parse_fasta 1.8.0 → 1.8.1

checksums.yaml

@@ -1,15 +1,15 @@
 ---
 !binary "U0hBMQ==":
   metadata.gz: !binary |-
-    MmIxNzgzM2E2MzkzYTc1ZjM2MWE0NzIxN2ZmZmRkMGUxMDQ0Y2MzYw==
+    YzlkNTQ5NGQ5YTFlNzVkOTJjOTJkMTM2YmUwN2FlMjhmOTg2ZDZlMQ==
   data.tar.gz: !binary |-
-    Njc4ZTVlN2EwMTVjZDkwNWZmYTM2MzcwY2FkY2NlNjVmNmJiNWFmYw==
+    OThhYTU5NTAzYzlkMTg2N2IxOWNjYTExMWEyODRiY2Q2OGFhMzQ4MQ==
 SHA512:
   metadata.gz: !binary |-
-    ZjNiOTc2NTM0OTc0M2U4NDNkYzhlNTBlNjM3MzI4Mjc4YWY4YmE4MjRhMDdj
-    MGRiNWNiMmExZTZjYmRhZDk0YzcxMTQ1OGQ0NzdkYTVmMmQ0Mzg1YmU0NmFl
-    NmNjYWM1MmY4ZWRhNTY5MDAzNTk1YzcyN2IzYWE2MTNkMzAwNDg=
+    NzMyYTNmZmQ0YThlMThkZmE3ZjZhZjAzNDM2MGQ4ZTcwODhkODY3NzI2NzU1
+    NDQzYmU1ZDBiZjljYzVhZmNlMDIzZDMxMDc4Zjk3N2E1YTAxOTUzZTIyOGNj
+    NzdjOWJiODA2ZDA0NGNmMjFkOGI1ZjgxZWY3NTRmMTQ1MDc5MTU=
   data.tar.gz: !binary |-
-    Mjk5ZjU3YWI3YTJlN2Q4NWJjMWY3NDczOTBhNzI3NzlkMDViZjRhZGFkN2I4
-    NmU5YzFhYTI4ZDc2N2RhYTE5ODdkODE5NTQ5ZjJmNzNmNjEyNzY3NzJiZTk3
-    ODA2MWI3YjUzZDdmMTE5MGM1MDA3ZTk1NmMyNGU3NjFmOTIyMWY=
+    ODU2NzMwZTk3ZmE0ZTIxYzMwOWVkMWUyY2U4MTE3YzAzMzI5MzU1ZDAzNWE3
+    OGQ3ODk2ZjQwYTNjNTJlZTVjYzg3MGU5YzliZjAyYjQ4ZDNmNjRlNzE2YmJk
+    MmY2OTRhYjI3NTM3ODFmYWYwNDk2ZjQ0YzI3YjIxMzI3MGU3MmE=

data/.gitignore

@@ -20,3 +20,4 @@ tmp
 *.o
 *.a
 mkmf.log
+.ruby-*

data/README.md

@@ -27,7 +27,7 @@ lightweight than BioRuby. And more fun! ;)
 ## Documentation ##
 
 Checkout
-[parse_fasta docs](http://rubydoc.info/gems/parse_fasta/1.8.0/frames)
+[parse_fasta docs](http://rubydoc.info/gems/parse_fasta)
 for the full api documentation.
 
 ## Usage ##
@@ -73,6 +73,27 @@ Add `Sequence#rev_comp`. It can handle IUPAC characters. Since
 an amino acid string, things will get weird as it will complement the
 IUPAC characters in the AA string and leave others.
 
+#### 1.8.1 ####
+
+An error will be raised if a fasta file has a `>` in the
+sequence. Sometimes files are not terminated with a newline
+character. If this is the case, then catting two fasta files will
+smush the first header of the second file right in with the last
+sequence of the first file. This is bad, raise an error! ;)
+
+Example
+
+    >seq1
+    ACTG>seq2
+    ACTG
+    >seq3
+    ACTG
+
+This will raise `ParseFasta::SequenceFormatError`.
+
+Also, headers with lots of `>` within are fine now.
+
+
 ### 1.7 ###
 
 Add `SeqFile#to_hash`, `FastaFile#to_hash` and `FastqFile#to_hash`.

data/lib/parse_fasta/fasta_file.rb

@@ -51,6 +51,8 @@ class FastaFile < File
   #
   # @return [Hash] A hash with headers as keys, sequences as the
   #   values (Sequence objects)
+  #
+  # @raise [ParseFasta::SequenceFormatError] if sequence has a '>'
   def to_hash
     hash = {}
     self.each_record do |head, seq|
@@ -88,6 +90,8 @@ class FastaFile < File
   #   fasta record. If `separate_lines` is falsy (the default
   #   behavior), will be Sequence, but if truthy will be
   #   Array<String>.
+  #
+  # @raise [ParseFasta::SequenceFormatError] if sequence has a '>'
   def each_record(separate_lines=nil)
     begin
       f = Zlib::GzipReader.open(self)
@@ -100,11 +104,33 @@ class FastaFile < File
         header, sequence = parse_line_separately(line)
         yield(header.strip, sequence)
       end
+
+      # f.each_with_index(">") do |line, idx|
+      #   if idx.zero?
+      #     if line != ">"
+      #       raise ParseFasta::DataFormatError
+      #     end
+      #   else
+      #     header, sequence = parse_line_separately(line)
+      #     yield(header.strip, sequence)
+      #   end
+      # end
     else
       f.each("\n>") do |line|
         header, sequence = parse_line(line)
         yield(header.strip, Sequence.new(sequence || ""))
       end
+
+      # f.each_with_index(sep=/^>/) do |line, idx|
+      #   if idx.zero?
+      #     if line != ">"
+      #       raise ParseFasta::DataFormatError
+      #     end
+      #   else
+      #     header, sequence = parse_line(line)
+      #     yield(header.strip, Sequence.new(sequence || ""))
+      #   end
+      # end
     end
 
     f.close if f.instance_of?(Zlib::GzipReader)
@@ -114,12 +140,12 @@ class FastaFile < File
   private
 
   def parse_line(line)
-    line.split("\n", 2).map { |s| s.gsub(/\n|>/, '') }
+    line.split("\n", 2).map { |s| s.gsub(/\n|^>|>$/, '') }
   end
 
   def parse_line_separately(line)
     header, sequence =
-      line.split("\n", 2).map { |s| s.gsub(/>/, '') }
+      line.split("\n", 2).map { |s| s.gsub(/^>|>$/, '') }
 
     if sequence.nil?
       sequences = []

data/lib/parse_fasta/seq_file.rb

@@ -29,6 +29,9 @@ class SeqFile < File
   #
   # @return [Hash] A hash with headers as keys, sequences as the
   #   values (Sequence objects)
+  #
+  # @raise [ParseFasta::SequenceFormatError] if sequence has a '>',
+  #   and file is a fastA file
   def to_hash
     first_char = get_first_char(self)
 
@@ -73,6 +76,9 @@ class SeqFile < File
   #   leading '>' or '@'
   #
   # @yieldparam sequence [Sequence] The sequence of the record.
+  #
+  # @raise [ParseFasta::SequenceFormatError] if sequence has a '>',
+  #   and file is a fastA file
   def each_record
     first_char = get_first_char(self)
 

data/lib/parse_fasta/sequence.rb

@@ -35,7 +35,13 @@ class Sequence < String
   #
   # @example Removes whitespace
   #   Sequence.new "AA CC TT" #=> "AACCTT"
+  #
+  # @raise [ParseFasta::SequenceFormatError] if sequence has a '>'
   def initialize(str)
+    if str.match(/>/)
+      raise ParseFasta::SequenceFormatError
+    end
+
     super(str.gsub(/ +/, ""))
   end
 

data/lib/parse_fasta/version.rb

@@ -17,5 +17,5 @@
 # along with parse_fasta.  If not, see <http://www.gnu.org/licenses/>.
 
 module ParseFasta
-  VERSION = "1.8.0"
+  VERSION = "1.8.1"
 end

data/lib/parse_fasta.rb

@@ -24,10 +24,17 @@ require 'parse_fasta/sequence'
 require 'parse_fasta/quality'
 
 module ParseFasta
+
+  class Error < StandardError
+  end
+
   # Error raised when FASTA file is malformed
   class DataFormatError < IOError
     def message
       "Data format error -- check input file"
     end
   end
+
+  class SequenceFormatError < Error
+  end
 end

data/spec/lib/fasta_file_spec.rb

@@ -49,6 +49,15 @@ describe FastaFile do
     let(:fname) { "#{File.dirname(__FILE__)}/../../test_files/test.fa.gz" }
     let(:fasta) { FastaFile.open(fname) }
 
+    context "with badly catted fasta" do
+      it "raises ParseFasta::SequenceFormatError" do
+        fname = "#{File.dirname(__FILE__)}/../../test_files/bad.fa"
+
+        expect { FastaFile.open(fname).to_hash }.
+          to raise_error ParseFasta::SequenceFormatError
+      end
+    end
+
     it "reads the records into a hash: header as key and seq as val" do
       expect(fasta.to_hash).to eq records
     end
@@ -66,6 +75,15 @@ describe FastaFile do
     let(:truthy_records) { Helpers::TRUTHY_RECORDS }
     let(:f_handle) { FastaFile.open(@fname).each_record { |s| } }
 
+    context "with badly catted fasta" do
+      it "raises ParseFasta::SequenceFormatError" do
+        fname = "#{File.dirname(__FILE__)}/../../test_files/bad.fa"
+
+        expect { FastaFile.open(fname).each_record {} }.
+          to raise_error ParseFasta::SequenceFormatError
+      end
+    end
+
     shared_examples_for "any FastaFile" do
       context "with no arguments" do
         it "yields proper header and sequence for each record" do

data/spec/lib/seq_file_spec.rb

@@ -26,6 +26,15 @@ describe SeqFile do
       let(:fname) { "#{File.dirname(__FILE__)}/../../test_files/test.fa.gz" }
       let(:fasta) { SeqFile.open(fname) }
 
+      context "with badly catted fasta" do
+        it "raises ParseFasta::SequenceFormatError" do
+          fname = "#{File.dirname(__FILE__)}/../../test_files/bad.fa"
+
+          expect { FastaFile.open(fname).to_hash }.
+            to raise_error ParseFasta::SequenceFormatError
+        end
+      end
+
       it "reads the records into a hash: header as key and seq as val" do
         expect(fasta.to_hash).to eq records
       end
@@ -77,6 +86,15 @@ describe SeqFile do
 
       let(:f_handle) { SeqFile.open(@fname).each_record { |s| } }
 
+      context "with badly catted fasta" do
+        it "raises ParseFasta::SequenceFormatError" do
+          fname = "#{File.dirname(__FILE__)}/../../test_files/bad.fa"
+
+          expect { FastaFile.open(fname).to_hash }.
+            to raise_error ParseFasta::SequenceFormatError
+        end
+      end
+
       shared_examples_for "parsing a fasta file" do
         it "yields proper header and sequence for each record" do
           expect { |b|

data/spec/lib/sequence_spec.rb

@@ -35,6 +35,14 @@ describe Sequence do
       s_no_spaces = "ACTACTACTGCT"
       expect(Sequence.new(s)).to eq s_no_spaces
     end
+
+    context "when sequence has a '>' in it" do
+      it "raises SequenceFormatError" do
+        s = "actg>sequence 3"
+        expect { Sequence.new(s) }.
+          to raise_error ParseFasta::SequenceFormatError
+      end
+    end
   end
 
   describe "#gc" do

data/spec/spec_helper.rb

@@ -29,6 +29,7 @@ module Helpers
              ["empty seq 1", ""],
              ["empty seq 2", ""],
              ["seq3", "yyyyyyyyyyyyyyyNNN"],
+             ["seq 4 > has many '>' in header", "ACTGactg"],
              ["empty seq at end", ""]]
 
   RECORDS_MAP = {
@@ -38,16 +39,18 @@ module Helpers
     "empty seq 1" => "",
     "empty seq 2" => "",
     "seq3" => "yyyyyyyyyyyyyyyNNN",
+    "seq 4 > has many '>' in header" => "ACTGactg",
     "empty seq at end" => ""
   }
 
-
   TRUTHY_RECORDS = [["empty seq at beginning", []],
                     ["seq1 is fun", ["AACTGGNNN"]],
                     ["seq2", ["AAT", "CCTGNNN"]],
                     ["empty seq 1", []],
                     ["empty seq 2", []],
                     ["seq3", ["yyyyyyyyyy", "yyyyy", "NNN"]],
+                    ["seq 4 > has many '>' in header", ["ACTG" ,
+                                                        "actg"]],
                     ["empty seq at end", []]]
 
 end

data/test_files/bad.fa

@@ -0,0 +1,5 @@
+>seq1
+ACTG>seq2
+ACTG
+>seq3
+ACTG

data/test_files/test.fa

@@ -15,4 +15,7 @@ yyyyyyyyyy
 
 yyyyy
 NNN
+>seq 4 > has many '>' in header
+ACTG
+actg
 >empty seq at end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: parse_fasta
 version: !ruby/object:Gem::Version
-  version: 1.8.0
+  version: 1.8.1
 platform: ruby
 authors:
 - Ryan Moore
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-11-17 00:00:00.000000000 Z
+date: 2016-03-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -138,6 +138,7 @@ files:
 - spec/lib/seq_file_spec.rb
 - spec/lib/sequence_spec.rb
 - spec/spec_helper.rb
+- test_files/bad.fa
 - test_files/benchmark.rb
 - test_files/bogus.txt
 - test_files/test.fa