csv_parser 0.0.1

data/lib/csv_parser/parser_extensions.rb

@@ -0,0 +1,75 @@
+module CsvParser
+  module ParserExtensions
+    def field_sep
+      @field_sep ||= ','
+    end
+
+    def field_sep=(str)
+      @field_sep = str
+    end
+
+    def record_sep
+      @record_sep ||= "\n"
+    end
+
+    def record_sep=(str)
+      @record_sep = str
+    end
+
+    def quote_char
+      @quote_char ||= '"'
+    end
+
+    def quote_char=(str)
+      @quote_char = str
+    end
+
+    def allow_empty_record?
+      if defined? @allow_empty_record
+        @allow_empty_record
+      else
+        true
+      end
+    end
+
+    def allow_empty_record=(bool)
+      @allow_empty_record = bool
+    end
+
+    def skip_empty_record?
+      if defined? @skip_empty_record
+        @skip_empty_record
+      else
+        true
+      end
+    end
+
+    def skip_empty_record=(bool)
+      @skip_empty_record = bool
+    end
+
+    def allow_uneven_records?
+      if defined? @allow_uneven_records
+        @allow_uneven_records
+      else
+        true
+      end
+    end
+
+    def allow_uneven_records=(bool)
+      @allow_uneven_records = bool
+    end
+
+    def failure_type
+      @failure_type
+    end
+
+    def failure_index
+      @failure_index || super
+    end
+
+    def warnings
+      @warnings ||= []
+    end
+  end
+end

data/lib/csv_parser/result.rb

@@ -0,0 +1,9 @@
+module CsvParser
+  class Result
+    attr_reader :data, :warnings
+    def initialize(data, warnings)
+      @data = data
+      @warnings = warnings
+    end
+  end
+end

data/lib/csv_parser/version.rb

@@ -0,0 +1,3 @@
+module CsvParser
+  VERSION = "0.0.1"
+end

data/test/helper.rb

@@ -0,0 +1,16 @@
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+
+require 'test/unit'
+require 'byebug'
+
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+require 'csv_parser'

data/test/test_csv_parser.rb

@@ -0,0 +1,226 @@
+require 'helper'
+
+class TestCsvParser < Test::Unit::TestCase
+  def parse(string)
+    parser = CsvParser::CsvParser.new
+    [parser.parse(string), parser.failure_reason]
+  end
+
+  test "one record with two fields" do
+    result, error = parse("foo,bar")
+    assert result, error
+    assert_equal [['foo', 'bar']], result.value
+  end
+
+  test "one record with one field" do
+    result, error = parse("foo")
+    assert result, error
+    assert_equal [['foo']], result.value
+  end
+
+  test "empty records" do
+    result, error = parse("")
+    assert result, error
+    assert_equal [], result.value
+  end
+
+  test "empty record is skipped by default" do
+    result, error = parse("foo\n\nbar")
+    assert result, error
+    assert_equal [['foo'], ['bar']], result.value
+  end
+
+  test "skipping empty record at end" do
+    result, error = parse("foo\nbar\n")
+    assert result, error
+    assert_equal [['foo'], ['bar']], result.value
+  end
+
+  test "not skipping an empty record" do
+    parser = CsvParser::CsvParser.new
+    parser.skip_empty_record = false
+    result = parser.parse("foo\n\nbar")
+    assert_equal [['foo'], [], ['bar']], result.value
+  end
+
+  test "not skipping empty record at end" do
+    parser = CsvParser::CsvParser.new
+    parser.skip_empty_record = false
+    result = parser.parse("foo\nbar\n")
+    assert_equal [['foo'], ['bar'], []], result.value
+  end
+
+  test "two records" do
+    result, error = parse("foo,bar\nbaz,qux")
+    assert result, error
+    assert_equal [['foo', 'bar'], ['baz', 'qux']], result.value
+  end
+
+  test "quoted field" do
+    result, error = parse(%{"foo,bar"})
+    assert result, error
+    assert_equal [["foo,bar"]], result.value
+  end
+
+  test "missing closing quote" do
+    parser = CsvParser::CsvParser.new
+    result = parser.parse(%{"foo})
+    assert !result
+    assert_equal :missing_quote, parser.failure_type
+  end
+
+  test "quote inside unquoted field" do
+    parser = CsvParser::CsvParser.new
+    result = parser.parse(%{f"oo})
+    assert !result
+    assert_equal :stray_quote, parser.failure_type
+  end
+
+  test "missing fields gets warning by default" do
+    parser = CsvParser::CsvParser.new
+    result = parser.parse(%{foo,bar\nbaz})
+    assert result, parser.failure_reason
+    assert_equal [[:missing_fields, 2, 4]], parser.warnings
+  end
+
+  test "missing fields when disallowed" do
+    parser = CsvParser::CsvParser.new
+    parser.allow_uneven_records = false
+    result = parser.parse(%{foo,bar\nbaz})
+    assert !result
+    assert_equal :missing_fields, parser.failure_type
+  end
+
+  test "extra fields gets warning by default" do
+    parser = CsvParser::CsvParser.new
+    result = parser.parse(%{foo\nbar,baz})
+    assert result, parser.failure_reason
+    assert_equal [[:extra_fields, 2, 5]], parser.warnings
+  end
+
+  test "extra fields when disallowed" do
+    parser = CsvParser::CsvParser.new
+    parser.allow_uneven_records = false
+    result = parser.parse(%{foo\nbar,baz})
+    assert !result
+    assert_equal :extra_fields, parser.failure_type
+  end
+
+  test "single-character custom field separator" do
+    parser = CsvParser::CsvParser.new
+    parser.field_sep = "\t"
+    result = parser.parse("foo\tbar")
+    assert result, parser.failure_reason
+    assert_equal [['foo', 'bar']], result.value
+  end
+
+  test "multi-character custom field separator" do
+    parser = CsvParser::CsvParser.new
+    parser.field_sep = "foo"
+    result = parser.parse("bazfoobar")
+    assert result, parser.failure_reason
+    assert_equal [['baz', 'bar']], result.value
+  end
+
+  test "single-character custom record separator" do
+    parser = CsvParser::CsvParser.new
+    parser.record_sep = "x"
+    result = parser.parse("fooxbar")
+    assert result, parser.failure_reason
+    assert_equal [['foo'], ['bar']], result.value
+  end
+
+  test "multi-character custom record separator" do
+    parser = CsvParser::CsvParser.new
+    parser.record_sep = "foo"
+    result = parser.parse("barfoobaz")
+    assert result, parser.failure_reason
+    assert_equal [['bar'], ['baz']], result.value
+  end
+
+  test "custom quote character" do
+    parser = CsvParser::CsvParser.new
+    parser.quote_char = "'"
+    result = parser.parse("'foo,bar'")
+    assert result, parser.failure_reason
+    assert_equal [['foo,bar']], result.value
+  end
+
+  test "parse helper" do
+    result = CsvParser.parse("foo,bar")
+    assert_equal [['foo', 'bar']], result.data
+  end
+
+  test "parse helper with options" do
+    result = CsvParser.parse("foo\tbar", :field_sep => "\t")
+    assert_equal [['foo', 'bar']], result.data
+  end
+
+  test "parse helper with missing closing quote" do
+    error = nil
+    begin
+      CsvParser.parse(%{"foo})
+    rescue CsvParser::MissingQuoteError => error
+      assert_equal 1, error.line
+      assert_equal 1, error.column
+      assert_equal "no ending quote found for quote on line 1, column 1", error.message
+    end
+    assert error
+  end
+
+  test "parse helper with stray quote" do
+    error = nil
+    begin
+      CsvParser.parse(%{f"oo})
+    rescue CsvParser::StrayQuoteError => error
+      assert_equal 1, error.line
+      assert_equal 2, error.column
+      assert_equal "invalid quote found on line 1, column 2", error.message
+    end
+    assert error
+  end
+
+  test "parse helper with allowed short records" do
+    result = CsvParser.parse(%{foo,bar\nbaz})
+    assert_equal 1, result.warnings.length
+    assert_kind_of CsvParser::MissingFieldsError, result.warnings[0]
+    error = result.warnings[0]
+    assert_equal 2, error.line
+    assert_equal 4, error.column
+    assert_equal "record on line 2 had too few fields", error.message
+  end
+
+  test "parse helper with disallowed short records" do
+    error = nil
+    begin
+      CsvParser.parse(%{foo,bar\nbaz}, :allow_uneven_records => false)
+    rescue CsvParser::MissingFieldsError => error
+      assert_equal 2, error.line
+      assert_equal 4, error.column
+      assert_equal "record on line 2 had too few fields", error.message
+    end
+    assert error
+  end
+
+  test "parse helper with allowed long records" do
+    result = CsvParser.parse(%{foo\nbar,baz})
+    assert_equal 1, result.warnings.length
+    assert_kind_of CsvParser::ExtraFieldsError, result.warnings[0]
+    error = result.warnings[0]
+    assert_equal 2, error.line
+    assert_equal 5, error.column
+    assert_equal "record on line 2 had too many fields", error.message
+  end
+
+  test "parse helper with disallowed long records" do
+    error = nil
+    begin
+      CsvParser.parse(%{foo\nbar,baz}, :allow_uneven_records => false)
+    rescue CsvParser::ExtraFieldsError => error
+      assert_equal 2, error.line
+      assert_equal 5, error.column
+      assert_equal "record on line 2 had too many fields", error.message
+    end
+    assert error
+  end
+end

metadata

@@ -0,0 +1,116 @@
+--- !ruby/object:Gem::Specification
+name: csv_parser
+version: !ruby/object:Gem::Version
+  version: 0.0.1
+platform: ruby
+authors:
+- Jeremy Stephens
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-11-14 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: treetop
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: test-unit
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+description: CSV parser with advanced error reporting
+email:
+- jeremy.f.stephens@vanderbilt.edu
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- csv_parser.gemspec
+- lib/csv_parser.rb
+- lib/csv_parser.treetop
+- lib/csv_parser/csv_parser.rb
+- lib/csv_parser/parser_extensions.rb
+- lib/csv_parser/result.rb
+- lib/csv_parser/version.rb
+- test/helper.rb
+- test/test_csv_parser.rb
+homepage: https://github.com/coupler/csv_parser
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.0.3
+signing_key: 
+specification_version: 4
+summary: CSV parser with advanced error reporting
+test_files:
+- test/helper.rb
+- test/test_csv_parser.rb