WEBPACK 0.7beta
===============

What is it?

 Webpack is an open-source command-line tool for automatically packing websites
by shrinking them down without affecting the way they look or behave.
 Webpack is also useful for losslessly shrinking image collections and locating
corrupted files.
 Webpack works by stripping unnecesary information from images, optimising the
compression of images & other compressed files, and removing comments/whitespace
from html.
 This makes for a faster website for your visitors, and lower bandwidth
usage/costs for you! (oh, and it's free - both as in speech and as in beer).

Features
    * Losslessly packs GIF, JPG & PNG images (images are pixel-for-pixel identical, no quality loss)
    * Strips HTML of unneeded comments/whitespace (with HTML::Clean)
    * Optimises compression of ZIP, GZ, & BZ2 files
    * Always copies the original file if it can't be improved (files never get bigger)
    * Prints lots of useful statistics, including size breakdown by image type
    * Maintains original time/date stamps of copied files
    * quiet mode -q (show only errors)
    * brute mode -b (try really hard - makes some files even smaller)
    * update mode -u (update existing destination files if the source file is newer)
    * force overwrite mode -f (replace any existing files)

 Webpack is "safe" in that it is designed to create a compressed replica of a
directory tree, so your original files are not touched.
 Any files it doesn't know what to do with are copied verbatim.
 Essentially you run webpack on your websites original files and upload the
"webpacked" version to your webserver - easy!


How Do I Use It?
1. First, install prerequisites

On your typical Ubuntu based distro you only need this:
sudo apt install libjpeg-progs gifsicle optipng advancecomp

Cygwin - see homepage for a separate package for Cygwin with precompiled binaries

(Read further if that doesn't work for you, otherwise skip to step 2.)
 Webpack relies on several readily available command line tools to process
various types of files.
 You will need to install some (preferably all) of these for it to be useful.
 Most of them either come with your linux distribution or are easy to compile
(usually ./configure; make; make install or less).
 At an absolute minimum, install jpegtran (usually already installed) and gifsicle.
 If you don't have some of these tools installed webpack will just copy the
particular type of file that tool would have processed uncompressed.

Preferred Tools:
    * jpegtran - supplied with most linux distributions as part of the jpeg, libjpeg, libjpeg-progs or jpegtools package
        o Homepage: http://http://jpegclub.org/jpegtran/
    * gifsicle - excellent GIF optimizer
        o Homepage: http://www.lcdf.org/gifsicle/
    * OptiPNG - advanced PNG optimizer
        o Homepage: http://optipng.sourceforge.net/
    * advancecomp - used to optimise compression of gz/zip/png files
        o Homepage: http://www.advancemame.it/comp-readme
    * HTML::Clean - Perl module to compress HTML
        o Homepage: http://search.cpan.org/~lindner/HTML-Clean-0.8/    

Backup Tools (only use these if you can't use or get the above for some reason):
    * pngcrush - this may be supplied with some linux distributions as package pngcrush
        o Homepage: http://pmt.sourceforge.net/pngcrush/
    * giftrans - supplied with many linux distributions as package giftrans
        o Download source: http://ftp.debian.org/debian/pool/main/g/giftrans/giftrans_1.12.2.orig.tar.gz
    
    
2. Install the webpack script

Simply copy "webpack" into ~/bin or /usr/local/bin (whichever is
your preference). You may need to set it executable ("chmod +x webpack")


3. Run It!

a) change into the directory you want to 'webpack' eg "cd ~/mysite"
b) do a "webpack <output directory>" eg "webpack ~/mysitepacked"
.. and that's it - webpack will create a packed replica of the current
directory (and subdirectories) in <output directory>.

Use the -b (brute) option if you want more file reduction at the expense of
speed.

Use the -q (quiet) option if you want to see only warnings/errors.

 Webpack will remind you what to do with no arguments, and there is some error
checking to stop accidents.
NOTE you can't use webpack in-place. It is only intended to create a compressed
copy of a directory tree, because it is DESTRUCTIVE to the files, and because
stuff happens, you could lose your original website files.
Think of it as a compiler - the output is the compressed version of your website.
You wouldn't ask a compiler to destroy your source code!



FeedBack

 Feedback/suggestions/bugreports are encouraged!
 Webpack is in development and works for us, but we need testers and feedback
from you, so please contact us if you use it!
 
Homepage - http://www.kludgesoft.com/nix/webpack.html
Email feedback, suggestions, money etc. to: ezza (at) kludgesoft [dot] com


LICENSE

Webpack is released under the terms of the GNU General Public License (GPL).



History:

WebPack 0.1 alpha 18/04/2004
- first release
- processes gif, jpg & png files only
- limited sanity checking, but won't overwrite anything

0.2 alpha 8/05/2004
- track input & output filesizes
  - print #files, inbytes, outbytes & % reduction
  - separate stats for images/remainder/total
  - break down to individual image/file type
  - print % and byte reduction for each file
- check if file got bigger (or was same size) and copy original if it did
  (prints "!")
  - for jpg, also try non-progressive to see if it's smaller before copying
  original (prints "*")
- try to maintain flags, modified time etc if original was copied
- update/make clear licensing (GPL)

0.3 beta 5/07/2004
- it's now beta because it works for me :)
- implemented options/flags logic
- use HTMLCrunch for compressing html (results not that good unfortunately)
- implement brute mode (-b)
  - always try both interlaced & non-interlaced for jpgs
  - supply -brute flag to pngcrush
- recompress bzip2 files if -9 wasn't used (-9 is default)
  - checks file for what method was used (not sure if this is 100% reliable?)

0.4 beta 19/07/2004
- now also work with gifsicle for GIFs (WAY better than giftrans)
- quiet mode (-q) - show only errors/warnings
- show time taken in m:s
- simplify output a little

0.5 beta 13/09/2004
- catch (more) pngcrush failures
- minor cleanups
- more sanity checking
- implement process-existing-files-only-if-source-newer-than-target (-u)
  - now can use in an automated way to keep a website updated
- implement force overwrite (-f) (hidden for the moment)

0.6 beta 4/10/2004
- WAY more sanity checking
  - check if source or destination is the root directory
  - check if output is same as, subdirectory of or parent of current directory
  - now checks all stupid combinations of source/destination directories I could imagine
- enable force overwrite (-f) option
- works with (more) paths with spaces in them
- add support for HTML::Clean for html compression
- add support for advancecomp (compression optimizer):
  - use advdef to recompress *.gz (and *.png if pngcrush not installed)
  - use advzip to recompress *.zip
  - brute mode (-b) enables -4 (compress extreme) option on advdef/advzip

0.7 beta 20/02/2018
- Drag webpack kicking and screaming into the current decade and put up on github for others to play with
- Add support for OptiPNG which seems to work better than pngcrush, so OptiPNG is the default if both are installed
- Remove support for htmlcrunch as it's no longer maintained (website doesn't exist)
- Don't strip JPG files of additional metadata to preserve EXIF tags etc.
- Remove pngcrush from cygwin distribution as it's available natively from the cygwin setup

  

To-Do

This is a rough list of things to implement or suggestions from users:

    * Investigate other modern tools, e.g. jpegoptim, libhtml-packer-perl
    * Debug/verbose mode (print input & output files, directory creation etc)
    * Find/create a better html compressor (better than htmlcrunch)
    * Perform duplicate checking of output dir & inform user of duplicates

   
 Let us know if you find this useful and if you have suggestions/patches/
praise/bugreports for it as they are most welcome.

 Regards, the KludgeSoft development team.

www.kludgesoft.com