From: Robert Klemme <shortcutter@...>
Date: 2008-03-25T17:19:28+09:00
Subject: Re: Find.find sorting: files before directories

2008/3/24, Adam Boyle <briefcase.speakers@gmail.com>:
> > Why do you need that?  I mean, if a folder is excluded then you want to
>  > exclude all files and subfolders, don't you?
>
>
> Yes, an excluded folder would also exclude its children files and
>  folders.
>
>  I'm thinking that I haven't exactly made it clear what my goal is...
>
>  Using Find.find, I want to traverse through a directory structure and
>  make an NSIS-style list of files and their paths for use in an NSIS
>  installer script.  A list of this sort would be best organized if a
>  directory's file children are listed before the directory children.
>
>  Example:
>
> SetOutPath "jruby-1.0.3/docs"
>  File       "jruby-1.0.3/docs/README.rails"
>  File       "jruby-1.0.3/docs/README.coverage"
>  File       "jruby-1.0.3/docs/Readline-HOWTO.txt"
>  File       "jruby-1.0.3/docs/LICENSE.bouncycastle"
>  File       "jruby-1.0.3/docs/LICENSE.ant"
>  File       "jruby-1.0.3/docs/LICENCE.bsf"
>  File       "jruby-1.0.3/docs/Glossary.txt"
>  File       "jruby-1.0.3/docs/getting_involved.html"
>
> SetOutPath "jruby-1.0.3/docs/rbyaml"
>  File       "jruby-1.0.3/docs/rbyaml/README"
>  File       "jruby-1.0.3/docs/rbyaml/LICENSE"
>
> SetOutPath "jruby-1.0.3/docs/jvyaml"
>  File       "jruby-1.0.3/docs/jvyaml/README"
>  File       "jruby-1.0.3/docs/jvyaml/LICENSE"
>  File       "jruby-1.0.3/docs/jvyaml/CREDITS"
>
> ...
>
>
>  The "SetOutPath" lines are the directories, the "File" lines are the
>  files.
>
>  The code you gave will gave results like this (once I used Ruby
>  instead of JRuby :)...):
>
> SetOutPath jruby-1.0.3
>
> SetOutPath jruby-1.0.3/samples
>  File       jruby-1.0.3/samples/xslt.rb
>  File       jruby-1.0.3/samples/thread.rb
>  File       jruby-1.0.3/samples/swing2.rb
>  File       jruby-1.0.3/samples/scripting.rb
>  File       jruby-1.0.3/samples/javascript.rb
>  File       jruby-1.0.3/samples/java2.rb
>  File       jruby-1.0.3/samples/error.rb
>  File       jruby-1.0.3/samples/dom-applet.html
>  File       jruby-1.0.3/samples/applet.html
>  File       jruby-1.0.3/README
>  SetOutPath jruby-1.0.3/lib
>  SetOutPath jruby-1.0.3/lib/ruby
>  SetOutPath jruby-1.0.3/lib/ruby/site_ruby
>  SetOutPath jruby-1.0.3/lib/ruby/site_ruby/1.8
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/ubygems.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/securerandom.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems.rb
>  SetOutPath jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/version.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/validator.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  user_interaction.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/timer.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  specification.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  source_info_cache_entry.rb
>  ...
>
>
>  The issue is that it is very important that the SetOutPath call for
>  any particular file comes directly before it in the list (ie,
>  SetOutPath jruby-1.0.3 for the line File jruby-1.0.3/README).  That
>  way the output path is set correctly when the file is extracted from
>  the installer executable.
>
>  The selected lines from the previous example would ideally be listed
>  this way:
>
> SetOutPath jruby-1.0.3
>
> File       jruby-1.0.3/README
>  SetOutPath jruby-1.0.3/samples
>  File       jruby-1.0.3/samples/xslt.rb
>  File       jruby-1.0.3/samples/thread.rb
>  File       jruby-1.0.3/samples/swing2.rb
>  File       jruby-1.0.3/samples/scripting.rb
>  File       jruby-1.0.3/samples/javascript.rb
>  File       jruby-1.0.3/samples/java2.rb
>  File       jruby-1.0.3/samples/error.rb
>  File       jruby-1.0.3/samples/dom-applet.html
>  File       jruby-1.0.3/samples/applet.html
>  SetOutPath jruby-1.0.3/lib/ruby/site_ruby/1.8
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/ubygems.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/securerandom.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems.rb
>  SetOutPath jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/version.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/validator.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  user_interaction.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/timer.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  specification.rb
>  File       jruby-1.0.3/lib/ruby/site_ruby/1.8/rubygems/
>  source_info_cache_entry.rb
>  ...
>
>
>  It's beginning to seem to me that Find.find just doesn't have an easy
>  way of sorting the elements being traversed.  Any additional help is
>  greatly appreciated.

Thanks for clarifying.  You could do something like this:

require 'find'

base='.'
excludes = []
dirs = Hash.new {|h,p| h[p]=[]}

Find.find base do |path|
  if File.directory? path
    Find.prune if excludes.include? path
  else
    dirs[File.dirname(path)] << path
  end
end

dirs.sort.each do |dir,files|
  puts "SetOutPath #{dir}"
  files.each {|f| puts "File #{f}"}
end

If you want to do printing while traversing then the code becomes more
complicated (either using Find or manual traversal via Dir[]).

Kind regards

robert

-- 
use.inject do |as, often| as.you_can - without end