hasherize_csv 0.0.7 → 0.1.0

data/.travis.yml

@@ -9,6 +9,7 @@ rvm:
   - jruby-head
   - 1.9.2
   - 1.9.3
+  - 2.0.0
 gemfile:
   - Gemfile
 notifications:

data/README.md

@@ -3,6 +3,7 @@
 
 Dead simple CSV parsing, with configurable regex selectors if required. 
 Reads line-by-line, so you can parse big CSV files without running out of memory.
+The first line is always assumed to be column headers, and thus keys in the corresponding hash.
 
 ### Simple case
 Given sample_csv.csv
@@ -28,3 +29,33 @@ The output will be
 {'Col1' => 'Val1', 'Col2' => 'Val2', 'Col3' => 'Val3'}
 {'Col1' => 'Val4', 'Col2' => 'Val5', 'Col3' => 'Val6'}
 ```
+
+### Complex case: parse strange line endings and elegantly select quoted fields
+Given complex_sample.csv
+```csv
+"Col1";"Col2";"Col3"\r
+"Val1";"Val2";"Val3"\r
+"Val4";"Val5
+
+Oh my, newlines in the record!";"Val6"\r
+```
+
+and
+```ruby
+require 'hasherize_csv'
+@f = File.new("complex_sample.csv")
+
+#HasherizeCsv yields the value of the first match group in the :value_pattern regex
+@csv = HasherizeCsv::Csv.new(@f, :separator => "\r", :value_pattern => /\"(.*?)\"/m)
+
+@csv.each do |hash|
+   puts hash.inspect
+end
+```
+
+The output will be
+```
+{'Col1' => 'Val1', 'Col2' => 'Val2', 'Col3' => 'Val3'}
+{'Col1' => 'Val4', 'Col2' => 'Val5\n\nOh my, newlines in the record!', 'Col3' => 'Val6'}
+```
+

data/lib/hasherize_csv.rb

@@ -7,37 +7,35 @@ module HasherizeCsv
   end
 
   class Csv 
+    include Enumerable
     attr_accessor :keys, :file, :separator
     def initialize file, opts = {}
       @file = file
       @separator = opts[:separator] || DefaultOpts::DEFAULT[:separator]
       @value_pattern = opts[:value_pattern] || DefaultOpts::DEFAULT[:value_pattern]
       @keys = []
-      next_line { |l| @keys = values_from_line l if !l.nil? }
-    end
-
-    def next_item 
-      next_line { |l|
-        if l.nil?
-          yield nil
-        else
-          yield hashify_values( values_from_line l ) 
-        end
-      }
+      @keys =  values_from_line next_line 
     end
 
     def each
-      while(1) 
-        self.next_item { |hash|
-          return if hash.nil?
-          yield hash
-        }
+      return self.to_enum if !block_given?
+
+      until (hash = next_item).nil?
+        yield hash
       end 
     end
 
     private 
+    def next_item 
+      if (l = next_line).nil?
+        return nil
+      else
+        return hashify_values( values_from_line l ) 
+      end
+    end
+
     def next_line
-      yield(@file.gets(@separator) ? $_.chomp : $_) 
+      @file.gets(@separator) ? $_.chomp : $_
     end
 
     def hashify_values values

data/lib/hasherize_csv/version.rb

@@ -1,3 +1,3 @@
 module HasherizeCsv
-  VERSION = "0.0.7"
+  VERSION = "0.1.0"
 end

data/test/test_hasherize_csv.rb

@@ -25,6 +25,13 @@ class HasherizeCsvTest < Test::Unit::TestCase
     assert_equal ["Heading1", "Heading2", "H_EA_3__c"], @csv.keys
   end
 
+  def test_hash_file_enumerator
+    result = []
+    enumerator = @csv.each
+
+    assert_equal({"Heading1"=>"Item1", "Heading2"=>"Item2", "H_EA_3__c"=>"Item3"}, enumerator.next)
+  end
+
   def test_hash_file
     result = []
     @csv.each { |hash|

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hasherize_csv
 version: !ruby/object:Gem::Version
-  version: 0.0.7
+  version: 0.1.0
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-02-06 00:00:00.000000000 Z
+date: 2013-10-05 00:00:00.000000000 Z
 dependencies: []
 description: Turns csv files into hashes without reading the entire csv into memory
 email: