I'm creating a zipped archive with zip4j using the following method to add single files to the archive:
zipParameters = new ZipParameters();
zipParameters.setCompressionMethod(Zip4jConstants.COMP_DEFLATE);
zipParameters.setCompressionLevel(Zip4jConstants.DEFLATE_LEVEL_ULTRA);
zipFile = new ZipFile(fullQualifiedZipFile);
...
zipFile.addFile(new File(fullQualilfiedPath), zipParameters);
Doing so, the resulting archive doesn't contain the folder structure. Any idea how to resolve this?
Thanks in advance for any hint!
Kind regards,
Michael
Just if somebody is interested, I've found the solution. Before adding a file to the archive, method setRootFolderInZip must be called:
zipParameters.setRootFolderInZip(folder);
This will lead to the desired result, that the folder structure will be preserved. You may want to adjust the folder. At least the root part should be removed.
Michael
Related
I'm trying to create some template to simplify some boring repeating stuff in existing project.
I need to create something like entity and it's behavior classes, e.g. I want to add new entity "Hero" and I need the following structure in the end:
Entities/Hero/Hero.h
Entities/Hero/Hero.cpp
Entities/Hero/StateControllers/HeroStateController.h
Entities/Hero/StateControllers/HeroStateController.cpp
So I've created the following structure in my xcode template
The problem is when I use "New file..." and create "Hero" all files are added but folders are blue what means that this folder will be considered as assets folder. But I don't need it. I need it to be group referenced to folder(yellow one). I've searched the whole internet and could not find anything for file templates.
Can anybody please help me?
Thanks in advance!
How could I possible open a symbolic link and get the content of the file instead of the file it is pointing to?
By doing:
with open('/home/symlink.txt', 'rb') as f:
data=f.read()
If the symbolic link points to /foo/faa.txt, the variable data will contain the content of faa.txt. This is a big security and file problem from my server because I'm generating zip archives.
If for example, a folder contains multiple symbolic links with different names to avoid duplicating files, the zip archive will contain multiple files instead of multiple symbolic links!
I hope to be clear enough!
An extra explanation:
The point of this is to allow downloading symlinks in a django server. The way of returning files is the following one:
response = HttpResponse()
response.write(data))
return response
This means that data must contain the content that the user will download. I can not just give it a path. So what I need to do is to give it a symbolic link. The problem is that reading a symbolic link makes python read the content where it is pointing to instead of its real content. In a few words, the user downloads the real file instead of the symbolic link!
A possible solution to this would be to get the path where the symlink points to, and then generate the link in the buffer. Is this possible?
It looks like there are 2 questions here: How can you read a symlink from the filesystem, and how can you store this in a .zip file such that it will be recreated when you unzip it.
Reading a symlink
The contents of a symlink are defined here:
http://man7.org/linux/man-pages/man7/symlink.7.html
A symbolic link is a special type of file whose contents are a string that is the pathname of another file, the file to which the link refers
You can read that path by using os.readlink (https://docs.python.org/2/library/os.html#os.readlink) - this is analogous to C's readlink function.
It's also important to note that these symlinks aren't distinguished by their content or file attributes, but by the fact that the file entry on disk points to a string rather than a file object:
In other words, a symbolic link is a pointer to another name, and not to an underlying object.
This means that there isn't really a "file" you could store in the ZIP. So how do the existing zip & unzip utilities do it?
Storing a symlink in a zip file
The spec for the ZIP format is here: https://pkware.cachefly.net/webdocs/casestudies/APPNOTE.TXT
Note that section 4.5.7 (defining UNIX Extra Field) says:
The variable length data field will contain file type specific data. Currently the only values allowed are the original "linked to" file names for hard or symbolic links, and the major and minor device node numbers for character and block device nodes. [...] Link files will have the name of the original file stored.
This means that to store a symlink, all you need to do is add the UNIX extra field block to the data you are writing (these appear to live immediately after the filename is written, and you need to set the extra field length accordingly), and populate its "Variable length data field" with the path you get from readlink. The content you store for the node will be empty.
If you're using a library to generate the zip data (recommended!), it will probably have an abstraction available for that. If not, I'd suggest you put in a feature request!
Of course, most existing zip and unzip utilities follow the same definition, which is why you are able to zip and unzip symbolic links as if they were regular files.
I am building a sync app with a customprovider and a filesyncprovider. I based my provider on this example:
https://code.msdn.microsoft.com/File-Sync-with-Simple-c497bf87
Now I want to extend to a hierarchical folderstructure. So in the EnumerateItems method of the custom syncprovider I return all files and folders just like I did before with only the files in the directory. Now on the filesyncprovider side, this results in a creation of folders with the name of the file and the file being placed in this folder. E.g.
Folder1\textfile.txt\textfile.txt
I have no idea, what I am doing wrong and I find it hard to know the part of the MS filesyncprovider where I could debug to see, what's happening.
My question is, what am I doing wrong and how can I correct it, so that the correct output would be
Folder1\textfile.txt?
Best regards,
Tobias
// Must return the relative path without the filename
public string RelativeDirectoryPath
{
get
{
return _relativeFilePath;
}
Read first - then ask: I returned the path to the file instead of the path to the folder... comment above even warns not to do that...
I would like to compress the folder having the following structure using 7zip or.. using any other zip archive utility available for windows that can use the command line argument in windows and achieve what I want to accomplish.. Here is my folder structure:
/images
/cache
/skin/common/images
/skin/skin1/images
/extensions/extension1/images
What I want to do is, I just want to exclude the top level "images" and "cache" folders from the root folder (The very first two entries in my listing above).
I have tried using the xr!images or xr!/images but.. xr!images removes all images folder. xr!/images just doesn't work.
Can someone please point me to the right direction?
Thanks
Use -x!images instead of -xr!images.
at the moment I'm writing a kind of lib, which gets from outside the file name 'sFilename'. With it data were written to a file it will be created, data were append to an existing file with data, data were updated in an existing file with data or the data were read from an existing data.
The user of the application with my lib should get as much as possible on information about errors of file handling.
For this purpose I code a method FileExists(...) which uses _stat() to determine if a file exists or not and a method "bool checkPermission(std::string sFilename, CFile::EOpenmode iOpenmode)" which gives back a bool if the specified file (sFilename) with the iOpenmode (Read, Write, Readwrite) have the permission to be read, written or read and written.
This method works with _stat(sFilename.c_str(), &buf) too and gives the desired information of the file back in buf.
Before checking any file I want to check if the directory containing the specified file has the desired permissions and for that I want to use the checkPermission method [works with _stat()] for the directory!
Now the problem: how can I determine easyly the containing directory? If the user just give a filename "test.txt" the file will be created or read in working directory. So its easy to get the up-directory. Its the same like the working directory. (And there its simple to use checkPermission to get detailed information about the directory).
But what about when the user not only give the file name? For exaample "....\test.txt" or "dir1\dir2\test.txt". How to combine the working directory with a specific path to gain the up-directory of the file and then to check the permissions?
Phew, I hope all is clear and it was'nt too long ;-)
Rumo
I'd suggest using the Boost FileSystem library at www.boost.org. In particular, check out the path class, which has methods such as make_absolute and parent_path.
This is Windows example code GetFileNameFromHandle to show you how to get the path from a HANDLE. I think it is what you are looking for.
http://msdn.microsoft.com/en-us/library/aa366789%28v=vs.85%29.aspx
I found out that _stat() and _access() doesn't really works for the permissions of the directories. See this stackoverflow page.
With _stat() you can't use ".\" to get information about the current directory. But _access() at least can check if a directory exists as well ".\" or "..\".
In conclusion I use _access() to check the existence of a directory and _stat() to check the permissions of an existing file. If a file should be created I'll check it by doing.
And by the way ;-) I don't need to combine working directory with the user specified file because I can use the specified file alone in _access() to determine if directory exists.
Rumo