bio-sambamba 0.0.2 → 0.0.3

data/Gemfile

@@ -1,7 +1,7 @@
 source "http://rubygems.org"
 
 gem "bio", "~> 1.4.2"
-gem "oj", "~> 1.2.9"
+gem "oj", "~> 1.3.4"
 
 group :development do
   gem "bundler", "~> 1.1.4"

data/README.md

@@ -1,5 +1,6 @@
 # bio-sambamba
 
+[![Code Climate](https://codeclimate.com/badge.png)](https://codeclimate.com/github/lomereiter/bioruby-sambamba)
 [![Build Status](https://secure.travis-ci.org/lomereiter/bioruby-sambamba.png)](http://travis-ci.org/lomereiter/bioruby-sambamba)
 
 Full description goes here
@@ -13,7 +14,7 @@ Note: this software is under active development!
     rake install
 ```
 
-In order to use the gem, you also need <code>sambamba</code> tool installed.
+In order to use the gem, you also need <code>sambamba</code> tool installed (version >= 0.2.4).
 
 If you use Debian, you can download a package for your architecture from
 [github downloads](http://github.com/lomereiter/sambamba/downloads).

data/VERSION

@@ -1 +1 @@
-0.0.2
+0.0.3

data/features/filtering.feature

@@ -0,0 +1,74 @@
+Feature: custom filters for BAM data
+
+    In order to filter alignment records faster,
+    As a user who deals with a lot of data,
+    I want to be able to create custom filters and have them executed
+    not in Ruby but in sambamba view tool.
+
+    Background: 
+        Given I have a BAM file
+          And I have an iterator for alignment records in this file
+
+    Scenario: setting filter for alignments
+         When I create a filter with Bio::Bam::filter function
+         Then I should be able to pass this filter to the 'with_filter' method of the iterator
+          And it should give me a enumerator for those alignments which pass the filter
+
+    Scenario Outline: setting conditions for flags
+        When I create a filter with Bio::Bam::filter { flag.<name>.<is_set_or_unset> }
+        Then I should get all alignments where flag called <name> <is_set_or_unset> correspondingly.
+
+        Examples:
+            |       name       |    is_set_or_unset  |
+            |   proper_pair    |      is_set         |
+            | mate_is_unmapped |     is_unset        |
+
+    Scenario Outline: setting conditions for integer and string fields
+        When I create a filter with Bio::Bam::filter { <field> <comparison_operator> <value> }
+        Then I should get all alignments where <comparison_operator> for <field> and <value> is true
+
+        Examples:
+            |     field       | comparison_operator |   value   |
+            |  mate_position  |        <            |   1.Kbp   |
+            | mapping_quality |        >=           |    39     |
+            |   read_name     |        >            | 'EAS114_' |
+
+    Scenario Outline: setting conditions for tags
+        When I create a filter with Bio::Bam::filter { tag(:<tagname>) <comparison_operator> <value> }
+        Then I should get all alignments where tag with name <tagname> exists
+        And <comparison_operator> for tag with name <tagname> and <value> is true
+
+        Examples:
+            | tagname | comparison_operator | value |
+            |   Aq    |         ==          |   72  |
+            |   UQ    |         !=          |   0   | 
+            |   NM    |         >=          |   3   |
+
+    Scenario Outline: regex matching for tags and string fields
+        When I create a filter with Bio::Bam::filter { <field_or_tag> =~ <regex> }
+        Then I should get all alignments where <field_or_tag> matches given <regex>
+
+        Examples:
+            | field_or_tag |       regex         |
+            |  read_name   | /^B7_(?:\d+):+\d+$/ |
+            |    cigar     | /[^M\d]/            |
+
+    Scenario Outline: logical operations on filters
+        Given I have several <conditions>
+         When I enclose them by a <n-ary operation> block
+         Then I should get a condition representing <n-ary operation> of those
+
+        Examples:
+            |            conditions                                |  n-ary operation   |
+            | ["position >= 100.bp", "mate_position <= 200.bp"]    |      union         |
+            | ["tag(:NM) == 3", "tag(:UQ) == 42"]                  |    intersection    |         
+
+    Scenario Outline: negation of filter
+        Given I have a condition <condition>
+         When I enclose it in 'negate' block
+         Then I should have a condition representing the same alignments as <'not' equivalent>
+
+         Examples:
+            |            condition          |      'not' equivalent          |
+            |       flag.paired.is_set      |     flag.paired.is_unset       | 
+            |     mapping_quality >= 50     |     mapping_quality < 50       |

data/features/step_definitions/filtering_steps.rb

@@ -0,0 +1,86 @@
+Given /^I have an iterator for alignment records in this file$/ do
+    @iter = @bam.alignments
+end
+
+When /^I create a filter with Bio::Bam::filter function$/ do
+    @filter = Bio::Bam::filter { mapping_quality > 50 }
+end
+
+Then /^I should be able to pass this filter to the 'with_filter' method of the iterator$/ do
+    @iter.should respond_to(:with_filter).with(1).argument
+    @alignments = @iter.with_filter @filter
+end
+
+Then /^it should give me a enumerator for those alignments which pass the filter$/ do
+    @alignments.each do |read|
+      read.mapping_quality.should > 50
+    end
+end
+
+When /^I create a filter with (.*?{[^}]*?})\s*$/ do |query|
+    @filter = eval(query)
+end
+
+Then /^I should get all alignments where (.*?) for (\w+?) and (.*?) is true$/ do |op, field, val|
+    @bam.alignments.with_filter(@filter).take(100).each do |read|
+      eval("read.#{field} #{op} #{val}").should be true
+    end
+end
+
+Then /^I should get all alignments where tag with name (\w{2}) exists$/ do |tagname|
+    @bam.alignments.with_filter(@filter).take(100).each do |read|
+      read.tags[tagname].should_not be nil
+    end
+end
+
+Then /^(.*?) for tag with name (\w{2}) and (.*?) is true$/ do |op, tagname, val|
+    @bam.alignments.with_filter(@filter).take(100).each do |read|
+      eval("read.tags['#{tagname}'] #{op} #{val}").should be true
+    end
+end
+
+Then /^I should get all alignments where (\w+) matches given (.*?)$/ do |field, regex|
+    @bam.alignments.with_filter(@filter).take(100).each do |read|
+      eval("read.#{field}.should =~ #{regex}")
+    end
+end
+
+Then /^I should get all alignments where flag called (\w+) (\w+) correspondingly.$/ do |flagname, op|
+    @bam.alignments.with_filter(@filter).take(100).each do |read|
+      read.send(flagname.to_sym).should == (op == 'is_set') 
+    end
+end
+
+Given /^I have several (.*?)$/ do |conditions|
+    @conditions = eval(conditions)
+    @filters = @conditions.map {|condition| Bio::Bam::filter { eval(condition) }} 
+end
+
+When /^I enclose them by a (\w+) block$/ do |op|
+    all = @conditions.join "\n"
+    @all_filter = Bio::Bam::filter { eval("#{op} { #{all} }") }
+end
+
+Then /^I should get a condition representing (\w+) of those$/ do |op|
+    seq_c = @filters.map {|f| @bam.alignments.with_filter(f).map(&:sequence).to_a}
+    seq_f = @bam.alignments.with_filter(@all_filter).map(&:sequence).to_a
+    if op == 'union' then
+        seq_c.reduce(&:|).sort.should == seq_f.uniq.sort
+    elsif op == 'intersection'
+        seq_c.reduce(&:&).sort.should == seq_f.uniq.sort
+    else 
+        raise 'unknown op: ' + op
+    end
+end
+
+Given /^I have a condition (.*?)$/ do |cond|
+    @negation = Bio::Bam::filter { negate { eval(cond) }}
+end
+
+When /^I enclose it in 'negate' block$/ do
+end
+
+Then /^I should have a condition representing the same alignments as (.*?)$/ do |equiv|
+    @equiv = Bio::Bam::filter { eval(equiv) }
+    @bam.alignments.with_filter(@negation).map(&:read_name).should == @bam.alignments.with_filter(@equiv).map(&:read_name)
+end

data/features/step_definitions/iterate-alignments_steps.rb

@@ -23,7 +23,7 @@ Then /^I should be able to access all fields mentioned in SAM\/BAM format specif
   @read.position.should == 100
   @read.flag.should == 69
   @read.mapping_quality.should == 0
-  @read.cigar_string.should == '*'
+  @read.cigar.should == '*'
   @read.reference.should == 'chr1'
   @read.quality.should == [27, 27, 27, 22, 27, 27, 27, 26, 27, 27, 27, 27, 27, 27, 27, 27, 23, 26, 26, 27, 22, 26, 19, 27, 26, 27, 26, 26, 26, 26, 26, 24, 19, 27, 26]
 end

data/features/step_definitions/syntax-sugar_steps.rb

@@ -0,0 +1,42 @@
+Given /^the associated BAI file$/ do
+    @bam.has_index?.should be_true
+end
+
+When /^I say "(.*?)"$/ do |expr|
+    bam = @bam
+    @alignments = eval(expr)
+end
+
+Then /^I should get an enumerator for alignments$/ do
+    @alignments.first.should be_instance_of Bio::Bam::Alignment
+end
+
+Then /^each of them should have reference sequence "(.*?)"$/ do |sequence|
+    @alignments.each do |read|
+        read.reference.should == sequence
+    end
+end
+
+Then /^each of them should overlap region \[(\d+), (\d+)\] \(1-based\)$/ do |begpos, endpos|
+    @alignments.each do |read|
+        read.position.should <= endpos.to_i
+        (read.position + read.bases_covered).should >= begpos.to_i
+    end
+end
+
+Given /^an alignment iterator$/ do
+    @alignments = @bam.alignments
+end
+
+When /^I use 'select' method of this iterator with a block$/ do
+    @selected_reads = @alignments.select { read_name =~ /^EAS2/ }
+end
+
+Then /^it should be the same as calling Bio::Bam::filter with this block$/ do
+    @filter = Bio::Bam::filter { read_name =~ /^EAS2/ } 
+end
+
+Then /^passing it as an argument to the 'with_filter' method$/ do
+    # TODO: make comparison function for alignments
+    @selected_reads.to_a.length.should == @alignments.with_filter(@filter).to_a.length
+end

data/features/syntax-sugar.feature

@@ -4,14 +4,25 @@ Feature: syntax sugar
     As a Rubyista,
     I want some syntax sugar.
 
-    Scenario: fetching alignments
+    Scenario: human-readable requests
         Given I have a BAM file
-          And associated BAI file
-         When I say "bam.alignments.referencing(chromosome).overlapping(500.kbp .. 600.kbp)"
-         Then I should get these alignments
+          And the associated BAI file
+         When I say "bam.alignments.referencing('chr1').overlapping(1.Kbp .. 2.Kbp)"
+         Then I should get an enumerator for alignments
+          And each of them should have reference sequence "chr1"
+          And each of them should overlap region [1000, 2000] (1-based)
 
-    Scenario: using shortcuts
+    Scenario: shortcuts for programmers who are too lazy to type that much
         Given I have a BAM file
-          And associated BAI file
-         When I say "bam[chromosome][500.kbp .. 600.kbp]"
-         Then I should get these alignments
+          And the associated BAI file
+         When I say "bam['chr1'][500.bp .. 1.Kbp]"
+         Then I should get an enumerator for alignments
+          And each of them should have reference sequence "chr1"
+          And each of them should overlap region [500, 1000] (1-based)
+
+    Scenario: shortcut for filtering
+        Given I have a BAM file
+          And an alignment iterator
+         When I use 'select' method of this iterator with a block
+         Then it should be the same as calling Bio::Bam::filter with this block
+          And passing it as an argument to the 'with_filter' method

data/lib/bio-sambamba.rb

@@ -1,8 +1,11 @@
 require 'bio/command'
 require 'oj'
 
+require 'bio-sambamba/exception.rb'
 require 'bio-sambamba/samheader.rb'
 require 'bio-sambamba/alignment.rb'
 require 'bio-sambamba/alignmentiterator.rb'
 require 'bio-sambamba/bamfile.rb'
 require 'bio-sambamba/samfile.rb'
+require 'bio-sambamba/basepair.rb'
+require 'bio-sambamba/filtering.rb'

data/lib/bio-sambamba/alignment.rb

@@ -15,6 +15,10 @@ module Bio
         @json['tags'][tag]
       end
 
+      def ==(read)
+        read.json == json
+      end
+
       # Hash of record tags
       attr_reader :tags if false
 
@@ -31,7 +35,7 @@ module Bio
       attr_reader :mapping_quality if false
 
       # CIGAR string
-      attr_reader :cigar_string if false
+      attr_reader :cigar if false
 
       # Observed template length
       attr_reader :template_length if false
@@ -52,18 +56,36 @@ module Bio
       # 1-based leftmost position of the mate/next segment
       attr_reader :mate_position if false
 
-      {'tags' => 'tags',
-       'reference' => 'rname',
-       'read_name' => 'qname',
-       'position' => 'pos',
+      # The number of reference bases covered
+      def bases_covered
+        return 0 if cigar == '*'
+        cigar.split(/([MIDNSHP=X])/).each_slice(2).reduce(0) {|res, op| 
+          res += op[0].to_i unless ('M=XDN'.index op[1]).nil?
+          res
+        }
+      end
+
+      {'position' => 'pos',
        'mapping_quality' => 'mapq',
-       'cigar_string' => 'cigar',
        'template_length' => 'tlen',
        'flag' => 'flag',
+       'mate_position' => 'pnext'
+      }.each do |k, v|
+        eval <<-DEFINE_READER
+          def #{k}
+            @json['#{v}'].to_i
+          end
+        DEFINE_READER
+      end
+
+      {'tags' => 'tags',
+       'reference' => 'rname',
+       'read_name' => 'qname',
+       'cigar' => 'cigar',
        'quality' => 'qual',
        'sequence' => 'seq',
-       'mate_reference' => 'rnext',
-       'mate_position' => 'pnext'}.each do |k, v|
+       'mate_reference' => 'rnext'
+       }.each do |k, v|
         eval <<-DEFINE_READER
           def #{k}
             @json['#{v}']
@@ -125,6 +147,10 @@ module Bio
       def is_duplicate             
         (flag & 0x400) != 0
       end
+
+      private
+      attr_accessor :json
+
     end
 
   end

data/lib/bio-sambamba/alignmentiterator.rb

@@ -4,6 +4,7 @@ module Bio
     # Class for iterating through alignments
     class AlignmentIterator
       include Enumerable
+      include SambambaStderrParser
 
       # Creates a new AlignmentIterator object which will
       # parse JSON outputted by a specified command.
@@ -26,19 +27,89 @@ module Bio
         end
       end
 
+      private
+
+      def get_command
+        command = @command
+
+        if not @chromosome.nil? then
+          if not @region.nil? then
+            command.push "#{@chromosome}:#{@region.min}-#{@region.max}"       
+          else
+            command.push "#{@chromosome}"
+          end
+        elsif not @region.nil? then
+          raise 'must specify a reference when doing a region query'
+        end
+
+        command
+      end
+
+      public
+
       # Iterate through all alignments skipping
       # validation checks
       def each
-
         return enum_for(:each) if not block_given?
 
-        Bio::Command.call_command(@command) do |io|
-          io.each do |line|
-            raise line unless line[0] == '{'
-            yield Bio::Bam::Alignment.new(Oj.load(line))
+        command = get_command
+
+        Bio::Command.call_command_open3(command) do |pin, pout, perr|
+          pout.each do |line|
+            json = Oj.load(line)
+            yield Bio::Bam::Alignment.new(json)
           end
+          raise_exception_if_stderr_is_not_empty(perr)
         end
       end
+
+      # Set filter for alignments
+      def with_filter(filter)
+        iter = self.clone
+        iter.command.push('-F')
+        iter.command.push(filter.to_s)
+        iter
+      end
+
+      def select(&block)
+        with_filter (Bio::Bam::filter &block)
+      end
+
+      def count
+        command = get_command
+        command.push('-c')
+        Bio::Command.call_command_open3(command) do |pin, pout, perr|
+          raise_exception_if_stderr_is_not_empty(perr)
+          pout.readline.to_i
+        end
+      end
+
+      def referencing(chr)
+        iter = self.clone
+        iter.chromosome = chr
+        iter
+      end
+
+      def overlapping(reg)
+        iter = self.clone
+        iter.region = reg
+        iter
+      end
+
+      def [](reg)
+        overlapping(reg)
+      end
+
+      def clone
+        iter = AlignmentIterator.new @command
+        iter.chromosome = chromosome
+        iter.region = region
+        iter
+      end
+
+      attr_accessor :chromosome
+      attr_accessor :region
+      attr_accessor :command
     end
 
   end

data/lib/bio-sambamba/bamfile.rb

@@ -1,14 +1,25 @@
+require 'fileutils'
+
 module Bio
 
+  private
+  RubyFile = File
+
+  public
   # Module for reading BAM files
   module Bam
 
+    public
+
     # Class providing access to BAM files
     class File
-     
+    
+      include FileExistenceChecker
+
       # Creates an object for access to BAM file
       def initialize(filename)
         @filename = filename
+        check_file_existence filename
       end
      
       # SAM header
@@ -18,13 +29,14 @@ module Bio
 
       # Returns an AlignmentIterator object for iterating over all alignments in the file
       def alignments
-        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format=json', @filename]
+        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format', 'json', @filename]
       end
 
       # True if index file was found 
       def has_index?
-        File::File.exists?(@filename + '.bai') || 
-        File::File.exists?(@filename[0...-1] + 'i')
+        fn1 = @filename + '.bai'
+        fn2 = @filename.chomp(RubyFile.extname(@filename)) + '.bai'
+        RubyFile.exists?(fn1) || RubyFile.exists?(fn2)
       end
 
       # Fetches alignments overlapping a region. 
@@ -35,11 +47,17 @@ module Bio
       # * _chr_: reference sequence
       # * _region_: a Range representing an interval. Coordinates are 1-based.
       def fetch(chr, region)
-        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format=json', 
-                                         @filename,
-                                         "#{chr}:#{region.min}-#{region.max}"]
+        iter = Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format=json', 
+                                               @filename]
+        iter.chromosome = chr
+        iter.region = region
+        iter
+      end
+
+      def [](chr)
+        fetch(chr, nil)
       end
-    end
+    end # class File
     
-  end
-end
+  end # module Bam
+end # module Bio

data/lib/bio-sambamba/basepair.rb

@@ -0,0 +1,17 @@
+class Integer
+    def bp
+        self
+    end
+
+    def Kbp
+        self * 1_000
+    end
+
+    def Mbp
+        self * 1_000_000
+    end
+
+    def Gbp
+        self * 1_000_000_000
+    end
+end

data/lib/bio-sambamba/exception.rb

@@ -0,0 +1,16 @@
+module Bio
+
+  module SambambaStderrParser
+    private
+    def raise_exception_if_stderr_is_not_empty(perr)
+      msg = perr.read
+      raise msg unless msg.empty?
+    end
+  end
+
+  module FileExistenceChecker
+    def check_file_existence filename
+      raise "file #{filename} does not exist" unless File.exists? filename
+    end
+  end
+end

data/lib/bio-sambamba/filtering.rb

@@ -0,0 +1,201 @@
+module Bio
+  module Bam
+
+    def self.filter &block
+      AlignmentFilter.new &block
+    end
+
+    class AlignmentFilter
+      def to_s
+        @expression
+      end
+
+      def initialize &block
+        qb = QueryBuilder.new
+        qb.instance_eval &block
+        @expression = qb.expression
+      end
+    end
+
+    private
+    BINARY_OPS = [:>, :<, :>=, :<=, :==, :!=] 
+
+    module ComparisonQueries
+      BINARY_OPS.each do |operator|
+        self.send :define_method, operator do |other|
+          if other.kind_of? String then
+            other = '\'' + other.gsub('\'', '\\\\\'') + '\''
+          end
+          @querybuilder.subexpressions << "#{@name} #{operator} #{other}"
+        end
+      end
+    end
+
+    # see comment above for NumberQueries
+    module StringQueries
+      include ComparisonQueries
+      def =~ regex
+        raise 'operand must be Regexp' unless regex.kind_of? Regexp
+        @querybuilder.subexpressions << "#{@name} =~ #{regex.inspect}"
+      end
+    end
+
+    module FlagQueries
+      def is_set
+        @querybuilder.subexpressions << "#{@name}"
+      end
+
+      def is_unset
+        @querybuilder.subexpressions << "not #{@name}"
+      end
+    end
+
+    class QueryBuilder
+      attr_accessor :subexpressions
+
+      def initialize
+        @subexpressions = []
+        class << @subexpressions
+          def pjoin str
+            self.map{|expr| '(' + expr + ')'}.join str
+          end
+        end
+      end
+
+      def flag
+        FlagQueryBuilder.new(self)
+      end
+
+      def tag tagname
+        TagQueryBuilder.new(self, tagname)
+      end
+
+      @@default_value = { :ref_id          =>  -1,
+                          :mate_ref_id     =>  -1,
+                          :position        =>   0,
+                          :mate_position   =>   0,
+                          :template_length =>   0,
+                          :mapping_quality => 255
+                        }
+
+      [:ref_id, :mate_ref_id,
+       :position, :mate_position,
+       :mapping_quality,
+       :sequence_length,
+       :template_length].each do |integer_field|
+
+        self.send :define_method, integer_field do
+          nb = NumberQueryBuilder.new(self, integer_field)
+
+          if not @@default_value[integer_field].nil?
+            class << nb
+              def is_unknown
+                expr = "#{@name} == #{@@default_value[@name]}"
+                @querybuilder.subexpressions << expr
+              end
+            end
+          end
+
+          nb
+        end
+
+      end
+
+      [:read_name, :sequence, :cigar].each do |string_field|
+        self.send :define_method, string_field do
+          StringQueryBuilder.new(self, string_field)
+        end
+      end
+
+      def negate &block
+        qb = QueryBuilder.new
+        qb.instance_eval &block
+        @subexpressions << ('not (' + qb.expression + ')')
+      end
+
+      def union &block
+        qb = QueryBuilder.new
+        qb.instance_eval &block
+        @subexpressions << (qb.subexpressions.pjoin ' or ')
+        nil
+      end
+
+      def intersection &block
+        qb = QueryBuilder.new
+        qb.instance_eval &block
+        @subexpressions << (qb.subexpressions.pjoin ' and ')
+      end
+
+      def expression
+        subexpressions.pjoin ' and '
+      end
+    end
+
+    class FlagQueryBuilder
+      include FlagQueries
+
+      def initialize(querybuilder)
+        @querybuilder = querybuilder
+      end
+
+      [:unmapped, :mate_is_unmapped, :paired, :proper_pair,
+       :first_of_pair, :second_of_pair, :reverse_strand,
+       :mate_is_reverse_strand, :secondary_alignment,
+       :failed_quality_control, :duplicate].each do |flagname|
+        self.send :define_method, flagname do
+          @name = flagname
+          self
+        end
+      end
+    end
+
+    class TagQueryBuilder
+      include ComparisonQueries
+      include StringQueries
+      def initialize(querybuilder, tagname)
+        @querybuilder = querybuilder
+        @name = '[' + tagname.to_s + ']'
+      end
+    end
+
+    class NumberQueryBuilder
+      include ComparisonQueries
+      BINARY_OPS.each do |op|
+        self.send :define_method, op do |rhs|
+          if not rhs.kind_of? Integer then
+            raise "right-hand side must be an integer, not #{rhs.inspect}"
+          end
+          # 1-based -> 0-based
+          if @name == :position || @name == :mate_position then
+            rhs -= 1
+          end
+          super(rhs)
+        end
+      end
+
+      def initialize(querybuilder, symbol)
+        @querybuilder = querybuilder
+        @name = symbol
+      end
+    end
+
+    class StringQueryBuilder
+      include StringQueries
+
+      BINARY_OPS.each do |op|
+        self.send :define_method, op do |rhs|
+          if not rhs.kind_of? String then
+            raise "right-hand side must be a string, not #{rhs.inspect}"
+          end
+          super(rhs)
+        end
+      end
+
+      def initialize(querybuilder, symbol)
+        @querybuilder = querybuilder
+        @name = symbol
+      end
+    end
+
+  end
+end

data/lib/bio-sambamba/samfile.rb

@@ -4,10 +4,13 @@ module Bio
 
     # Class providing access to SAM files
     class File
+      
+      include FileExistenceChecker
 
       # Creates an object for access to SAM file
       def initialize(filename)
         @filename = filename
+        check_file_existence filename
       end
 
       # SAM header
@@ -17,7 +20,7 @@ module Bio
 
       # Returns an AlignmentIterator object for iterating over all alignments in the file
       def alignments
-        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format=json', '-S', @filename]
+        Bio::Bam::AlignmentIterator.new ['sambamba', 'view', '--format', 'json', '-S', @filename]
       end
     end
 

data/lib/bio-sambamba/samheader.rb

@@ -3,6 +3,7 @@ module Bio
 
     # Represents SAM header
     class SamHeader
+      include SambambaStderrParser
      
       # Creates a new SamHeader object for a specified file,
       # specifying additional options to pass to sambamba tool
@@ -14,9 +15,10 @@ module Bio
       # Raw text of SAM header
       def raw_contents
         if @raw_contents.nil? then
-          @raw_contents = Bio::Command.query_command(['sambamba', 'view', '-H', @filename] + @opts)
-          if @raw_contents.start_with? "sambamba" then
-            raise @raw_contents
+          cmd = ['sambamba', 'view', '-H', @filename] + @opts
+          Bio::Command.call_command_open3(cmd) do |pin, pout, perr|
+            @raw_contents = pout.read
+            raise_exception_if_stderr_is_not_empty(perr)
           end
         end
         @raw_contents
@@ -55,9 +57,12 @@ module Bio
       private
       # Calls sambamba to get underlying JSON object
       def get_json
-        command = ['sambamba', 'view', '-H', '--format=json', @filename] + @opts
-        line = Bio::Command.query_command(command)
-        raise line if line[0] != '{'
+        cmd = ['sambamba', 'view', '-H', '--format=json', @filename] + @opts
+        line = ''
+        Bio::Command.call_command_open3(cmd) do |pin, pout, perr|
+          line = pout.read
+          raise_exception_if_stderr_is_not_empty(perr)
+        end
         @json = Oj.load(line)
       end
     end

metadata

@@ -1,121 +1,128 @@
---- !ruby/object:Gem::Specification
+--- !ruby/object:Gem::Specification 
 name: bio-sambamba
-version: !ruby/object:Gem::Version
-  version: 0.0.2
+version: !ruby/object:Gem::Version 
+  hash: 1889055196096400351
   prerelease: 
+  segments: 
+  - 0
+  - 0
+  - 3
+  version: 0.0.3
 platform: ruby
-authors:
+authors: 
 - Artem Tarasov
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-07-17 00:00:00.000000000 Z
-dependencies:
-- !ruby/object:Gem::Dependency
+
+date: 2012-08-15 00:00:00 Z
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  type: :runtime
+  prerelease: false
   name: bio
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id001 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
+      - !ruby/object:Gem::Version 
+        hash: 3725784833518553922
+        segments: 
+        - 1
+        - 4
+        - 2
         version: 1.4.2
+  requirement: *id001
+- !ruby/object:Gem::Dependency 
   type: :runtime
   prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.4.2
-- !ruby/object:Gem::Dependency
   name: oj
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id002 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.2.9
-  type: :runtime
+      - !ruby/object:Gem::Version 
+        hash: 3635686038185777981
+        segments: 
+        - 1
+        - 3
+        - 4
+        version: 1.3.4
+  requirement: *id002
+- !ruby/object:Gem::Dependency 
+  type: :development
   prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.2.9
-- !ruby/object:Gem::Dependency
   name: bundler
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id003 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
+      - !ruby/object:Gem::Version 
+        hash: 1287940794363108929
+        segments: 
+        - 1
+        - 1
+        - 4
         version: 1.1.4
+  requirement: *id003
+- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.1.4
-- !ruby/object:Gem::Dependency
   name: jeweler
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id004 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
+      - !ruby/object:Gem::Version 
+        hash: 4544990976642784663
+        segments: 
+        - 1
+        - 8
+        - 3
         version: 1.8.3
+  requirement: *id004
+- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.8.3
-- !ruby/object:Gem::Dependency
   name: rspec
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id005 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
+      - !ruby/object:Gem::Version 
+        hash: 3576421176010082098
+        segments: 
+        - 2
+        - 7
+        - 0
         version: 2.7.0
+  requirement: *id005
+- !ruby/object:Gem::Dependency 
   type: :development
   prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 2.7.0
-- !ruby/object:Gem::Dependency
   name: cucumber
-  requirement: !ruby/object:Gem::Requirement
+  version_requirements: &id006 !ruby/object:Gem::Requirement 
     none: false
-    requirements:
+    requirements: 
     - - ~>
-      - !ruby/object:Gem::Version
+      - !ruby/object:Gem::Version 
+        hash: 242951009964279202
+        segments: 
+        - 1
+        - 2
+        - 0
         version: 1.2.0
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    none: false
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: 1.2.0
-description: New Sambamba library comes with a command-line tool for working with
-  SAM/BAM files. This gem brings some of its functionality to Ruby.
+  requirement: *id006
+description: New Sambamba library comes with a command-line tool for working with SAM/BAM files. This gem brings some of its functionality to Ruby.
 email: lomereiter@gmail.com
 executables: []
+
 extensions: []
-extra_rdoc_files:
+
+extra_rdoc_files: 
 - LICENSE.txt
 - README.md
-files:
+files: 
 - .document
 - .travis.yml
 - Gemfile
@@ -123,12 +130,15 @@ files:
 - README.md
 - Rakefile
 - VERSION
+- features/filtering.feature
 - features/iterate-alignments.feature
 - features/random-access.feature
 - features/sam-header.feature
+- features/step_definitions/filtering_steps.rb
 - features/step_definitions/iterate-alignments_steps.rb
 - features/step_definitions/random-access_steps.rb
 - features/step_definitions/sam-header_steps.rb
+- features/step_definitions/syntax-sugar_steps.rb
 - features/step_definitions/validation-steps.rb
 - features/support/env.rb
 - features/syntax-sugar.feature
@@ -137,6 +147,9 @@ files:
 - lib/bio-sambamba/alignment.rb
 - lib/bio-sambamba/alignmentiterator.rb
 - lib/bio-sambamba/bamfile.rb
+- lib/bio-sambamba/basepair.rb
+- lib/bio-sambamba/exception.rb
+- lib/bio-sambamba/filtering.rb
 - lib/bio-sambamba/samfile.rb
 - lib/bio-sambamba/samheader.rb
 - test/data/bins.bam
@@ -154,31 +167,37 @@ files:
 - test/data/wrong_bc_subfield_length.bam
 - test/data/wrong_extra_gzip_length.bam
 homepage: http://github.com/lomereiter/bioruby-sambamba
-licenses:
+licenses: 
 - MIT
 post_install_message: 
 rdoc_options: []
-require_paths:
+
+require_paths: 
 - lib
-required_ruby_version: !ruby/object:Gem::Requirement
+required_ruby_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
-      segments:
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 2002549777813010636
+      segments: 
       - 0
-      hash: 1006609026242934180
-required_rubygems_version: !ruby/object:Gem::Requirement
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
   none: false
-  requirements:
-  - - ! '>='
-    - !ruby/object:Gem::Version
-      version: '0'
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 2002549777813010636
+      segments: 
+      - 0
+      version: "0"
 requirements: []
+
 rubyforge_project: 
-rubygems_version: 1.8.24
+rubygems_version: 1.8.12
 signing_key: 
 specification_version: 3
 summary: Ruby wrapper for Sambamba tool
 test_files: []
+