchive();
if ($zip->open($destination, $overwrite ? ZIPARCHIVE::OVERWRITE : ZIPARCHIVE::CREATE) !== true) {
return false;
}
//add the files
foreach ($valid_files as $file) {
$filearr = explode('/', $file);
$zip->addFile($file, end($filearr));
}
$zip->close();

If $valid_files is a glob'd array, use basename() instead of end(), your zip might not actually have added any files causing for it to be an invalid zip (however that would be visible in the size of the zip file).
Also try winrar/winzip/7zip and see what they return, microsoft's internal zip engine might not be up to date enough to open the zips.

I have also encountered this problem, using 7z solved the problem but we need to send the zip to somebody else so 7z is a nono.
I found that, in my case it is that the file path is too long:
When I use this:
$zip->addFile($files_path.'/people.txt');
And it generated a zip folder nested very deep e.g. ["/tmp/something/something1/something2/people.txt"]
So I need to use this instead
$zip->addFile($files_path.'/people.txt', 'people.txt');
Which generate a a zip folder with only 1 layer ["people.txt"], and Windows Zip read perfectly~
Hope this helps somebody that also have this problem!

Related

PHP — Information on the zip:// protocol

I am using ZipArchive class to unzip a file and put its contents somewhere useful. Using information derived from the comments on php.net, I ended up writing this function:
function unzip(string $zipFile, string $destination) {
$zip = new ZipArchive();
$zip->open($zipFile);
for($i=0; $i<$zip->numFiles; $i++) {
$file=$zip->getNameIndex($i);
if(substr($file,-1) == '/') continue; // skip containing folder
$name=basename($file);
copy("zip://$zipFile#$file","$destination/$name");
}
$zip->close();
}
This is to copy the individual files without the folder structure.
I can understand most of the code, but I cannot get any information on the following expression:
"zip://$zipFile#$file"
I know what it doing (obviously it is extracting one of the files from the Zip archive), but can anyone tell me more about the zip:// protocol, and why it uses the # to reference a particular file?
Check out source of Zip extension
https://github.com/php/php-src/blob/master/ext/zip/zip_stream.c
line 135. fragment = strchr(path, '#');
get pointer of entry in path/to/zip#entry
line 141.
if (strncasecmp("zip://", path, 6) == 0)
{
path += 6;
}
if zip:// is equal first 6 characters of path
I won't go into the logic of this code. It just exists here (zip_stream.c) and maybe somewhere else.
It seems like this extension "creates" zip:// protocol over php executable that is laying over apache server.

Reading files from ZIP without extracting to disk

Is it possible to open a ZIP file on a server, read a file from its content and display it / send it to a client directly without extracting it to disk first ? I am talking about pdf's and images. Haven't found any hints in the php sites.
Well,there is a PHP Extension.
If you use the extractTo method, you would be able to extract a single file, checkout the documentation.
From the documentation, extracting two files:-
<?php
$zip = new ZipArchive;
$res = $zip->open('test_im.zip');
if ($res === TRUE) {
$zip->extractTo('/my/destination/dir/', array('pear_item.gif', 'testfromfile.php'));
$zip->close();
echo 'ok';
} else {
echo 'failed';
}
?>
You would need to provide an array of path inside the zip.

PharData offsetExists on filename prefixed with ".\"

I have a .tar.gz file downloaded from an external API which we have to implement. It contains images for an object.
I'm not sure how they managed to compress it this way, but the files are basically prefixed with the "current directory". It looks like this in WinRAR:
And like this in 7-Zip, note the .tar first level, and "." second level:
-> ->
When calling
$file = 'archive.tar.gz';
$phar = new PharData($file, FilesystemIterator::CURRENT_AS_FILEINFO);
var_dump($phar->offsetGet('./12613_s_cfe3e73.jpg'));
I get the exception:
Cannot access phar file entry '/12613_s_cfe3e73.jpg' in archive '{...}/archive.tar.gz'
Calling a file which does not exist, e.g.:
var_dump($phar->offsetGet('non-existent.jpg'));
Or calling it without the directory seperator, e.g.:
var_dump($phar->offsetGet('12613_s_cfe3e73.jpg'));
I get a
Entry 12613_s_cfe3e73.jpg does not exist
Exception.
It is not possible to get the archive formatted differently. Does anyone have an idea how to solve this?
Ended up using Archive_Tar. There must be something wrong in the source code of PHP, though I don't think this is the "normal" way of packaging a .tar either.
Unfortunately I'm not very good at C, but it's probably in here (line 1214) or here.
This library seems to handle it just fine, using this example code:
$file = 'archive.tar.gz';
$zip = new Archive_Tar($file);
foreach ($zip->listContent() as $file) {
echo $file['filename'] . '<br>';
}
Result:
./12613_s_f3b483d.jpg
./12613_s_cfe3e73.jpg
./1265717_s_db141dc.jpg
./1265717_s_af5de56.jpg
./1265717_s_b783547.jpg
./1265717_s_35b11f9.jpg
./1265716_s_83ef572.jpg
./1265716_s_9ac2725.jpg
./1265716_s_c5af3e9.jpg
./1265716_s_c070da3.jpg
./1265715_s_4339e8a.jpg
Note the filenames are still prefixed with "./" just like they are in WinRAR.
If you want to stick to using PharData, i suggest a more conservative, two-step approach, where you first decompress the gz and then unarchive all files of the tar to a target folder.
// decompress gz archive to get "/path/to/my.tar" file
$gz = new PharData('/path/to/my.tar.gz');
$gz->decompress();
// unarchive all files from the tar to the target path
$tar = new PharData('/path/to/my.tar');
$tar->extractTo('/target/path');
But it looks like you want to select individual files from the tar.gz archive directly, right?
It should work using fopen() with a StreamReader (compress.zlib or phar) and selecting the individual file. Some examples:
$f = fopen("compress.zlib://http://some.website.org/my.gz/file/in/the/archive", "r");
$f = fopen('phar:///path/to/my.tar.gz//file/in/archive', 'r');
$filecontent = file_get_contents('phar:///some/my.tar.gz/some/file/in/the/archive');
Streaming should also work, when using Iterators:
$rdi = new RecursiveDirectoryIterator('phar:///path/to/my.tar.gz')
$rii = new RecursiveIteratorIterator($rdi, RecursiveIteratorIterator::CHILD_FIRST);
foreach ($rii as $splFileInfo){
echo file_get_contents($splFileInfo->getPathname());
}
The downside is that you have to buffer the stream and save it to file.
Its not a direct file extraction to a target folder.

Saving Archive Comment to ePub file

I am working on a PHP application which downloads an ePub file. As part of the code I am adding a unique identifier which can be used to trace the original download information in a database to combat piracy.
I am using PHP ZipArchive functions to add an archive comment to the file but it seems that Adobe Digital editions is unable to open the ePub file if there is a comment present. When the comment is removed it opens fine so I just want to see if anyone else has had this problem before and if there is a way to add a comment to the ePub file which will allow Adobe Digital Editions to open it. Below is the code I'm using:
public function secureEpub($file, $download_id, $filepath){
//GENERATE RANDOM FILE NAME
$newFile = uniqid().".epub";
//COPY EPUB AND WORK FROM THAT VERSION
copy($file, $filepath.$newFile);
//TREAT EPUB FILE AS ORDINARY ZIP ARCHIVE
//OPEN AND EXTRACT EPUB
$zip = new ZipArchive;
$res = $zip->open($filepath.$newFile);
if($res === TRUE){
//ZIP ARCHIVE HAS OPENED SUCCESSFULLY - SET ARCHIVE COMMENT TO DOWNLOAD ID
$zip->setArchiveComment($download_id);
$zip->close();
} else {
error_log('Unable to open ePub (' . $filepath.$newFile . ') as Directory');
return false;
exit;
}
//ONCE CHANGE HAS BEEN MADE CLOSE THE ARCHIVE AND RETURN TEMP FILE
return $filepath.$newFile;
}
As you can see it copies the original, creates an Archive comment then returns the file (which is later deleted). Is there a way of achieving this without corrupting the ePub (for ADE)?
Let me know if I've not made any sense :D
P.S: Just to clarify, the download works fine and the comment is applied correctly. It is just that ADE is unable to open it while other ePub readers manage fine. Am I going about it in the wrong way?
OK, I couldn't find a way around this so I kind of came up with a solution. You can add a new file or directory inside the ZIP file which had the details included.
So, you can use something like:
$zip->addFromString('test.txt', $download_id);
or create a directory named after the download like:
$zip->addEmptyDir($download_id);
Would probably recommend being a bit more obscure with naming though if you get stuck like I did :)
Thanks

zip files in PHP from another directory without a path structure

I'm trying to zip two files in another directory without zipping the folder hierarchy as well.
The event is triggered by a button press, which causes Javascript to send information using AJAX to PHP. PHP calls a Perl script (to take advantage of Perl's XLSX writer module and the fact that PHP kind of sucks, but I digress...), which puts the files a few folders down the hierarchy. The relevant code is shown below.
system("createFiles.pl -ids ${rows} -test ${test} -path ${path}",$retVal);
`zip ${path}/{$test}_both.zip ${path}/${test}.csv ${path}/${test}.xlsx`;
`zip ${path}/{$test}_csv.zip ${path}/${test}.csv`;
The problem is the zip file has ${path} hierarchy that has to be navigated before the files are shown as seen below:
I tried doing this (cd before each zip command):
system("createFiles.pl -ids ${rows} -test ${test} -path ${path}",$retVal);
`cd ${path}; zip {$test}_both.zip ${test}.csv ${test}.xlsx`;
`cd ${path}; zip {$test}_csv.zip ${test}.csv`;
And it worked, but it seems like a hack. Is there a better way?
The ZipArchive answer by Oldskool is good. I've used ZipArchive and it works. However, I recommend PclZip instead as it is more versatile (e.g. allows for zipping with no compression, ideal if you are zipping up images which are already compressed, much faster). PclZip supports the PCLZIP_OPT_REMOVE_ALL_PATH option to remove all file paths. e.g.
$zip = new PclZip("$path/{$test}_both.zip");
$files = array("$path/$test.csv", "$path/$test.xlsx");
// create the Zip archive, without paths or compression (images are already compressed)
$properties = $zip->create($files, PCLZIP_OPT_REMOVE_ALL_PATH);
if (!is_array($properties)) {
die($zip->errorInfo(true));
}
If you use PHP 5 >= 5.2.0 you can use the ZipArchive class. You can then use the full path as source filename and just the filename as target name. Like this:
$zip = new ZipArchive;
if($zip->open("{$test}_both.zip", ZIPARCHIVE::OVERWRITE) === true) {
// Add the files here with full path as source, short name as target
$zip->addFile("${path}/${test}.csv", "${test}.csv");
$zip->addFile("${path}/${test}.xlsx", "${test}.xlsx");
$zip->close();
} else {
die("Zip creation failed.");
}
// Same for the second archive
$zip2 = new ZipArchive;
if($zip2->open("{$test}_csv.zip", ZIPARCHIVE::OVERWRITE) === true) {
// Add the file here with full path as source, short name as target
$zip2->addFile("${path}/${test}.csv", "${test}.csv");
$zip2->close();
} else {
die("Zip creation failed.");
}

Categories