csv2hash 0.6.8 → 0.7.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 7c09804fb3f85eacdc90cbfa95ccee0acc8a4fb8
-  data.tar.gz: d29ac384ef4275b13e087fd0eb504eba2f93d337
+  metadata.gz: d0017cb5da55df2b12a7a2a859cb6735dcfe2e17
+  data.tar.gz: a720666766679ea852bd1337f6d0795073d30f3f
 SHA512:
-  metadata.gz: f2b07b5cfcb24197dec4195873a285f353974960ba29deee2a805da2ee45ce5c8a09838ec04b49fab4e6683672226b46587808ffdaacf066b3c6770f139451e1
-  data.tar.gz: 5f2fda3e1d08e971ad39ad05bcfbf094f29d1d2e5ca9a86a42cf413d5c7eb502a2ce08bdcfca25c0f1afa5ccd31fe4410bea0cc190ff53fec83e1b7bc319dc54
+  metadata.gz: 686218a873857c290b1352f83313031cc29bfb6fb23871eb83c2d3f2b520a59d654c0bc30be9141ea43f83816bc36828b7a44fca86a7e071541dab7186ff6cab
+  data.tar.gz: 7d77cc8c42c5facd840c4b8aa34bb9c140822f0947ab54bb229c1866e98993eb9e0b48cc8b6242451538ef8bd27e239c06cf0447b1b77195c3306b786038aff5

data/CHANGELOG.md

@@ -1,8 +1,15 @@
+### VERSION 0.7.0
+
+* feature
+  * Auto discover is also available for collection.
+
+* [fullchanges](https://github.com/FinalCAD/csv2hash/pull/20)
+
 ### VERSION 0.6.8
 
 * bug fix
   * fix typo on generator
-  
+
 ### VERSION 0.6.7
 
 * enhancements
@@ -10,7 +17,7 @@
 
 * refactoring
   * following rails way for naming file
-  
+
 * [fullchanges](https://github.com/FinalCAD/csv2hash/pull/17)
 
 ### VERSION 0.6.6

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    csv2hash (0.6.8)
+    csv2hash (0.7.0)
       activesupport (~> 4.1)
 
 GEM

data/README.md

@@ -26,8 +26,9 @@ It is a DSL to validate and map a CSV to a Ruby Hash.
     * [define where your data are expected](#define-where-your-data-are-expected)
   * [Samples](#samples)
     * [[MAPPING] Validation of cells with defined precision](#mapping-validation-of-cells-with-defined-precision)
-    * [Auto discover position feature](#auto-discover-position-feature)
+    * [Auto discover position feature in Mapping](#auto-discover-position-feature-in-mapping)
     * [[COLLECTION] Validation of a collection (Regular CSV)](#collection-validation-of-a-collection-regular-csv)
+    * [Auto discover position feature in Collection](#auto-discover-position-feature-in-collection)
     * [Structure validation rules](#structure-validation-rules)
     * [CSV Headers](#csv-headers)
     * [Parser and configuration](#parser-and-configuration)
@@ -210,7 +211,7 @@ class MyParser
 end
 ```
 
-### Auto discover position feature
+### Auto discover position feature in Mapping
 
 This is a special feature for finding the Y index of row where you data start. For instance you have this following data :
 
@@ -242,6 +243,7 @@ became
 cell position: [[0, /Employment/],1], key: 'employment'
 ```
 
+
 ### [COLLECTION] Validation of a collection (Regular CSV)
 
 Consider the following CSV:
@@ -288,6 +290,34 @@ class MyParser
 end
 ```
 
+### Auto discover position feature in Collection
+
+This is a special feature for finding a specific column index on header. For example you have the following data:
+
+
+```
+| Name          | Age           |
+|---------------|---------------|
+| John Doe      | 23            |  
+| Jane Doe      | 28            |   
+|               |               |     
+|               |               |  
+```
+
+You want to extract `Name` and 'Age' for all rows but you want the order of the columns to be able to change. 
+You change the position to the regex of column index you are looking for. So this how the position 
+
+```
+cell position: 0, key: 'name'
+```
+
+can be change to
+
+```
+cell position: /Name/ key: 'name'
+```
+
+
 ### Structure validation rules
 
 You may want to validate some structure, like min or max number of columns, definition accepts structure_rules as a key for the third parameter.

data/UPGRADE.md

@@ -1,5 +1,9 @@
 # Upgrading
 
+# Upgrading from 0.6.8 to 0.7.0
+
+nothing
+
 # Upgrading from 0.6.7 to 0.6.8
 
 nothing

data/lib/csv2hash/discover.rb

@@ -1,22 +1,42 @@
 module Csv2hash
   module Discover
 
-    def find_dynamic_position cell
-      y, x = cell.rules.fetch :position
-      column, matcher = y
-      dynamic_y_axe = data_source.index { |entries| entries[column] =~ matcher }
+    def find_dynamic_position cell, header = nil
+      header.present? ? find_dynamic_position_x(cell, header) : find_dynamic_position_y(cell)
+    end
+
+    private
+      def find_dynamic_position_y cell
+        y, x = cell.rules.fetch :position
+        column, matcher = y
+        dynamic_y_axe = data_source.index { |entries| entries[column] =~ matcher }
 
-      if dynamic_y_axe.nil?
-        if cell.rules.fetch(:allow_blank)
-          return nil
+        if dynamic_y_axe.nil?
+          if cell.rules.fetch(:allow_blank)
+            return nil
+          else
+            raise "Y doesn't found for #{cell.rules[:position]} on :#{cell.rules.fetch(:key)}"
+          end
         else
-          raise "Y doesn't found for #{cell.rules[:position]} on :#{cell.rules.fetch(:key)}"
+          cell.rules[:position] = [dynamic_y_axe, x]
+          cell
         end
-      else
-        cell.rules[:position] = [dynamic_y_axe, x]
-        cell
       end
-    end
 
+      def find_dynamic_position_x cell, header
+        x = cell.rules.fetch :position
+        dynamic_x_axe = header.index { |column| column =~ x }
+
+        if dynamic_x_axe.nil?
+          if cell.rules.fetch(:allow_blank)
+            return nil
+          else
+            raise "Column doesn't found in #{definition.name}"
+          end
+        else
+          cell.rules[:position] = dynamic_x_axe
+          cell
+        end
+      end
   end
 end

data/lib/csv2hash/validator.rb

@@ -5,8 +5,7 @@ module Csv2hash
     include Discover
 
     def validate_rules y=nil
-      find_or_remove_dynamic_fields! if definition.type == Definition::MAPPING
-
+      definition.type == Definition::MAPPING ? find_or_remove_dynamic_fields_on_mapping! : find_or_remove_dynamic_fields_on_collection!(y)
       definition.cells.each do |cell|
         _y, x = position cell.rules.fetch(:position)
         begin
@@ -75,14 +74,14 @@ module Csv2hash
       end
     end
 
-    def find_or_remove_dynamic_fields!
+    def find_or_remove_dynamic_fields_on_mapping!
       cells = definition.cells.dup
       # cells without optional and not found dynamic field
       definition.cells = [].tap do |_cells|
         while(!cells.empty?) do
           cell = cells.pop
           _y, x = cell.rules.fetch(:position)
-          if dynamic_field?(_y)
+          if dynamic_field_for_mapping?(_y)
             begin
               _cell = find_dynamic_position cell
               _cells << _cell
@@ -98,9 +97,35 @@ module Csv2hash
       nil
     end
 
-    def dynamic_field? field
+    def find_or_remove_dynamic_fields_on_collection! y
+      cells = definition.cells.dup
+      # cells without optional and not found dynamic field
+      definition.cells = [].tap do |_cells|
+        while(!cells.empty?) do
+          cell = cells.pop
+          x = cell.rules.fetch(:position)
+          if dynamic_field_for_collection?(x)
+            begin
+              _cell = find_dynamic_position cell, data_source.first
+              _cells << _cell
+            rescue => e
+              self.errors << { y: y, x: x, message: e.message, key: cell.rules.fetch(:key) }
+              raise if break_on_failure
+            end
+          else
+            _cells << cell
+          end
+        end
+      end.compact
+      nil
+    end
+
+    def dynamic_field_for_mapping? field
       field.is_a?(Array)
     end
 
+    def dynamic_field_for_collection? field
+      field.is_a?(Regexp)
+    end
   end
 end

data/lib/csv2hash/version.rb

@@ -1,3 +1,3 @@
 module Csv2hash
-  VERSION = '0.6.8'
+  VERSION = '0.7.0'
 end

data/spec/csv2hash/parser/collection_spec.rb

@@ -34,6 +34,47 @@ module Csv2hash
       end
     end
 
+    context 'discover' do
+
+      context 'when header keyword present' do
+        let(:data_source) { [ ['Name','Age'], [ 'John Doe',34 ], [ 'Jane Doe', 25] ] }
+        before do
+          subject.definition.header_size = 1
+          definition.cells << Cell.new({ position: /Age/, key: 'age' })
+          definition.cells << Cell.new({ position: /Name/, key: 'name' })      
+        end
+        it {
+          expect(subject.tap { |c| c.parse! }.data).to eql(
+            { data: [ { 'name' => 'John Doe', 'age' => 34 }, { 'name' => 'Jane Doe', 'age' => 25} ] }
+          )
+        }
+      end
+
+      context 'when no header keyword found and no break on failure' do
+        let(:data_source) { [ [ 'John Doe',34 ], [ 'Jane Doe', 25] ] }
+        before do
+          subject.break_on_failure = false
+          definition.cells << Cell.new({ position: /Name/, key: 'name' })
+          subject.parse
+        end
+        it{ expect(subject.data).to eql(nil)}
+        it{ expect(subject.errors.first[:message]).to eql("Column doesn't found in foo")}
+      end
+
+      context 'when no header keyword found and break on failure' do
+        let(:data_source) { [ [ 'John Doe',34 ], [ 'Jane Doe', 25] ] }
+        before do
+          subject.break_on_failure = true
+          definition.cells << Cell.new({ position: /Name/, key: 'name' })
+        end
+        it 'should throw exception' do
+          expect {
+            subject.parse
+          }.to raise_error("Column doesn't found in foo")
+        end
+      end
+    end
+
     context 'with nested' do
       let(:data_source) { [ [ 'John Doe', 22 ], [ 'Jane Doe', 19 ] ] }
       before do

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: csv2hash
 version: !ruby/object:Gem::Version
-  version: 0.6.8
+  version: 0.7.0
 platform: ruby
 authors:
 - Joel AZEMAR