rbbt 1.1.7 → 2.0.0

data/lib/rbbt/util/rake.rb

@@ -1,183 +0,0 @@
-require 'rake'
-
-# Include the step_def and step methods to simplify Pipelines. Steps depend on
-# the step strictly above by default. The output of the step is save marshaled,
-# except for Strings which are save as text. The input of the step, the output
-# of the previous step if availabe is accessed with the input method
-#
-# Example::
-#
-#     step :text do 
-#       "Text to revert"
-#     end
-#
-#     step :revert do
-#       text = input
-#       text.reverse
-#     end
-#
-module Rake::Pipeline
-
-  module Rake::Pipeline::Step
-
-    class << self
-
-      @@step_descriptions = {}
-      def step_descriptions
-        @@step_descriptions
-      end
-
-      def add_description(re, step, message)
-        @@step_descriptions[re] = "#{ step }: #{ message }"                       
-      end
-
-      @@last_step = nil
-      def step_def(name, dependencies = nil)
-
-        re = Regexp.new(/(?:^|\/)#{name}\/.*$/)
-
-        # Take the last_description and associate it with the name
-        if Rake.application.last_description
-          add_description(re, name, Rake.application.last_description)
-        end
-
-        if dependencies.nil? && ! @@last_step.nil?
-          dependencies = @@last_step
-        end
-        @@last_step = name
-
-        # Generate the Hash definition
-        case 
-        when dependencies.nil?
-          re
-        when String === dependencies || Symbol === dependencies
-          {re => lambda{|filename| filename.sub(name.to_s,dependencies.to_s) }}
-        when Array === dependencies
-          {re => lambda{|filename| dependencies.collect{|dep| filename.sub(name.to_s, dep.to_s) } }}
-        when Proc === dependencies
-          {re => dependencies}
-        end
-
-      end
-
-    end
-  end
-
-  module Rake::Pipeline::Info
-
-    def self.info_file(filename)
-      filename.sub(/^(.*?)(?:[^\/]*)\/([^\/]*)$/, '\1.info/\2.yaml')
-    end
-
-    def self.load_info(t)
-      filename = t.name
-      info_filename = info_file(filename)
-
-      if File.exists? info_filename
-        YAML.load(File.open(info_filename))
-      else
-        {}
-      end
-    end
-
-    def self.save_info(t, info = {})
-      filename = t.name
-      info_filename = info_file(filename)
-
-      FileUtils.mkdir_p File.dirname(info_filename) unless File.exists? File.dirname(info_filename)
-      File.open(info_filename,'w'){|file|
-        file.write YAML.dump info
-      }
-    end
-
-  end
-
-
-  NON_ASCII_PRINTABLE = /[^\x20-\x7e\s]/
-  def is_binary?(file)
-    binary = file.read(1024) =~ NON_ASCII_PRINTABLE
-    file.rewind
-    binary
-  end
-
-  def step_descriptions
-    Rake::Pipeline::Step.step_descriptions
-  end
-
-
-  def step_def(*args)
-    Rake::Pipeline::Step.step_def(*args)
-  end
-
-  def infile(t, &block)
-    File.open(t.prerequisites.first) do |f|
-      block.call(f)
-    end
-  end
-
-  def outfile(t, &block)
-    File.open(t.name, 'w') do |f|
-      block.call(f)
-    end
-  end
-
-  def load_input(t)
-    return nil if t.prerequisites.first.nil?
-    infile(t){|f| 
-      if is_binary?(f)
-        Marshal.load(f) 
-      else
-        f.read
-      end
-    }
-  end
-
-  def save_output(t, output)
-    case 
-    when output.nil?
-      nil
-    when String === output
-      outfile(t){|f| f.write output } 
-    else
-      outfile(t){|f| f.write Marshal.dump(output) } 
-    end
-
-  end
-
-  # We cannot load the input variable before the block.call, so we need another method
-
-  # Load the input data from the previous step
-  def input
-    load_input(@@current_task) if @@current_task
-  end
-
-  # Add values to the info file
-  def info(values = {})
-    info = Rake::Pipeline::Info.load_info(@@current_task)
-    info = info.merge values
-    Rake::Pipeline::Info.save_info(@@current_task, info)
-    info
-  end
-  
-  
-  # Define a new step, it depends on the previously defined by default. It
-  # saves the output of the block so it can be loaded by the input method of
-  # the next step
-  def step(name, dependencies = nil, &block)
-    rule step_def(name, dependencies) do |t| 
-
-      # Save the task object to be able to load the input
-      @@current_task = t
-      
-      output = block.call(t)
-      
-      save_output(t, output)
-    end
-
-  end
-end
-
-if __FILE__ == $0
-
-  p Rake::Pipeline::Info.info_file('work/diseases/t')
-end

data/lib/rbbt/util/simpleDSL.rb

@@ -1,87 +0,0 @@
-require 'parse_tree_extensions'
-require 'parse_tree'
-require 'ruby2ruby'
-
-# This class helps designing DSL in ruby based on method_missing. Class
-# is initialize with a block of code or a file with the code, and it is
-# given a method to be invoked instead of method missing. This class
-# deals simply with making the method_missing alias and removing it and
-# executing the block of file with code.
-class SimpleDSL
-
-  class ConfigFileMissingError < StandardError; end
-
-  private
-
-  def hook_method(method = nil)
-    method ||= :DSL_action 
-    @@restore_name = ("restore_DSL_" + method.to_s).to_sym
-    @@method_name = method.to_sym
-    
-    class << self
-      @restore_stack ||= []
-      @restore_stack << @@restore_name
-      alias_method(@@restore_name, :method_missing)
-      alias_method(:method_missing, @@method_name)
-    end
-  end
-
-  def unhook_method
-    class << self
-      alias_method(:method_missing, @restore_stack.pop)
-    end
-  end
-
-  public
-
-  def parse(method = nil, actions = nil, &block)
-
-    actions ||= block
-    
-    hook_method(method)
-
-    # Execute
-    if actions.is_a? Proc
-     
-      @config[@@method_name] = actions.to_ruby.collect[1..-2].join
-
-      instance_eval &actions
-    elsif File.exists?(actions)
-
-      @config[@@method_name] = File.open(actions).read
-
-      eval File.open(actions).read
-    end
-
-    unhook_method
-
-  end
-
-
-  # Processes a DSL. +method+ is the name of the method executed instead
-  # of method_missing. The code to be evaluated as a DSL is either
-  # specified in +&block+ or in the file pointed by +file+.
-  def initialize(method = nil, file = nil, &block)
-    @config = {}
-    if file
-      raise ConfigFileMissingError.new "File '#{ file }' is missing. Have you installed the config files? (use rbbt_config)." unless File.exists? file
-      parse(method, file)
-    end
-
-    if block
-      parse(method, block)
-    end
-  end
-
-  # Returns the code with the DSL that was executed. If it came from a
-  # block it was turned to string using ruby2ruby.
-  def config(action = nil)
-    if action
-      @config[action.to_sym]
-    else
-      @config[:DSL_action]
-    end
-  end
-end
-
-

data/lib/rbbt/util/tmpfile.rb

@@ -1,19 +0,0 @@
-require 'fileutils'
-require 'rbbt' 
-
-
-module TmpFile
-
-  # Creates a random file name, with the given suffix and a random number
-  # up to +max+
-  def self.random_name( s="",max=10000000)
-    n = rand(max)
-    s << n.to_s
-    s
-  end
-
-  # Creates a random filename in the temporary directory
-  def self.tmp_file(s = "",max=10000000)
-    File.join(Rbbt.tmpdir,random_name(s,max))
-  end
-end

data/tasks/install.rake

@@ -1,124 +0,0 @@
-require 'rbbt'
-
-$datadir = Rbbt.datadir
-$scriptdir = File.join(Rbbt.rootdir, '/install_scripts')
-
-
-task 'abner' do
-  directory = "#{$datadir}/third_party/abner/"
-  if !File.exists?(File.join(directory, 'abner.jar')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_abner.sh;cd -`
-  end
-end
-
-task 'banner' do
-  directory = "#{$datadir}/third_party/banner/"
-  if !File.exists?(File.join(directory, 'banner.jar')) || $force 
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_banner.sh;cd -`
-  end
-end
-
-task 'crf++' do
-  directory = "#{$datadir}/third_party/crf++/"
-  if !File.exists?(File.join(directory, 'ruby/CRFPP.so')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_crf++.sh;cd -`
-  end
-end
-
-
-
-task 'wordlists' do
-  FileUtils.cp_r File.join($scriptdir, 'wordlists/'), $datadir
-end
-
-task 'polysearch' do
-  directory = "#{$datadir}/dbs/polysearch/"
-  if !File.exists?(File.join(directory,'disease.txt')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_polysearch.sh;cd -`
-  end
-end
-
-
-task '3party' => %w(abner banner crf++)
-
-task 'entrez' do
-  directory = "#{$datadir}/dbs/entrez/"
-  if !File.exists?(File.join(directory,'gene_info')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_entrez.sh;cd -`
-  end
-end
-
-task 'go' do
-  directory = "#{$datadir}/dbs/go/"
-  if !File.exists?(File.join(directory,'gene_ontology.obo')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory}/; rm * -Rf; #{$scriptdir}/get_go.sh;cd -`
-  end
-end
-
-task 'biocreative' do
-  directory = "#{$datadir}/biocreative/"
-  if !File.exists?(File.join(directory, 'BC2GN')) || $force
-    FileUtils.mkdir_p directory
-    `cd #{directory};rm -Rf *; #{$scriptdir}/get_biocreative.sh;cd -`
-  end
-end
-
-
-task 'datasets' => %w(entrez biocreative)
-
-task 'organisms' do
-  directory = "#{$datadir}/organisms"
-  FileUtils.mkdir_p directory
-  %w(Rakefile rake-include.rb).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "organisms/#{ f }"), directory
-  }
-  Dir.glob(File.join($scriptdir, "organisms/*.Rakefile")).each{|f|
-    org = File.basename(f).sub(/.Rakefile/,'')
-    if !File.exists?(File.join(directory, org))
-      FileUtils.mkdir_p File.join(directory, org)
-    end
-    FileUtils.cp f , File.join(directory, "#{ org }/Rakefile")
-  }
-  `cd #{directory}; rake names`
-end
-
-task 'ner' do
-  directory = "#{$datadir}/ner"
-  FileUtils.mkdir_p directory
-  %w(Rakefile config).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "ner/#{ f }"), directory 
-  }
-  
-  %w(data model results).each{|d|
-    FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-
-task 'norm' do
-  directory = "#{$datadir}/norm"
-  FileUtils.mkdir_p directory
-  %w(Rakefile config functions.sh).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "norm/#{ f }"), directory 
-  }
- %w(results models).each{|d|
-  FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-
-task 'classifier' do
-  directory = "#{$datadir}/classifier"
-  FileUtils.mkdir_p directory
-  %w(Rakefile R).each{|f|
-    FileUtils.cp_r File.join($scriptdir, "classifier/#{ f }"), directory
-  }
-  %w(data model results).each{|d|
-    FileUtils.mkdir_p File.join(directory, d)
-  }
-end
-