appwiki:compression [ShiningWiki]

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== File name encoding problem in File Compression Method ======

  * Problem: specially when cross-platform, file name encoding is difference in different compression tool, like Zip, Gz, Bz2.
    * Solution 1: using tar to compress
    * Solution 2 in linux: unzip -O CP936 non_english_name.zip (means using GBK, GB18030 Chinese code)
    * Solution 3 use Java: jar xvf non_english_name.zip
      * ref: http://www.111cn.net/sys/linux/72590.htm

  * Problem: when zip or winrar uncompress a non-english encoding archive file, sometimes require set system locale language and reboot to get name uncompressed right.
    * Solution 1 (windows method): using already-to-use-build zip and unzip tool from DotNetZip library, which support encoding and decode option <code>Unzip.exe -cp 936 chinese_name_content.zip</code>
      * download and it's under its tool folder: https://dotnetzip.codeplex.com/
      * ref: http://www.chengxuyuans.com/Ruby/41584.html
      * windows code page: https://en.wikipedia.org/wiki/Windows_code_page
        * 936 and 1386 for GBK
        * 932 and 943 for shift JIS
    * Solution 2 (cross platform) using Python (sometime works, sometimes error encoding): <code>python xZip.py non_english.zip decode_language</code> (such as gbk for Chinese, decode_language code refer to this https://docs.python.org/2/library/codecs.html )
      * here is the python code for xZip.py<code python xZip.py>
# full list of codec: https://docs.python.org/2/library/codecs.html
# note: 
# - input from command line is using commandline system default locale encoding
# - it read the zip file path in unicode format with the given decode method
# - if you use python print method to print those unicode path in window command windows,
#   it may error when system default locale codec can't print those unicode characters
import zipfile   
import os.path   
import os
import sys
   
class ZFile(object):   
    def __init__(self, filename, mode='r', basedir=''):   
        self.filename = filename   
        self.mode = mode   
        if self.mode in ('w', 'a'):   
            self.zfile = zipfile.ZipFile(filename, self.mode, compression=zipfile.ZIP_DEFLATED)   
        else:   
            self.zfile = zipfile.ZipFile(filename, self.mode)   
        self.basedir = basedir   
        if not self.basedir:   
            self.basedir = os.path.dirname(filename)   
          
    def addfile(self, path, arcname=None):   
        path = path.replace('//', '/')   
        if not arcname:   
            if path.startswith(self.basedir):   
                arcname = path[len(self.basedir):]   
            else:   
                arcname = ''   
        self.zfile.write(path, arcname)   
              
    def addfiles(self, paths):   
        for path in paths:   
            if isinstance(path, tuple):   
                self.addfile(*path)   
            else:   
                self.addfile(path)   
              
    def close(self):   
        self.zfile.close()   
          
    def extract_to(self, path, decode):   
        for p in self.zfile.namelist():   
            self.extract(p, path, decode)   
              
    def extract(self, filename, path, decode):   
        if not filename.endswith('/'):   
            f = os.path.join(path, filename.decode(decode))   #gbk,gb18030, GB2312, utf-8
            dir = os.path.dirname(f)
            if not os.path.exists(dir):
                os.makedirs(dir)   
            file(f, 'wb').write(self.zfile.read(filename))   
              
          
def create(zfile, files):   
    z = ZFile(zfile, 'w')   
    z.addfiles(files)   
    z.close()   
      
def extract(zfile, path, decode):   
    z = ZFile(zfile)   
    z.extract_to(path, decode)   
    z.close() 

if __name__=="__main__":
    extract(unicode(sys.argv[1]), u'.', sys.argv[2])
</code>

  * Alternative solution: extract normally with wrong-encoding names, then fixing those name using python decode and encode

  * Site Notes:
    * in windows commands, chcp is used to change display page code (file name encoding) [[https://msdn.microsoft.com/en-us/library/windows/desktop/dd317756%28v=vs.85%29.aspx?f=255&MSPPError=-2147217396|page code list]]


  * additional reading:
    * http://www.docin.com/p-739332424.html
    * http://www.cnblogs.com/qq78292959/archive/2013/03/27/2985310.html
    * https://allencch.wordpress.com/2010/12/06/how-to-extract-zip-file-which-contains-filenames-with-shift_jis-encoding-in-ubuntu/
    * https://www.mkssoftware.com/docs/man1/unzip.1.asp

====== Common Problem on compressed File and Solution ======

  * Problem: Winrar has update the version recently, only winrar can't open some new winrar file.
    * Solution: get latest 7z to uncompress it, will be fine. http://www.7-zip.org/

====== Winrar ======

  * extract with cmd <code>unrar x Pack.rar</code>