Loading ...
Sorry, an error occurred while loading the content.
 

Re: [cc2-dev-l] Re: FCW compressed files

Expand Messages
  • Bradd W. Szonye
    ... That s right. That s why you need a script and not just PKZIP to do it. The easiest way to uncompress an FCW is with the PKWare library, but I don t have
    Message 1 of 10 , Mar 23, 2001
      >> On Wed, Mar 21, 2001 at 08:45:27PM -0000, bradds@... wrote:
      >>> However, the basics are there, so now I can convert a compressed FCW
      >>> to an uncompressed FCW on a Unix machine.

      On Fri, Mar 23, 2001 at 01:28:25AM -0700, Mike Riddle wrote:
      > Note that a compressed FCW file is NOT a PKZIP file - the file, after
      > the 128-byte file id block, is compressed using the same algorithms
      > used by PKZIP, but does not have a ZIP file's internal directory
      > structure.

      That's right. That's why you need a script and not just PKZIP to do it.
      The easiest way to uncompress an FCW is with the PKWare library, but I
      don't have $700 to buy the libraries for Windows and Unix for what is
      basically a hobby project. Fortunately, with a little extra work, PKZIP
      can do the job (for decompression anyway).

      The steps involved are:
      1. Extract and save the FCW file ID header.
      2. Wrap the compressed stream with a ZIP header and central directory.
      3. Decompress with PKZIP.
      4. Re-attach the data to the FCW header.

      The tricky part is to do it in such a way that PKZIP doesn't leave
      droppings all over the place.

      > As to PKZIP in UNIX being awful, I can't say, but I have always liked
      > and used PkWare's programs (and Data Compression Library) and have
      > found them stable and fast. And I've *never* had bugs in their
      > products cause me trouble. I really respect their products.

      Oh, PKZIP for Unix seems to work well enough -- it's not buggy or
      anything. It's just that it has some quirks that are acceptable in a
      Windows/DOS environment that are inappropriate on a Unix system. For
      example, every time you run it, it writes a configuration file in the
      current directory. In Unix, that sort of thing belongs in a hidden
      file in your home directory -- dropping files in the current directory
      is not only not always possible, but it's a data integrity and security
      risk.

      I was also going to complain that there's no way to decompress to
      standard output (instead of to a file), but I just found that one! I
      still haven't figured out whether it's possible to supply the ZIP file
      on standard input instead of as a filename.

      In short, it's not that it's a poor program, it's that it has an
      extremely poor Unix command interface. Now, for somebody who's
      accustomed to using PKZIP-DOS, that's an advantage, but it makes it very
      difficult for seasoned Unix users, especially if you're trying to write
      a script.
      --
      Bradd W. Szonye Work: bradd@...
      Software Design Engineer Home: bradds@...
      Hewlett-Packard Cupertino Site, iFL Phone: 408-447-4832
    • Bradd W. Szonye
      ... Well, that turned out not to be the tricky part after all. On the surface, the files appear to decompress correctly, but when I try to actually read the FC
      Message 2 of 10 , Mar 24, 2001
        On Fri, Mar 23, 2001 at 09:50:01AM -0800, Bradd W. Szonye wrote:
        > The steps [to decompress an FCW] are:
        > 1. Extract and save the FCW file ID header.
        > 2. Wrap the compressed stream with a ZIP header and central directory.
        > 3. Decompress with PKZIP.
        > 4. Re-attach the data to the FCW header.
        >
        > The tricky part is to do it in such a way that PKZIP doesn't leave
        > droppings all over the place.

        Well, that turned out not to be the tricky part after all. On the
        surface, the files appear to decompress correctly, but when I try to
        actually read the FC chunks, it falls apart. There are corrupted bits
        here and there, and the files appear to get truncated slightly. Unless I
        make some sort of radical breakthrough, it looks like I'm going to need
        to leave this alone for now. Oh well -- I have lots of other projects
        I'm working on.

        Sorry if I got anyone's hopes up! (Assuming, of course, that anyone
        other than me is interested in this.)
        --
        Bradd W. Szonye Hewlett-Packard: Home:
        Software Engineer bradd@... bradds@...
        408-447-4832 http://www.concentric.net/~Bradds
      Your message has been successfully submitted and would be delivered to recipients shortly.