parsely 0.1.3 → 0.1.4

data/.travis.yml

@@ -0,0 +1,38 @@
+# Passes arguments to bundle install (http://gembundler.com/man/bundle-install.1.html)
+#bundler_args: --binstubs
+
+# Specify which ruby versions you wish to run your tests on, each version will be used
+rvm:
+  - 1.9.2
+  - ruby-head
+
+# Define how to run your tests (defaults to `bundle exec rake` or `rake` depending on whether you have a `Gemfile`)
+#script: "bundle exec rake db:drop db:create db:migrate test"    
+
+# Define tasks to be completed before and after tests run . Will allow folding of content on frontend
+#before_script:
+#  - command_1
+#  - command_2
+
+#after_script:
+#  - command_1
+#  - command_2
+
+# Specify an ENV variable to run before: 'bundle install' and 'rake' (or your defined 'script')
+#env: "RAILS_ENV='test' " 
+
+# Specify the recipients for email notification
+#notifications:
+#  recipients:
+#    - email-address-1
+#    - email-address-2   
+#  disabled: true # Disable email notifications
+
+# Specify branches to build
+# You can either specify only or except. If you specify both, except will be ignored.
+#branches:
+#  only:
+#    - master
+#  except:
+#    - legacy
+

data/LICENSE

@@ -0,0 +1,12 @@
+THE DRINK-WARE LICENSE (forked from BEER-WARE r42)
+
+Gabriele Renzi (http://www.riffraff.info) 
+wrote this code, whereas not specified otherwise. 
+
+As long as you retain this notice you can do whatever you want with this stuff. 
+If we meet some day, and you think this stuff is worth it, you can buy me 
+a beer, coffee, palinka shot or any other drink in return.
+
+This program is free software; you can redistribute it and/or modify
+it under the terms of the Artistic License version 2.0.
+ 

data/README.mkd

@@ -0,0 +1,45 @@
+# DISCLAIMER
+
+Do Not Use This. 
+At the moment it is mostly a code dump, and a bad one at
+that, while I slowly add the stuff I need. One day I will refactor the
+code and add useful options and make it reliable and it may become a
+useful project for other people. As it stands, it is not, just move on.
+
+
+## Summary
+parsely is a tool to extract and manipulate text files.
+Basically, it allows you to run ruby one liners (think `-n/-p`) with some additional
+shortcuts. 
+
+parsely is intended as a replacement for all those single-use-and-discard scripts 
+in sed/awk/perl/ruby that I constantly end up rewriting, such as counting frequencies, 
+summing fields, selecting (c,t)sv rows by field values etc
+
+It does nothing you can't do with a few pipes, sed, awk, grep, ack, perl, 
+ruby, sort, uniq, bc, ministats and comm.
+
+It is useful for me because 
+* I am very bad at remembering options for command line tools, and get
+  confused when BSD and GNU tools don't match
+* I always get confused escaping stuff in the shell
+* I have written or googled a freq.awk a dozen times
+
+This is most likely useless to you.
+
+## INSTALLATION
+
+Running
+
+    gem install parsely
+
+should be enough to install. 
+I use ruby (YARV) 1.9.2 and have not tested this anywhere else.
+
+## SUPPORT
+
+Open a ticket at http://github.com/riffraff/parsely/issues if you want
+something in parsely, but I don't think you should use this tool, at
+least for the next couple of years. 
+Or you can write me an email at rff.rff+parsely@gmail.com if you want.
+

data/Rakefile

@@ -17,3 +17,8 @@ task :newtest, :name do |t, args|
   new = new + '-' + args[:name] if args[:name]
   cp_r 'test/basic/'+old, 'test/basic/'+new
 end
+
+desc 'run tests with profiler'
+task :profile do
+  sh "ruby-prof test/cli-runner.rb"
+end

data/TODO

@@ -3,7 +3,7 @@
 - _2 in dump by _1 in other
 + _1 if /www/
 + _1 if _2 =~ x
-- .parselyrc
++ .parselyrc
 - split by: if a X or not, or if _1 in other_file (subsides comm)
 + in "foo 1\nbar 2" "sum(_2) if _2 < 3" to see how much I can delete
 - _1._1 for submatch (e.g. user agent)
@@ -18,3 +18,7 @@ skip jsonp bit
 summaryse is a cool extension from which to steal stuff. Maybe provide basic functionalities through that
 - given [/m/09_pbpl      /type/object/key        /soft/isbn/9780491035453,..] and [/m/077601h      /book/isbn/book_editions        /m/09_pbpl] get [/m/09_pbpl /soft/isbn/9780491035453]
 - use merge sorted and merge unsorted probably
+- find a simpler way to do aggregates
+- top k
+selext top/max N or N% ditto for minus
+select outliers

data/lib/parsely.rb

@@ -1,13 +1,22 @@
 require 'set'
 require 'English'
 $OUTPUT_FIELD_SEPARATOR = ' '
-
+module Kernel
   def p args
     STDERR.puts(args.inspect) #if $DEBUG
   end
+end
 
 class PseudoBinding
   class PerlVar < String
+    # this is not defined in terms of <=>
+    def == other
+      if other.is_a? Numeric
+        to_f == other
+      else
+        super
+      end
+    end
     def <=> other
       if other.is_a? Numeric
         to_f <=> other
@@ -18,11 +27,24 @@ class PseudoBinding
     def inspect
       "PerlVar(#{super})"
     end
+    #unneed as of now
+    #def coerce something
+    #  [something, to_f]
+    #end
+    def + other
+      case other
+      when Numeric
+        PerlVar.new((to_i + other).to_s)
+      when String
+        PerlVar.new((to_s + other).to_s)
+      end
+    end
   end
   PerlNil = PerlVar.new ''
   attr :line
+  attr :vals
   def initialize lineno, vals
-    @line, @vals = lineno, vals.map {|x| PerlVar.new(x)}
+    @line, @vals = PerlVar.new(lineno.to_s), vals.map {|x| PerlVar.new(x)}
   end
   def method_missing name, *args
     if args.empty?
@@ -36,6 +58,11 @@ class PseudoBinding
     end
   end
 end
+class Array
+  def value
+    self
+  end
+end
 class String
   def value
     to_s
@@ -48,33 +75,18 @@ class Proc
   end
 end
 class Parsely
-  VERSION = "0.1.3"
+  VERSION = "0.1.4"
   def self.cmd(&block)
-    klass = Struct.new :value, &block
-    klass.class_eval do 
-      def process(items)
-        value.assign(items)
-        _process(value)
-      end
-    end
-    klass
+    Struct.new :value, &block
   end
   RGX= /"(.*?)"|\[(.*?)\]|([^\s]+)/
-  Value = Struct.new :index, :value do
-    def assign(items)
-      self.value = items[index]
-    end
-    def process(items)
-      items[index]
-    end
-    def to_i
-      value.to_i
-    end
-    def to_f
-      value.to_f
+
+  Expression = Struct.new :code, :items do
+    def process(pb)
+      result = pb.instance_eval(code)
     end
     def to_s
-      value.to_s
+      code.to_s
     end
   end
   Ops = {
@@ -85,27 +97,27 @@ class Parsely
         @result = proc { @running_value }
         @result.single = true
       end
-      def _process(value)
+      def process(value)
         if value.to_f < @running_value  
           @running_value = value.to_f
         end
         @result
       end
     end,
-      :max => cmd do
+    :max => cmd do
       def initialize index
         super
         @running_value = Float::MIN #-Inf would be better
         @result = proc { @running_value }
         @result.single = true
       end
-      def _process(value)
+      def process(value)
         if value.to_f > @running_value
           @running_value = value.to_f
         end
         @result
       end
-      end,
+    end,
     :sum => cmd do
       def initialize index
         super
@@ -113,7 +125,7 @@ class Parsely
         @result = proc { @running_value }
         @result.single = true
       end
-      def _process(value)
+      def process(value)
         @running_value += value.to_i
         @result
       end
@@ -126,7 +138,7 @@ class Parsely
         @result = proc { @running_value/@running_count.to_f }
         @result.single = true
       end
-      def _process(value)
+      def process(value)
         @running_value += value.to_i
         @running_count += 1
         @result
@@ -137,7 +149,7 @@ class Parsely
         super
       	@running_freqs = Hash.new(0)
         @running_count = 0
-	as_ary=nil
+	      as_ary=nil
         @result = proc do
           if as_ary.nil?
             as_ary=@running_freqs.sort_by do |k,v| [-v,k] end.each
@@ -146,7 +158,7 @@ class Parsely
           [v, k]
         end
       end
-      def _process(value)
+      def process(value)
         @running_freqs[value.to_s]+=1
         @running_count += 1
         @result
@@ -180,39 +192,32 @@ class Parsely
           cached.next
         end
       end
-      def _process(value)
+      def process(value)
         @running_values << value.to_i
         @result
       end
     end,
   }
+  PseudoBinding.class_eval do
+    Ops.each do |k,v|
+      #instantiating the object is expensive and we are not using 99% of them
+      obj = nil
+      define_method k do |values|
+        obj ||= v.new(nil)
+        obj.process(values)
+      end
+    end
+  end
 
   def parse(expr)
     val, cond = expr.split(/ if /)
                               # p [ val, cond]
-    elems=val.split
-    r=elems.map do |e|
-      case e
-      when /(\w+)\(\_(\d+)\)/
-        opname = $1.to_sym
-        klass=Ops[opname]
-        if klass.nil?
-=begin
-          if respond_to? opname
-            klass = cmd do
-              def _process(value)
-                send opname, value
-              end
-            end
-          else
-=end
-          abort "unknown op '#$1'"
-        end
-        klass.new(Value.new($2.to_i))
-      when /\_(\d+)/
-        Value.new($1.to_i)
-      end
+
+    val = '['+val+']'
+    if val =~ /([\(\)\w])( ([\(\)\w]))+/
+      val = val.split(" ").join(",")
     end
+    r = Expression.new(val)
     [r, parse_cond(cond)]
   end
 
@@ -223,7 +228,7 @@ class Parsely
     when nil, ''
       proc { |bnd| true }
     else
-    proc { |bnd| bnd.instance_eval(str) }
+      proc { |bnd| bnd.instance_eval(str) }
     end
   end
 
@@ -250,7 +255,7 @@ class Parsely
   end
 
   def main_loop(expr,lines)
-    ast, cond =parse(expr)
+    expression, cond =parse(expr)
     result = []
     result = lines.map.with_index do |line, lineno|
       line.chomp!
@@ -263,15 +268,14 @@ class Parsely
       #XXX ugly
       next unless items
       b = PseudoBinding.new(lineno, items)
-      ast.map do |a| 
-        a.process(items) if cond[b]
-      end 
+      expression.process(b) if cond[b]
     end
     last = []
     result.each do |cols|
+      next if cols.nil? #when test fails
       result_line = cols.map do |col|
         next if col.nil?
-        col.value 
+        col.value
       end.join.strip 
       same_results = cols.zip(last).map do |a,b| 
         a.respond_to?(:single) && a.single && a.object_id == b.object_id && !a.is_a?(Numeric) 

data/parsely.gemspec

@@ -10,6 +10,7 @@ Gem::Specification.new do |s|
   s.homepage    = "http://github.com/riffraff/parsely"
   s.summary     = %q{a simple tool for text file wrangling}
   s.description = %q{parsely is a simple tool for managing text files.
+                     DO NOT USE IT.
                      Mostly to replace a lot of awk/sed/ruby/perl one-off scripts.
                      This is an internal release, guaranteed to break and ruin your life.}
 

data/test/basic/017-number-lines/command

@@ -0,0 +1 @@
+parsely 'line + ": " + _0' lines.txt

data/test/basic/017-number-lines/lines.txt

@@ -0,0 +1,3 @@
+a
+b
+c

data/test/basic/017-number-lines/output

@@ -0,0 +1,3 @@
+0: a
+1: b
+2: c

data/test/basic/018-increment-lineno/command

@@ -0,0 +1 @@
+parsely 'line + 1 + ": " + _0' lines.txt

data/test/basic/018-increment-lineno/lines.txt

@@ -0,0 +1,3 @@
+a
+b
+c

data/test/basic/018-increment-lineno/output

@@ -0,0 +1,3 @@
+1: a
+2: b
+3: c

data/test/basic/019-increment-lineno-and-values/command

@@ -0,0 +1 @@
+parsely 'line + ": " + (_0 + 1)' lines.txt

data/test/basic/019-increment-lineno-and-values/lines.txt

@@ -0,0 +1,3 @@
+1
+2
+3

data/test/basic/019-increment-lineno-and-values/output

@@ -0,0 +1,3 @@
+0: 2
+1: 3
+2: 4

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: parsely
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-07-14 00:00:00.000000000Z
+date: 2011-12-10 00:00:00.000000000Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
-  requirement: &2168739940 !ruby/object:Gem::Requirement
+  requirement: &2152641300 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,10 +21,11 @@ dependencies:
         version: '0'
   type: :development
   prerelease: false
-  version_requirements: *2168739940
-description: ! "parsely is a simple tool for managing text files.\n                     Mostly
-  to replace a lot of awk/sed/ruby/perl one-off scripts.\n                     This
-  is an internal release, guaranteed to break and ruin your life."
+  version_requirements: *2152641300
+description: ! "parsely is a simple tool for managing text files.\n                     DO
+  NOT USE IT.\n                     Mostly to replace a lot of awk/sed/ruby/perl one-off
+  scripts.\n                     This is an internal release, guaranteed to break
+  and ruin your life."
 email:
 - rff.rff+parsely@gmail.com
 executables:
@@ -33,7 +34,10 @@ extensions: []
 extra_rdoc_files: []
 files:
 - .gitignore
+- .travis.yml
 - Gemfile
+- LICENSE
+- README.mkd
 - Rakefile
 - TODO
 - bin/parsely
@@ -89,6 +93,15 @@ files:
 - test/basic/016-rcfile-home/counts.txt
 - test/basic/016-rcfile-home/myhome/.parselyrc
 - test/basic/016-rcfile-home/output
+- test/basic/017-number-lines/command
+- test/basic/017-number-lines/lines.txt
+- test/basic/017-number-lines/output
+- test/basic/018-increment-lineno/command
+- test/basic/018-increment-lineno/lines.txt
+- test/basic/018-increment-lineno/output
+- test/basic/019-increment-lineno-and-values/command
+- test/basic/019-increment-lineno-and-values/lines.txt
+- test/basic/019-increment-lineno-and-values/output
 - test/cli-runner.rb
 homepage: http://github.com/riffraff/parsely
 licenses: []
@@ -165,4 +178,14 @@ test_files:
 - test/basic/016-rcfile-home/counts.txt
 - test/basic/016-rcfile-home/myhome/.parselyrc
 - test/basic/016-rcfile-home/output
+- test/basic/017-number-lines/command
+- test/basic/017-number-lines/lines.txt
+- test/basic/017-number-lines/output
+- test/basic/018-increment-lineno/command
+- test/basic/018-increment-lineno/lines.txt
+- test/basic/018-increment-lineno/output
+- test/basic/019-increment-lineno-and-values/command
+- test/basic/019-increment-lineno-and-values/lines.txt
+- test/basic/019-increment-lineno-and-values/output
 - test/cli-runner.rb
+has_rdoc: