fastest-csv 0.0.1

data/.gitignore

@@ -0,0 +1,17 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp

data/Gemfile

@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in fastest-csv.gemspec
+gemspec

data/LICENSE

@@ -0,0 +1,22 @@
+Copyright (c) 2012 Maarten Oelering
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,68 @@
+# FastestCSV
+
+Fastest CSV class for MRI Ruby. Faster than faster_csv and fasterer-csv. 
+
+Uses native C code to parse CSV lines. Not (yet) compatible with JRuby.
+
+Supports standard CSV according to RFC4180. Not the so-called "csv" from Excel.
+
+The interface is a subset of the CSV interface in Ruby 1.9.3. The options parameter is not supported.
+
+Originally developed to parse large CSV log files from PowerMTA.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+    gem 'fastest-csv'
+
+And then execute:
+
+    $ bundle
+
+Or install it yourself as:
+
+    $ gem install fastest-csv
+
+## Usage
+
+Parse single line
+
+    FastestCSV.parse_line("one,two,three")
+     => ["one", "two", "three"]
+    
+    "one,two,three".parse_csv
+     => ["one", "two", "three"]
+
+Parse file without header
+
+    FastestCSV.foreach("path/to/file.csv") do |row|
+      while row = csv.shift
+        #
+      end
+    end
+
+Parse file with header
+
+    FastestCSV.open("path/to/file.csv") do |csv|
+      fields = csv.shift
+      while values = csv.shift
+        # 
+      end
+    end
+
+Parse file in array of arrays
+
+    rows = FastestCSV.read("path/to/file.csv")
+
+Parse string in array of arrays
+
+    rows = FastestCSV.parse(csv_data)
+
+## Contributing
+
+1. Fork it
+2. Create your feature branch (`git checkout -b my-new-feature`)
+3. Commit your changes (`git commit -am 'Added some feature'`)
+4. Push to the branch (`git push origin my-new-feature`)
+5. Create new Pull Request

data/Rakefile

@@ -0,0 +1,2 @@
+#!/usr/bin/env rake
+require "bundler/gem_tasks"

data/ext/csv_parser/extconf.rb

@@ -0,0 +1,9 @@
+#!/usr/bin/ruby -w
+
+require 'mkmf'
+
+if RUBY_VERSION =~ /1.8/ then
+  $CPPFLAGS += " -DRUBY_18"
+end
+
+create_makefile('csv_parser')

data/ext/csv_parser/parser.c

@@ -0,0 +1,94 @@
+/*
+ * Copyright (c) Maarten Oelering, BrightCode BV
+ */
+
+#include "ruby.h"
+#ifdef RUBY_18
+  #include "rubyio.h"
+#else
+  #include "ruby/io.h"
+#endif
+
+/* default allocated size is 16 */
+#define DEF_ARRAY_LEN 32
+
+static VALUE cFastestCSV;
+
+static VALUE parse_line(VALUE self, VALUE str)
+{
+    if (NIL_P(str))
+        return Qnil;
+    
+    const char *ptr = RSTRING_PTR(str);
+    int len = (int) RSTRING_LEN(str);  /* cast to prevent warning in 64-bit OS */
+
+    if (len == 0)
+        return Qnil;
+    
+    VALUE array = rb_ary_new2(DEF_ARRAY_LEN); 
+    char value[len];  /* field value, no longer than line */
+    int state = 0;
+    int index = 0;
+    int i;
+    char c;
+    for (i = 0; i < len; i++)
+    {
+        c = ptr[i];
+        switch (c)
+        {
+            case ',':
+                if (state == 0) {
+                    rb_ary_push(array, (index == 0 ? Qnil: rb_str_new(value, index)));
+                    index = 0;
+                }
+                else if (state == 1) {
+                    value[index++] = c;
+                }
+                else if (state == 2) {
+                    rb_ary_push(array, rb_str_new(value, index));
+                    index = 0;
+                    state = 0;  /* outside quoted */
+                }
+                break;
+            case '"':
+                if (state == 0) {
+                    state = 1;  /* in quoted */
+                }
+                else if (state == 1) {
+                    state = 2;  /* quote in quoted */
+                }
+                else if (state == 2) {
+                    value[index++] = c;  /* escaped quote */
+                    state = 1;  /* in quoted */
+                }
+                break;
+            case 13:  /* \r */
+            case 10:  /* \n */
+                if (state == 1) { /* quoted */
+                    value[index++] = c;
+                }
+                else {
+                    /* only do first line */
+                    i = len;
+                }
+                /* else eat it ??? or return so far */
+                break;
+            default:
+                value[index++] = c;
+        }
+    }
+    
+    if (state == 0) {
+        rb_ary_push(array, (index == 0 ? Qnil: rb_str_new(value, index)));
+    }
+    else if (state == 2) {
+        rb_ary_push(array, rb_str_new(value, index));
+    }
+    return array;
+}
+
+void Init_csv_parser()
+{
+    cFastestCSV = rb_define_class("FastestCSV", rb_cObject);
+    rb_define_singleton_method(cFastestCSV, "parse_line", parse_line, 1);
+}

data/fastest-csv.gemspec

@@ -0,0 +1,19 @@
+# -*- encoding: utf-8 -*-
+require File.expand_path('../lib/fastest-csv/version', __FILE__)
+
+Gem::Specification.new do |gem|
+  gem.authors       = ["Maarten Oelering"]
+  gem.email         = ["maarten@brightcode.nl"]
+  gem.description   = %q{Fastest standard CSV parser for MRI Ruby}
+  gem.summary       = %q{Fastest standard CSV parser for MRI Ruby}
+  gem.homepage      = "https://github.com/brightcode/fastest-csv"
+
+  gem.files         = `git ls-files`.split($\)
+  #gem.executables   = gem.files.grep(%r{^bin/}).map{ |f| File.basename(f) }
+  gem.test_files    = gem.files.grep(%r{^(test|spec|features)/})
+  gem.name          = "fastest-csv"
+  gem.require_paths = ["lib"]
+  gem.version       = FastestCSV::VERSION
+  
+  gem.extensions    = ['ext/csv_parser/extconf.rb']
+end

data/lib/fastest-csv.rb

@@ -0,0 +1 @@
+require 'fastest_csv'

data/lib/fastest-csv/version.rb

@@ -0,0 +1,3 @@
+class FastestCSV
+  VERSION = "0.0.1"
+end

data/lib/fastest_csv.rb

@@ -0,0 +1,91 @@
+require 'csv_parser'
+require 'stringio'
+
+class FastestCSV
+  
+  # This method opens an accounting file and passes each record to the provided +block+.
+  def self.foreach(path, &block)
+    open(path) do |reader|
+      reader.each(&block)
+    end
+  end
+
+  # This method opens a csv file. It will pass a Reader object to the provided block,
+  # or return a Reader object when no block is provided.
+  def self.open(path, mode = "rb")
+    csv = new(File.open(path, mode))
+    if block_given?
+      begin
+        yield csv
+      ensure
+        csv.close
+      end
+    else
+      csv
+    end
+  end
+
+  def self.read(path)
+    open(path, "rb") { |csv| csv.read }
+  end
+
+  def self.readlines(path)
+    read(path)
+  end
+
+  def self.parse(data, &block)
+    csv = new(StringIO.new(data))
+    if block.nil?
+      begin
+        csv.read
+      ensure
+        csv.close
+      end
+    else
+      csv.each(&block)
+    end
+  end
+
+  def initialize(io)
+    @io = io
+  end
+  
+  def each
+    while row = shift
+      yield row
+    end
+  end
+  
+  def read
+    table = Array.new
+    each {|row| table << row}
+    table
+  end
+  alias_method :readlines, :read
+
+  def shift
+    if line = @io.gets
+      FastestCSV.parse_line(line)
+    else
+      nil
+    end
+  end
+  alias_method :gets,     :shift
+  alias_method :readline, :shift
+  
+  def close
+    @io.close
+  end
+  
+  def closed?
+    @io.closed?
+  end
+end
+
+class String
+  # Equivalent to <tt>FasterCSV::parse_line(self, options)</tt>.
+  def parse_csv
+    FastestCSV.parse_line(self)
+  end
+end
+

data/test/tc_csv_parsing.rb

@@ -0,0 +1,126 @@
+#
+# Tests copied from faster_csv by James Edward Gray II
+#
+
+require 'test/unit'
+require 'fastest_csv'
+
+# 
+# Following tests are my interpretation of the 
+# {CSV RCF}[http://www.ietf.org/rfc/rfc4180.txt].  I only deviate from that 
+# document in one place (intentionally) and that is to make the default row
+# separator <tt>$/</tt>.
+# 
+class TestCSVParsing < Test::Unit::TestCase
+
+  def test_mastering_regex_example
+    ex = %Q{Ten Thousand,10000, 2710 ,,"10,000","It's ""10 Grand"", baby",10K}
+    assert_equal( [ "Ten Thousand", "10000", " 2710 ", nil, "10,000",
+                    "It's \"10 Grand\", baby", "10K" ],
+                  FastestCSV.parse_line(ex) )
+  end
+
+  # Pulled from:  http://www.ruby-lang.org/cgi-bin/cvsweb.cgi/ruby/test/csv/test_csv.rb?rev=1.12.2.2;content-type=text%2Fplain
+  def test_std_lib_csv
+    [ ["\t", ["\t"]],
+      ["foo,\"\"\"\"\"\",baz", ["foo", "\"\"", "baz"]],
+      ["foo,\"\"\"bar\"\"\",baz", ["foo", "\"bar\"", "baz"]],
+      ["\"\"\"\n\",\"\"\"\n\"", ["\"\n", "\"\n"]],
+      ["foo,\"\r\n\",baz", ["foo", "\r\n", "baz"]],
+      ["\"\"", [""]],
+      ["foo,\"\"\"\",baz", ["foo", "\"", "baz"]],
+      ["foo,\"\r.\n\",baz", ["foo", "\r.\n", "baz"]],
+      ["foo,\"\r\",baz", ["foo", "\r", "baz"]],
+      ["foo,\"\",baz", ["foo", "", "baz"]],
+      ["\",\"", [","]],
+      ["foo", ["foo"]],
+      [",,", [nil, nil, nil]],
+      [",", [nil, nil]],
+      ["foo,\"\n\",baz", ["foo", "\n", "baz"]],
+      ["foo,,baz", ["foo", nil, "baz"]],
+      ["\"\"\"\r\",\"\"\"\r\"", ["\"\r", "\"\r"]],
+      ["\",\",\",\"", [",", ","]],
+      ["foo,bar,", ["foo", "bar", nil]],
+      [",foo,bar", [nil, "foo", "bar"]],
+      ["foo,bar", ["foo", "bar"]],
+      [";", [";"]],
+      ["\t,\t", ["\t", "\t"]],
+      ["foo,\"\r\n\r\",baz", ["foo", "\r\n\r", "baz"]],
+      ["foo,\"\r\n\n\",baz", ["foo", "\r\n\n", "baz"]],
+      ["foo,\"foo,bar\",baz", ["foo", "foo,bar", "baz"]],
+      [";,;", [";", ";"]] ].each do |csv_test|
+      assert_equal(csv_test.last, FastestCSV.parse_line(csv_test.first))
+    end
+    
+    [ ["foo,\"\"\"\"\"\",baz", ["foo", "\"\"", "baz"]],
+      ["foo,\"\"\"bar\"\"\",baz", ["foo", "\"bar\"", "baz"]],
+      ["foo,\"\r\n\",baz", ["foo", "\r\n", "baz"]],
+      ["\"\"", [""]],
+      ["foo,\"\"\"\",baz", ["foo", "\"", "baz"]],
+      ["foo,\"\r.\n\",baz", ["foo", "\r.\n", "baz"]],
+      ["foo,\"\r\",baz", ["foo", "\r", "baz"]],
+      ["foo,\"\",baz", ["foo", "", "baz"]],
+      ["foo", ["foo"]],
+      [",,", [nil, nil, nil]],
+      [",", [nil, nil]],
+      ["foo,\"\n\",baz", ["foo", "\n", "baz"]],
+      ["foo,,baz", ["foo", nil, "baz"]],
+      ["foo,bar", ["foo", "bar"]],
+      ["foo,\"\r\n\n\",baz", ["foo", "\r\n\n", "baz"]],
+      ["foo,\"foo,bar\",baz", ["foo", "foo,bar", "baz"]] ].each do |csv_test|
+      assert_equal(csv_test.last, FastestCSV.parse_line(csv_test.first))
+     end
+  end
+
+  # From:  http://ruby-talk.org/cgi-bin/scat.rb/ruby/ruby-core/6496
+  def test_aras_edge_cases
+    [ [%Q{a,b},               ["a", "b"]],
+      [%Q{a,"""b"""},         ["a", "\"b\""]],
+      [%Q{a,"""b"},           ["a", "\"b"]],
+      [%Q{a,"b"""},           ["a", "b\""]],
+      [%Q{a,"\nb"""},         ["a", "\nb\""]],
+      [%Q{a,"""\nb"},         ["a", "\"\nb"]],
+      [%Q{a,"""\nb\n"""},     ["a", "\"\nb\n\""]],
+      [%Q{a,"""\nb\n""",\nc}, ["a", "\"\nb\n\"", nil]],
+      [%Q{a,,,},              ["a", nil, nil, nil]],
+      [%Q{,},                 [nil, nil]],
+      [%Q{"",""},             ["", ""]],
+      [%Q{""""},              ["\""]],
+      [%Q{"""",""},           ["\"",""]],
+      [%Q{,""},               [nil,""]],
+      [%Q{,"\r"},             [nil,"\r"]],
+      [%Q{"\r\n,"},           ["\r\n,"]],
+      [%Q{"\r\n,",},          ["\r\n,", nil]] ].each do |edge_case|
+        assert_equal(edge_case.last, FastestCSV.parse_line(edge_case.first))
+      end
+  end
+
+  def test_james_edge_cases
+    # A read at eof? should return nil.
+    assert_equal(nil, FastestCSV.parse_line(""))
+    # 
+    # With CSV it's impossible to tell an empty line from a line containing a
+    # single +nil+ field.  The standard CSV library returns <tt>[nil]</tt>
+    # in these cases, but <tt>Array.new</tt> makes more sense to me.
+    # 
+    #assert_equal(Array.new, FastestCSV.parse_line("\n1,2,3\n"))
+    assert_equal([nil], FastestCSV.parse_line("\n1,2,3\n"))
+  end
+
+  def test_rob_edge_cases
+    [ [%Q{"a\nb"},                         ["a\nb"]],
+      [%Q{"\n\n\n"},                       ["\n\n\n"]],
+      [%Q{a,"b\n\nc"},                     ['a', "b\n\nc"]],
+      [%Q{,"\r\n"},                        [nil,"\r\n"]],
+      [%Q{,"\r\n."},                       [nil,"\r\n."]],
+      [%Q{"a\na","one newline"},           ["a\na", 'one newline']],
+      [%Q{"a\n\na","two newlines"},        ["a\n\na", 'two newlines']],
+      [%Q{"a\r\na","one CRLF"},            ["a\r\na", 'one CRLF']],
+      [%Q{"a\r\n\r\na","two CRLFs"},       ["a\r\n\r\na", 'two CRLFs']],
+      [%Q{with blank,"start\n\nfinish"\n}, ['with blank', "start\n\nfinish"]],
+    ].each do |edge_case|
+      assert_equal(edge_case.last, FastestCSV.parse_line(edge_case.first))
+    end
+  end
+
+end

data/test/tc_interface.rb

@@ -0,0 +1,119 @@
+#
+# Tests copied from faster_csv by James Edward Gray II
+#
+
+require 'test/unit'
+require 'fastest_csv'
+
+class TestFastestCSVInterface < Test::Unit::TestCase
+
+  def setup
+    @path = File.join(File.dirname(__FILE__), "temp_test_data.csv")
+    
+    File.open(@path, "w") do |file|
+      file << "1,2,3\r\n"
+      file << "4,5\r\n"
+    end
+
+    @expected = [%w{1 2 3}, %w{4 5}]
+  end
+  
+  def teardown
+    File.unlink(@path)
+  end
+  
+  ### Test Read Interface ###
+  
+  def test_foreach
+    FastestCSV.foreach(@path) do |row|
+      assert_equal(@expected.shift, row)
+    end
+  end
+  
+  def test_open_and_close
+    csv = FastestCSV.open(@path, "r+")
+    assert_not_nil(csv)
+    assert_instance_of(FastestCSV, csv)
+    assert_equal(false, csv.closed?)
+    csv.close
+    assert(csv.closed?)
+    
+    ret = FastestCSV.open(@path) do |csv|
+      assert_instance_of(FastestCSV, csv)
+      "Return value."
+    end
+    assert(csv.closed?)
+    assert_equal("Return value.", ret)
+  end
+  
+  def test_parse
+    data = File.read(@path)
+    assert_equal( @expected,
+                  FastestCSV.parse(data) )
+
+    FastestCSV.parse(data) do |row|
+      assert_equal(@expected.shift, row)
+    end
+  end
+  
+  #def test_parse_line
+  #  row = FasterCSV.parse_line("1;2;3", :col_sep => ";")
+  #  assert_not_nil(row)
+  #  assert_instance_of(Array, row)
+  #  assert_equal(%w{1 2 3}, row)
+  #  
+  #  # shortcut interface
+  #  row = "1;2;3".parse_csv(:col_sep => ";")
+  #  assert_not_nil(row)
+  #  assert_instance_of(Array, row)
+  #  assert_equal(%w{1 2 3}, row)
+  #end
+  
+  def test_parse_line_with_empty_lines
+    assert_equal(nil,       FastestCSV.parse_line(""))  # to signal eof
+    #assert_equal(Array.new, FastestCSV.parse_line("\n1,2,3"))
+    assert_equal([nil], FastestCSV.parse_line("\n1,2,3"))
+  end
+  
+  def test_read_and_readlines
+    assert_equal( @expected,
+                  FastestCSV.read(@path) )
+    assert_equal( @expected,
+                  FastestCSV.readlines(@path))
+    
+    
+    data = FastestCSV.open(@path) do |csv|
+      csv.read
+    end
+    assert_equal(@expected, data)
+    data = FastestCSV.open(@path) do |csv|
+      csv.readlines
+    end
+    assert_equal(@expected, data)
+  end
+  
+  #def test_table
+  #  table = FastestCSV.table(@path)
+  #  assert_instance_of(FastestCSV::Table, table)
+  #  assert_equal([[:"1", :"2", :"3"], [4, 5, nil]], table.to_a)
+  #end
+  
+  def test_shift  # aliased as gets() and readline()
+    FastestCSV.open(@path, "r+") do |csv|
+      assert_equal(@expected.shift, csv.shift)
+      assert_equal(@expected.shift, csv.shift)
+      assert_equal(nil, csv.shift)
+    end
+  end
+
+  def test_long_line # ruby's regex parser may have problems with long rows
+    File.unlink(@path)
+
+    long_field_length = 2800
+    File.open(@path, "w") do |file|
+      file << "1,2,#{'3' * long_field_length}\r\n"
+    end
+    @expected = [%w{1 2} + ['3' * long_field_length]]
+    test_shift
+  end
+end

metadata

@@ -0,0 +1,61 @@
+--- !ruby/object:Gem::Specification
+name: fastest-csv
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+  prerelease: 
+platform: ruby
+authors:
+- Maarten Oelering
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2012-06-28 00:00:00.000000000 Z
+dependencies: []
+description: Fastest standard CSV parser for MRI Ruby
+email:
+- maarten@brightcode.nl
+executables: []
+extensions:
+- ext/csv_parser/extconf.rb
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- LICENSE
+- README.md
+- Rakefile
+- ext/csv_parser/extconf.rb
+- ext/csv_parser/parser.c
+- fastest-csv.gemspec
+- lib/fastest-csv.rb
+- lib/fastest-csv/version.rb
+- lib/fastest_csv.rb
+- test/tc_csv_parsing.rb
+- test/tc_interface.rb
+homepage: https://github.com/brightcode/fastest-csv
+licenses: []
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.24
+signing_key: 
+specification_version: 3
+summary: Fastest standard CSV parser for MRI Ruby
+test_files:
+- test/tc_csv_parsing.rb
+- test/tc_interface.rb