bel 0.2.1 → 0.3.0.beta1

data/bin/bel2rdf

@@ -0,0 +1,134 @@
+#!/usr/bin/env ruby
+# bel2rdf: Convert BEL to RDF triples.
+#
+# From BEL file
+# usage: bel2rdf -b file.bel
+#
+# From file
+# usage: bel2rdf --bel [FILE]
+#
+# From standard in
+# usage: echo "<BEL DOCUMENT STRING>" | bel2rdf
+#
+# Format option
+# usage: bel2rdf --bel [FILE] --format [ntriples | nquads | turtle]
+
+require 'bel'
+require 'optparse'
+require 'set'
+require 'open-uri'
+
+# setup and parse options
+options = {
+  format: 'ntriples',
+  schema: false
+}
+OptionParser.new do |opts|
+  opts.banner = '''Converts BEL statements to RDF triples.
+Usage: bel2rdf --bel [FILE]'''
+
+  opts.on('-b', '--bel FILE', 'BEL file to convert.  STDIN (standard in) can also be used for BEL content.') do |bel|
+    options[:bel] = bel
+  end
+
+  opts.on('-f', '--format FORMAT', 'RDF file format.') do |format|
+    options[:format] = format.downcase
+  end
+
+  opts.on('-s', '--[no-]schema', 'Write BEL RDF schema?') do |schema|
+    options[:schema] = schema
+  end
+end.parse!
+
+# option guards
+unless options[:bel] or not STDIN.tty?
+  $stderr.puts "No bel content provided.  Either use --bel option or STDIN (standard in).  Use -h / --help for details." 
+  exit 1
+end
+if options[:bel] and not File.exists? options[:bel]
+  $stderr.puts "No file for bel, #{options[:bel]}"
+  exit 1
+end
+unless ['ntriples', 'nquads', 'turtle'].include? options[:format]
+  $stderr.puts "Format was not one of: ntriples, nquads, or turtle"
+  exit 1
+end
+
+
+class Main
+  include BEL::Language
+
+  def initialize(content, writer)
+    BEL::Script.parse(content).find_all { |obj|
+      obj.is_a? Statement
+    }.each do |stmt|
+      triples = stmt.to_rdf[1]
+      triples.each do |triple|
+        writer << triple
+      end
+    end
+  end
+end
+
+class Serializer
+  attr_reader :writer
+
+  def initialize(stream, format)
+    rdf_writer = find_writer(format)
+    @writer = rdf_writer.new($stdout, {
+        :stream => stream
+      }
+    )
+  end
+
+  def <<(trpl)
+    @writer.write_statement(RDF::Statement(*trpl))
+  end
+
+  def done
+    @writer.write_epilogue
+  end
+
+  private
+
+  def find_writer(format)
+    case format
+      when 'nquads'
+        BEL::RDF::RDF::NQuads::Writer
+      when 'turtle'
+        begin
+          require 'rdf/turtle'
+          BEL::RDF::RDF::Turtle::Writer
+        rescue LoadError
+          $stderr.puts """Turtle format not supported.
+Install the 'rdf-turtle' gem."""
+          raise
+        end
+      when 'ntriples'
+        BEL::RDF::RDF::NTriples::Writer
+    end
+  end
+end
+
+# create writer for format
+rdf_writer = ::Serializer.new(true, options[:format])
+
+# first write schema if desired
+if options[:schema]
+  BEL::RDF::vocabulary_rdf.each do |trpl|
+    rdf_writer << trpl
+  end
+end
+
+# read bel content
+content =
+  if options[:bel]
+    File.open(options[:bel]).read
+  else
+    $stdin.read
+  end
+
+# parse and write rdf
+Main.new(content, rdf_writer)
+# vim: ts=2 sw=2:
+# encoding: utf-8

data/bin/bel_compare

@@ -0,0 +1,177 @@
+#!/usr/bin/env ruby
+# vim: ts=2 sw=2
+
+require 'optparse'
+require 'bel'
+include BEL::Language
+include BEL::Namespace
+
+USAGE = "Usage: #{File.basename(__FILE__)} [options] [file] [file]"
+
+options = {}
+OptionParser.new do |opts|
+  opts.banner = USAGE
+
+  opts.on("-v", "--[no-]verbose", "Run verbosely") do |v|
+    options[:verbose] = v
+  end
+  opts.on_tail("-h", "--help", "Show usage and options") do
+    puts opts
+    exit
+  end
+end.parse!
+
+if ARGV.length != 2
+  $stderr.puts "Need to specify two BEL files."
+  $stderr.puts USAGE
+  exit 1
+end
+
+def fetch_groups(bel)
+  groups = {}
+  active_group = nil
+  BEL::Script.parse(bel) do |obj|
+    if obj.is_a? BEL::Language::StatementGroup
+      active_group = obj
+    end
+    if obj.is_a? BEL::Language::UnsetStatementGroup
+      groups[obj.name] = active_group
+    end
+  end
+  groups
+end
+
+threads = []
+first_groups = {}
+second_groups = {}
+threads << Thread.new {
+  first_groups.update(fetch_groups(File.read(ARGV[0])))
+}
+
+threads << Thread.new {
+  second_groups.update(fetch_groups(File.read(ARGV[1])))
+}
+
+threads.each { |t| t.join }
+
+group_diff = second_groups.length - first_groups.length
+if group_diff == 0
+  puts "Equal number of statement groups."
+elsif group_diff > 0
+  puts "Second file has #{group_diff} more statement groups."
+else
+  puts "Second file has #{group_diff.abs} less statement groups."
+end
+
+first_statements = first_groups.values.
+  map {|group| group.statements.map {|s| s.to_bel} }.flatten
+puts "First file has #{first_statements.length} statements."
+second_statements = second_groups.values.
+  map {|group| group.statements.map {|s| s.to_bel}}.flatten
+puts "Second file has #{second_statements.length} statements."
+
+# statement groups in 1st, but not 2nd
+first_groups.each do |name,first|
+  second = second_groups[name]
+  if not second
+    puts "#{name}, in 1st but not 2nd."
+  end
+end
+
+# statement groups in 2nd, but not 1st
+second_groups.each do |name,second|
+  first = first_groups[name]
+  if not first
+    puts "#{name}, in 2nd but not 1st."
+  end
+end
+
+# differences between statement group intersection
+first_groups.each do |name,first|
+  second = second_groups[name]
+  if second
+    first_stmt = first.statements ? first.statements.map {|s| s.to_bel } : []
+    second_stmt = second.statements ? second.statements.map {|s| s.to_bel } : []
+    if first_stmt == second_stmt
+      if options[:verbose]
+        puts "#{name}, both contain zero statements" if first_stmt.empty?
+        puts "#{name}, both contain #{first_stmt.length} statements"
+      end
+    else
+      puts "#{name}:"
+      if options[:verbose]
+        puts "  1st contains #{first_stmt.length} statements"
+        puts "  2nd contains #{second_stmt.length} statements"
+        puts "  common to both, #{(first_stmt & second_stmt).length}"
+        puts (first_stmt & second_stmt).map {|stmt| stmt.to_s}.join("\n    ").prepend("    ")
+      end
+
+      ab = first_stmt - second_stmt
+      puts "  statements only in 1st, #{ab.length} (1st - 2nd)"
+      if ab.length > 0
+        puts ab.map {|stmt| stmt.to_s}.join("\n    ").prepend("    ")
+      end
+
+      ba = second_stmt - first_stmt
+      puts "  statements only in 2nd, #{ba.length} (2nd - 1st)"
+      if ba.length > 0
+        puts ba.map {|stmt| stmt.to_s}.join("\n    ").prepend("    ")
+      end
+    end
+  end
+end
+
+first_terms = first_groups.values.
+  map { |group| group.statements }.flatten.
+  map { |stmt|
+    l = [stmt.subject.to_bel]
+    if stmt.simple?
+      l << stmt.object.to_bel
+    elsif stmt.nested?
+      l << stmt.object.subject.to_bel
+      l << stmt.object.object.to_bel
+    end
+    l
+  }.flatten.uniq.
+  sort {|term1,term2| term1.to_s <=> term2.to_s}
+second_terms = second_groups.values.
+  map { |group| group.statements }.flatten.
+  map { |stmt|
+    l = [stmt.subject.to_bel]
+    if stmt.simple?
+      l << stmt.object.to_bel
+    elsif stmt.nested?
+      l << stmt.object.subject.to_bel
+      l << stmt.object.object.to_bel
+    end
+    l
+  }.flatten.uniq.
+  sort {|term1,term2| term1.to_s <=> term2.to_s}
+
+# statements
+puts "Statements:"
+ab = first_statements - second_statements
+puts "  only in 1st, #{ab.length} (1st - 2nd)"
+if ab.length > 0
+  puts ab.map {|t| t.to_s}.join("\n    ").prepend("    ")
+end
+
+ba = second_statements - first_statements
+puts "  only in 2nd, #{ba.length} (2nd - 1st)"
+if ba.length > 0
+  puts ba.map {|t| t.to_s}.join("\n    ").prepend("    ")
+end
+
+# terms
+puts "Terms:"
+ab = first_terms - second_terms
+puts "  only in 1st, #{ab.length} (1st - 2nd)"
+if ab.length > 0
+  puts ab.map {|t| t.to_s}.join("\n    ").prepend("    ")
+end
+
+ba = second_terms - first_terms
+puts "  only in 2nd, #{ba.length} (2nd - 1st)"
+if ba.length > 0
+  puts ba.map {|t| t.to_s}.join("\n    ").prepend("    ")
+end

data/bin/bel_parse

@@ -0,0 +1,60 @@
+#!/usr/bin/env ruby
+# bel_parse: Show parsed objects from BEL content for debugging purposes.
+#
+# From BEL file
+# usage: bel_parse -b file.bel
+#
+# From standard in
+# usage: echo "<BEL DOCUMENT STRING>" | bel_parse
+
+require 'bel'
+require 'optparse'
+
+# additive String helpers
+class String
+  def rjust_relative(distance, string)
+    rjust(distance - string.size + size)
+  end
+end
+
+# setup and parse options
+options = {}
+OptionParser.new do |opts|
+  opts.banner = "Usage: bel_parse [options] [.bel file]"
+  opts.on('-b', '--bel FILE', 'BEL file to parse.  STDIN (standard in) can also be used for BEL content.') do |bel|
+    options[:bel] = bel
+  end
+end.parse!
+
+# option guards
+unless options[:bel] or not STDIN.tty?
+  $stderr.puts "No bel content provided.  Either use --bel option or STDIN (standard in).  Use -h / --help for details." 
+  exit 1
+end
+
+if options[:bel] and not File.exists? options[:bel]
+  $stderr.puts "No file for bel, #{options[:bel]}"
+  exit 1
+end
+
+# read bel content
+content =
+if options[:bel]
+  File.open(options[:bel]).read
+else
+  $stdin.read
+end
+
+class Main
+
+  def initialize(content)
+    BEL::Script.parse(content) do |obj|
+      object_desc = obj.class.name.split('::').last
+      object_desc << obj.to_bel.rjust_relative(25, object_desc)
+      puts object_desc
+    end
+  end
+end
+Main.new(content)
+# vim: ts=2 sw=2:
+# encoding: utf-8

data/bin/bel_rdfschema

@@ -0,0 +1,72 @@
+#!/usr/bin/env ruby
+# bel_rdfschema: Dump RDF schema for BEL.
+# usage: bel_rdfschema --format [ntriples | nquads | turtle]
+
+require 'bel'
+require 'optparse'
+require 'set'
+require 'open-uri'
+
+options = {
+  format: 'ntriples'
+}
+OptionParser.new do |opts|
+  opts.banner = '''Dumps RDF schema for BEL.
+Usage: bel_rdfschema'''
+
+  opts.on('-f', '--format FORMAT', 'RDF file format.') do |format|
+    options[:format] = format.downcase
+  end
+end.parse!
+
+unless ['ntriples', 'nquads', 'turtle'].include? options[:format]
+  $stderr.puts "Format was not one of: ntriples, nquads, or turtle"
+  exit 1
+end
+
+class Serializer
+  attr_reader :writer
+
+  def initialize(stream, format)
+    rdf_writer = find_writer(format)
+    @writer = rdf_writer.new($stdout, {
+        :stream => stream
+      }
+    )
+  end
+
+  def <<(trpl)
+    @writer.write_statement(RDF::Statement(*trpl))
+  end
+
+  def done
+    @writer.write_epilogue
+  end
+
+  private
+
+  def find_writer(format)
+    case format
+      when 'nquads'
+        BEL::RDF::RDF::NQuads::Writer
+      when 'turtle'
+        begin
+          require 'rdf/turtle'
+          BEL::RDF::RDF::Turtle::Writer
+        rescue LoadError
+          $stderr.puts """Turtle format not supported.
+Install the 'rdf-turtle' gem."""
+          raise
+        end
+      when 'ntriples'
+        BEL::RDF::RDF::NTriples::Writer
+    end
+  end
+end
+
+@rdf_writer = ::Serializer.new(true, options[:format])
+BEL::RDF::vocabulary_rdf.each do |trpl|
+  @rdf_writer << trpl
+end
+# vim: ts=2 sw=2:
+# encoding: utf-8

data/bin/bel_summarize

@@ -0,0 +1,86 @@
+#!/usr/bin/env ruby
+# vim: ts=2 sw=2
+
+require 'bel'
+require 'csv'
+require 'optparse'
+include BEL::Language
+include BEL::Namespace
+
+options = {}
+OptionParser.new do |opts|
+  opts.banner = '''Statistic report for BEL script.
+Usage: bel_summarize --bel [FILE]'''
+
+  opts.on('-b', '--bel FILE', 'BEL file to summarize.  STDIN (standard in) can also be used for BEL content.') do |bel|
+    options[:bel] = bel
+  end
+end.parse!
+
+# read bel content
+content =
+  if options[:bel]
+    File.open(options[:bel]).read
+  else
+    $stdin.read
+  end
+
+CSV do |csv_out|
+  report = {
+    statement_group_count: 0,
+    empty_statement_groups: 0,
+    statement_count: 0,
+    evidence_count: 0
+  }
+
+  FUNCTIONS.each do |k, v|
+    report['fx_' + v[:long_form].to_s] = 0
+  end
+
+  RELATIONSHIPS.each do |r|
+    report['rel_' + r.to_s] = 0
+  end
+
+  active_group = nil
+  BEL::Script.parse(content) do |obj|
+    if obj.is_a? BEL::Language::StatementGroup
+      report[:statement_group_count] += 1
+      active_group = obj
+    end
+    if obj.is_a? BEL::Language::UnsetStatementGroup
+      if active_group.statements.empty?
+        report[:empty_statement_groups] += 1
+      end
+    end
+    if obj.is_a? BEL::Language::Term
+      report['fx_' + obj.fx[:long_form].to_s] += 1
+    end
+    if obj.is_a? BEL::Language::Statement
+      report[:statement_count] += 1
+      obj.relationship = case obj.relationship
+      when :"->"
+        :increases
+      when :"-|"
+        :decreases
+      when :"=>"
+        :directlyIncreases
+      when :"=|"
+        :directlyDecreases
+      when :"--"
+        :association
+      else
+        obj.relationship
+      end
+
+      if obj.relationship
+        report['rel_' + obj.relationship.to_s] += 1
+      end
+    end
+    if obj.is_a? BEL::Language::Annotation
+      report[:evidence_count] += 1 if obj.name == 'Evidence'
+    end
+  end
+
+  csv_out << report.keys
+  csv_out << report.values
+end