facets 1.0.0 → 1.0.1

data/CHANGELOG

@@ -1,254 +0,0 @@
-= Facets Change Log
-
-== 1.0.0
-* Sped up string#similarity.
-* By popular protest deprecated usage of "AClass.use Facets, :amethod".
-* Created nicer layout of facets/support, /group, /english (one day /method).
-* Odd sets of facets have been move to facets/group/, eg. facets/group/inflect.
-* All is now well preped for a Rolls release shoud that come about.
-
-== v0.10.30  Halloween
-* Change OrderedHash (ohash.rb) to Dictionary (dictionary.rb).
-* Change BlankSlate (blankslate.rb) to BasicObject (basicobject.rb).
-* Old versions of the above should still work, but throw warnings
-  and will stop working come next version.
-* 1.rb has been renamed to one.rb (I know, two more characters,
-  but you can do it! ;)
-* Methods enumerable/op_mod (%) and enumerable/eF are deprecated.
-  Use #every or #ew instead.
-
-== v0.10.11  Palapable
-* Merged Mega into Facets. I put the old Mega ChangeLog at
-  the bottom of this document.
-* This represents the One-Point-Oh realease (albeit RC1) of
-  Facets, but from now on I'll be using date versioning
-  --if you look at the versions below you may understand why.
-  The versioning sequence has no real rationality.
-  Instead I will use this simple Rational Versioning Policy:
-  * If YYYY-MM-DD it's a development release.
-  * If YYYY-MM it's an official release.
-  * If just YYYY it's hella stable.
-  One might argue this is not as robust as a compatability
-  significant point-based versioning system, but I'd say that's
-  all well and good in theory, now give me something that actually
-  works in practice.
-
-== v0.9.5  Rebirth
-* Rebirth of Facets! As cool as the name Nano is, it became clear
-  that most people are drawn toward the name Facets. So we are
-  reverting back. Nonetheless Nano will remain a viable alias for
-  Facets.
-* Must now use <code>require 'facets'</code> prior to using
-  other Facets. This does two things. First it loads the 
-  base methods, which ensures that all programs have them 
-  readibly available when using Facets. Secondly it supplements
-  #require to handle aliases. In essence, require 'facet/...',
-  require 'facets/...' and require 'nano/...' all point to the
-  same place(s). (NOTE This is a precursor to a more universal
-  system that will handle aliasing automatically.)
-* kernel/require_esc has been removed. #require_nano will be also
-  in the next release. These will no longer be needed thanks to 
-  the aliasing system.
-
-== v0.9.2  Littles
-* Added string/starts_with? and string/ends_with?
-* Fixed module/memoize to cache on a per class/module bases.
-* Added module/is as an alias for /include.
-* Added module/shadow_method and shadow_all.
-* Renamed module/superup (which was module/supers) to module/as. And defined a new
-  method called /superior which is like /super but skips to a sepecified ancestor.
-* Renamed module/super_send to module/send_as.
-
-== v0.9.1  Wraps
-* Added module/nesting (not to be confused with the class method Module::nesting).
-* Thanks goes to Gavien Kistner for new string/word_wrap methods 
-  (with additioanl thanks to Dyane Borderson for his suggestions).
-* Added hash/each_with_key and each_with_index, as well as array/each_with_key;
-  increasing polymorphism between the two classes.
-* Moved array/each_pair to enumerable/each_pair and changed how it worked to 
-  what one might expect.
-
-== v0.9.0  NotUorI
-* Deprecated object#special_class, which was a method for (class<<self;self;end).
-  Already have too many aliases for this, and though the name is fitting, the simpler
-  method #own is better (alternative: #singleton).
-* Fixed enumerable/uniq_by file which was misnamed 'unique_by.rb'.
-* BIG CHANGE! Got rid of URI encoding on file names and sub'd a converson table
-  of english readable names. Thanks Jeff Wood, Florain Gross, Dav Burt,
-  Gavin Kistner, James Edward Gray II, Brian Schröder, Mauricio Fernández,
-  David A. Black, Gavin Sinclair, Nikolai Weibull and Christian Neukirchen.
-
-== v0.8.3  UorI
-* Added kernel/uri and kernel/unuri
-* (Good idea?) removed methods module/sattr, sattr_reader, &c. Instead created object/meta
-  in which attr, attr_reader, &c. are made public. So, sattr -> meta.attr.
-* Added array/unzip
-* Added logger/format and logger/format_message (stub).
-* Added string/dequote.
-* Class method file names now begin with '::' to separate them from instance methods.
-
-== v0.8.2  Nano Revolution
-* Name Change! What was Ruby Facets is now Nano Methods!
-* New minor version as there has been a change of plans with the integration of what was Carats.
-  The classes and modules have been reseperated into their own project called Mega Modules.
-* Modified the #firstxxx/#lastxxx methods to be more congruent and comprehensible.
-* Renamed object/supers to object/superup. Better!
-* Got rid ot symbol/gen. Use object/generate_method_name or  module#generate_instance_method_name
-  instead ( Is that the longest non-foo method name ever? :)
-* The file 1st.rb has moved to object/method. This modifies #method to persist the returned
-  Method object. The original non-persitent version of #method is aliased to #original_method.
-  (This arrangement is not set in stone though, and is still be considered.)
-* Added module/clone_using, module/clone_ranaming and module/clone_removing.
-* Added module/integrate (I love this one!)
-* Added many new enumerable methods: #filter_map, #compact_map, #commonality, #frequency,
-  #probability and #find_collisions.
-* Moved some methods from array to enumerable: #each_permutation, #each_combination, 
-  #each_unique_pair and the class method ::combinations.
-* Added string/bytes per ruby-dev summary 26385-26467.
-* Moved #cattr_reader, #cattr_writer and #cattr_accessor from Class to Module.
-* Renamed numeric/octet_units to numeric/bytes_to_s and added a #bits_to_s as well.
-* Added class/descendants (alias subclasses) and class/remove_descendents, as well as
-  object/descendents_of and object/remove_descendents_of.
-* Added methods to comparable: #cap, #clip. There are also #at_least and
-  #at_most, although FYI #cap and #clip handle the same functionality.
-* Removed string/table_name and string/class_name (they were too Rails specific).
-  Note: A new module has been added to Mega Modules, called orm-inflect.rb. It contains numerous
-  methods for doing this sort of thing.
-* Added numeric/ceil_multiple (not sure about name though, perhaps a better name like "ceil_to"?).
-* Added string/soundex for calculating the soundex code of a word/name.
-* Added io/expect. I thought Ruby already had this but can't find it, so here it is for now.
-* Placed Florian's binding-of-caller.rb in binding/of_caller. Makes more sense there.
-* Fixed bug with Continuation::create (it wasn't defined as a class method and should have been).
-* Fixed bug with string/capitalized? which wasn't always giving the correct result.
-
-== v0.7.2  George II
-* Added a class method dir/recurse which allows one to loop through a dir and all its
-  subdirs, etc. It also has an alias #ls_r. Thanks goes to George Maschovitis for this.
-* Changed array/permute to array/each_permutation.
-
-== v0.7.1  Quick George
-* Added facet.rb, although expiremental it makes it possible to use atomic methods without
-  specificaly requiring them. It uses Object#method_missing to require them as needed.
-* Improved on "molecules", i.e. files that require numerous related atoms in a single go.
-  These will see a great deal of continued improvement in the future.
-
-== v0.7.0  Georgian Transform
-* All methods with names containing non-alphanumeric characters now have facet
-  files without those special characters. This removes some incompatibilites with 
-  certain systems (including Windows). All such symbols have been replaced with
-  CGI escape sequences, for instance 'in?.rb' becomes 'in%3F.rb'. To avoid having
-  to use these "ugly" names, a new method has been added, kernel/require_facet.
-* Methods that were grouped together in the same file have been separated 
-  into their own files. Pure Atomicity has been achieved!
-* The method array/put has been dropped in favor of array/top (as a better alias for #unshift).
-* Methods #matchdata/post_match_from and #matchdata/pre_match_from have deprecated
-  in favor of more generalized methods that serves the same purposes,
-  matchdata/matchtree and matchdata/matchset.
-* Added module/include_as.
-* Added array/permute.
-
-== v0.6.3  Gemstone
-* Gem is now avaliable!
-* Added kernel/require_all which allows one to require a pattern of files,
-  making it easy to require every file in a directory.
-* Added hash/traverse and hash/traverse!, which takes a block and iterates over
-  each key-value pair descending into subhash values and applying the block.
-  (Thanks goes to Ara T. Howard and robert for their help.)
-* Added hash/collect. This uses enumerable/graph so that
-  hash/collect will return a hash rather then an array.
-* Added numeric/before and numeric/after in place of #ago and #since
-  and aliased #ago and #since to them, per the suggestions of
-  Francis Hwang.
-* Added "poor man's profiler", time/elapse. (Thanks goes to Hal Fulton).
-* Changed enumerable/build_hash to enumerable/graph. The method #build_hash
-  has been aliased to #graph for the time being for backward compatability.
-* Added array/last_index. (Thanks goes to Jabari)
-
-== v0.6.2  Refinement I
-* Added kernel/resc as a shortcut for Regexp.escape.
-* Renamed hash/keys_to_string to hash/key_to_s and  
-  hash/keys_to_symbol to hash/keys_to_sym to be more consistant
-  with other methods and what these methods specifically do.
-  The old names have been deprecated.
-* Added hash/has_keys?, which allows for checking mutliple keys at once.
-  Also includes #has_only_keys?
-* Added Rail's Time and Byte modules for Numeric with
-  numeric/times.rb and numeric/bytes.rb (I've been informed that
-  thanks go to Richard Kilmer for this. Thanks!)
-* Added numeric/octet_unit which utilizes numeric/bytes.
-  (The name of this method may change in the future though.)
-* Added enumerable/uniq_by.
-* Added module/abstract.
-* Added module/redirect which is similar to alias, but does not
-  copy the method, but rather wraps it. It also takes a hash so
-  multiple methods can be redirected all at once.
-* Added string/shatter which is similar to split but includes
-  the matched portions of text.
-
-== v0.6.1  Florian's Mixes
-* Updated the current set of mix files which were still from 
-  version 0.5.0. (2004-12-31)
-
-== v0.6.0  Florian's Onslaught
-* Major revisions made by Florian's onslaught. (2004-12-28) 
-* First general public release.
-
-== v0.5.0  Fit for First
-* This is the beginning of offical releases. (2004-12-23)
-* Now called Facets.
-* Changed layout to be completely atomic! (2004-10-30)
-
-== v0.4.0  Raspberry
-* Scoured the Ruby world for useful additions.
-* Changed name again from A.B.C. to Raspberry.
-  I can't seem to settle on a name. Even the new 
-  subsections have changed five times. (2004-08-08)
-
-== v0.3.0  ABC
-* Reorganinzed all of this together in a nice neat way.
-* Changed name of library from Succ to A.B.C. 
-  which stands for All Base Common. (2004-02-02)
-
-== v0.2.0  Succ
-* A number of new methods added.
-* Renamed it Succ for "successor". (2003-05-31)
-
-== v0.1.0  TomsLib
-* Tomslib's just a few useful methods. (2002-07-01)
-
-
-== Old Mega/Carats ChangeLog
-
-== v0.4.0
-* Adjusted Tuple to use Tuple::[*args] instead of ::new, which now takes
-  a single argument instead, either an Array or a String which it splits.
-* Removed services.rb unitl in gets fixed and a better name.
-
-== v0.3.3
-* Added inheritable.rb, a lib for creating class vars inherit from ancestors
-* Addes annotations.rb, a lightweight metadata system good for annotating
-  methods esspecially. It is built on top of inheritable.rb.
-
-== 0.3.2
-* Added services.rb, a lib for managing methods as first class objects
-* Fixed/cleaned-up orm_supprt.rb
-
-== 0.3.1
-* Aliased #autoload_classes with #autorequire.
-* Added #strfbits and #strfbytes to binary_multiplers.rb.
-
-== 0.3.0
-* Name change to Mega Modules!
-* Added a few important files: multipliers.rb, binary_multipliers.rb,
-  time_in_english.rb and orm_inlectors.rb.
-* Each one of these is a "Methods Module", i.e. They are a collection of
-  closely related methods that modify one or more core classes/modules.
-  Individually these methods could have very well ended up in Nano Methods,
-  but given there quantity and interrelation they were more suitable to a
-  module, and thus placed here.
-
-== 0.2.0
-* Initial public release
-
-== 0.1.0
-* Early Development as Ruby Carats

data/Rakefile

@@ -1,173 +0,0 @@
-
-# $:.unshift('lib')
-
-#require 'uri'
-require 'yaml'
-require 'fileutils'
-
-require './lib/facets/2005-11-05/ext/file/self/read_list'
-
-VERSION = "2005-11-05"
-REQPATH = "facets/#{ver}"
-DATPATH = "data/facets"
-ETCPATH = "conf/facets"
-MTHPATH = "lib/facets/#{ver}/ext"
-GENPATH = "lib/facets/#{ver}/set/general"
-
-#-----------------------------------------------------
-# Generate facet/ redirect to nano/, mega/ and unit/
-#-----------------------------------------------------
-desc %{Build facet redirect directory}
-task :facet do
-  sh %{rm -R './lib/facet'}
-  ['ext','set/special','set/general'].each{ |d|
-    from = "lib/facets/2005-11-05"
-    rbs = Dir.glob( File.join( from, d, '**/*.rb' ) )
-    rbs.each { |e|
-      nolib = e.gsub( /^lib\//, '' )
-      re = Regexp.new( Regexp.escape( File.join( from, d ) + '/' ) )
-      link = e.gsub( re, 'lib/facet/')
-      FileUtils.mkdir_p( File.dirname( link ) )
-      File.open( link, "w+" ) { |w|
-         w << "require '#{nolib}'"
-      }
-    }
-  }
-end
-
-#-----------------------------------------------------
-# Read in the facets/ext directory and sort-out per class
-#-----------------------------------------------------
-Methods = {}
-task :methods do
-  Dir.chdir(MTHPATH) do
-    dg = Dir.glob('*/**/*.rb')
-    dg.each { |f|
-      klass, *meth= f.split('/')
-      Methods[ klass ] ||= []
-      Methods[ klass ] << "#{meth.join('/').chomp('.rb')}"
-    }
-    Methods.keys.each { |k| Methods[k].sort! }
-  end
-end
-
-#-----------------------------------------------------
-# Save method index to file in yaml format
-#-----------------------------------------------------
-desc %{Generate method index file}
-task :index => [ :methods ] do
-  print "Creating method index... "
-  loc = File.join( DATPATH, 'index.yml' )
-  File.open( loc, 'w' ) { |f|
-    f << Methods.to_yaml
-  }
-  puts "[done]"
-end
-
-#-----------------------------------------------------
-# Create per class feature files
-#-----------------------------------------------------
-desc %Q{Generate per-class feature files}
-task :class => [:methods ] do
-  print "Creating per class files... "
-
-  index  = Methods 
-  noauto = File.read_list( File.join( ETCPATH, 'noauto' ) )
-  allfiles = index.collect{ |k,ms| ms.collect{ |m| "#{k}/#{m}" } }.flatten
-  full = calc_full(allfiles, noauto)
-
-  index.each { |classname, method_files|
-    File.open( "#{GENPATH}/#{classname.downcase}.rb", 'w' ) { |f|
-      method_files.each { |rf|
-        mfn = "#{classname.downcase}/#{rf}"
-        if full.include?(mfn)
-          f << %Q{require '#{REQPATH}/#{mfn}'\n}
-        end
-      }
-    }
-  }
-  puts "[done]"
-end
-
-#-----------------------------------------------------
-# Do all prep tasks (do this to generate data index
-# file  and method group files)
-#-----------------------------------------------------
-desc %{Do all preperation (index, class)}
-task :prep => [ :index, :class, :facet ] do
-end
-
-#-----------------------------------------------------
-# List Methods
-#-----------------------------------------------------
-desc %{Output a list of all methods}
-task 'list' => [ :methods ] do
-  puts method_list_outline
-end
-
-# #-----------------------------------------------------
-# # HTML List MethodsFile.new(
-# #-----------------------------------------------------
-#
-# HEAD = <<-HERE
-# <html>
-# <head>
-#   <style>
-#     pre#methods {
-#       font-size: .8em;
-#       line-height: 1em;
-#     }
-#   </style>
-# </head>
-# <body>
-# <pre id="methods">
-# HERE
-# 
-# FOOT=<<-HERE
-# </pre>
-# </body>
-# </html>
-# HERE
-# 
-# desc %{generate method list html file}
-# task 'listHtml' => [ :scrape_methods ] do
-#   s = HEAD + method_list_outline + FOOT
-#   File.open('pub/www/methods.html', 'w') { |f| f << s }
-# end
-
-=begin
-desc "Check for open comment tests"
-task :testchk do
-  system %q{grep -r ^#=begin\ test ./lib}
-end
-=end
-
-#-----------------------------------------------------
-# Support Methods
-#-----------------------------------------------------
-
-private
-
-def method_list_outline
-  s = ''
-  Methods.keys.sort.each { |k|
-    ms = Methods[k].collect{|m| m.gsub(%r{^self\/},'::')}.sort
-    s << "#{k}:\n"
-    ms.each { |m| s << "  #{m}\n" }
-  }
-  s
-end
-
-# Full set of require files (allfiles minus noauto and unsafe)
-def calc_full( allfiles, noauto )
-  sf = allfiles.dup
-  noauto.each do |f|
-    if f.include?('*')
-      re = %r"#{f.gsub('*','.*')}"
-      sf.reject!{ |x| x =~ re }
-    elsif sf.include?(f)
-      sf.delete f
-    end
-  end
-  sf.sort
-end

data/TODO

@@ -1,46 +0,0 @@
-
-= Facets' To Do List
-
-IT IS ALWAYS A GOOD IDEA TO WRITE A TEST!
-AND ITS ALWAYS A FINE THING TO IMPROVE DOCUMENTATION!
-
-* Move methods to facets/method and link back to facet/ but prefer library alias
-  instead. (Rolls)
-
-* hash/invert: What about letting Hash#invert take a block that specifies
-  what to do when a collision occurs?
-
-* hash/at_rand: allow for a _number_ of values, keys and pairs to be returned.
-
-* > Yep. Gone. But I may replace with a version that deletes only the first 
-  > matching element. Thoughts?
-  If you don't want to do that I would rather add an option hash to .delete and
-  use :once => true or something similar. (More expressive)
-
-* numeric/bound_by: Numeric includes Comparable. Maybe drop this, but 
-  might it be reasonable for Numeric but no other comparables?
-
-* nil/[]: Why should nil respond to [], but not to .each? I don't think 
-  nil should be treated as a container.
-  nil/length: See above.
-
-* nil/method_missing: I'm not so sure about this. There is a lot of 
-  libraries depending on nil raising exceptions on failed method access. I 
-  also think that this could make programming errors very hard to find.
-
-* fileutils/head&tail: Hm, I think the block forms ought to be equivalent to 
-  tail -f and so on. Then the block forms should also return all lines by default
-  instead of only the first 10.
-
-* fileutils/which: Should probably be based on fileutils/where_is as they have 
-  lots of common code. The two methods should probably also use RUBY_PLATFORM instead
-  of checking for ALT_SEPARATOR.
-
-* module/memoize: Should the main memoize method be defined as it is? 
-  (see bottom of file).
-
-* kernel/demo: It can be implemented as demo "1+1"  using Binding.of_caller. 
-  Using only the block is possibly with SCRIPT_LINES__ or File access and is quite ugly.
-
-* hash/weave: Document how does it compare to Hash#merge?
-