remote_table 3.0.0.beta → 3.0.0.rc1

data/CHANGELOG

@@ -1,3 +1,10 @@
+3.0.0.rc1 / 2013-08-08
+
+* Breaking changes
+
+  * :parser object must respond to #call(row), not #parse(row)
+  * Whitespace now automatically "normalized" (/\s+/ -> ' ', stripped) for CSV
+
 3.0.0.beta / 2013-07-30
 
 * Breaking changes

data/lib/remote_table.rb

@@ -110,6 +110,13 @@ class RemoteTable
       uri.to_s
     end
 
+    def normalize_whitespace(v)
+      v = v.to_s.dup
+      v.gsub! WHITESPACE, SINGLE_SPACE
+      v.strip!
+      v
+    end
+
     private
 
     def basename(url)
@@ -129,6 +136,8 @@ class RemoteTable
     end
   end
 
+  WHITESPACE = /\s+/
+  SINGLE_SPACE = ' '
   EXTERNAL_ENCODING = 'UTF-8'
   EXTERNAL_ENCODING_ICONV = 'UTF-8//TRANSLIT'
   GOOGLE_DOCS_SPREADSHEET = [
@@ -343,14 +352,14 @@ class RemoteTable
 
   # @private
   class NullParser
-    def parse(row)
+    def call(row)
       [row]
     end
   end
 
-  # An object that responds to #parse(row) and returns an array of one or more rows.
+  # An object that responds to #call(row) and returns an array of one or more rows.
   #
-  # @return [#parse]
+  # @return [#call]
   def parser
     @final_parser ||= (@parser || NullParser.new)
   end
@@ -452,7 +461,7 @@ class RemoteTable
       mark_download!
       preprocess!
       memo = _each do |row|
-        parser.parse(row).each do |virtual_row|
+        parser.call(row).each do |virtual_row|
           virtual_row.row_hash = ::HashDigest.hexdigest row
           if errata
             next if errata.rejects? virtual_row

data/lib/remote_table/delimited.rb

@@ -31,7 +31,7 @@ class RemoteTable
 
           # represent the row as an array
           array = row.map do |v|
-            v = v.to_s
+            v = RemoteTable.normalize_whitespace v
             if not some_value_present and not keep_blank_rows and v.present?
               some_value_present = true
             end
@@ -46,7 +46,7 @@ class RemoteTable
           # represent the row as a hash
           hash = ::ActiveSupport::OrderedHash.new
           row.each do |k, v|
-            v = v.to_s
+            v = RemoteTable.normalize_whitespace v
             if not some_value_present and not keep_blank_rows and v.present?
               some_value_present = true
             end
@@ -62,14 +62,6 @@ class RemoteTable
       local_copy.cleanup
     end
 
-    # Passes user-specified options in PASSTHROUGH_CSV_SETTINGS.
-    #
-    # Also maps:
-    # * +:headers+ directly
-    # * +:keep_blank_rows+ to the CSV option +:skip_blanks+
-    # * +:delimiter+ to the CSV option +:col_sep+
-    #
-    # @return [Hash]
     def csv_options
       {
         skip_blanks: !keep_blank_rows,
@@ -88,7 +80,7 @@ class RemoteTable
         i = 0
         line = local_copy.encoded_io.gets
         Engine.parse_line(line).map do |v|
-          header = v.to_s.gsub(/\s+/, ' ').strip
+          header = RemoteTable.normalize_whitespace v
           header.present? ? header : "empty_#{i+=1}"
         end
       when Array

data/lib/remote_table/processed_by_nokogiri.rb

@@ -1,8 +1,6 @@
 class RemoteTable
   # Mixed in to process XML and XHTML.
   module ProcessedByNokogiri
-    WHITESPACE = /\s+/
-    SINGLE_SPACE = ' '
     SOFT_HYPHEN = '­'
 
     def preprocess!
@@ -34,8 +32,7 @@ class RemoteTable
         end.map do |cell|
           memo = cell.content.dup
           memo = assume_utf8 memo
-          memo.gsub! WHITESPACE, SINGLE_SPACE
-          memo.strip!
+          memo = RemoteTable.normalize_whitespace memo
           if not some_value_present and not keep_blank_rows and memo.present?
             some_value_present = true
           end

data/lib/remote_table/version.rb

@@ -1,3 +1,3 @@
 class RemoteTable
-  VERSION = '3.0.0.beta'
+  VERSION = '3.0.0.rc1'
 end

data/test/test_parser.rb

@@ -2,9 +2,9 @@ require 'helper'
 
 describe RemoteTable do
   describe ":parser option" do
-    it "takes a parser object that responds to #parse(row) and returns an array of rows" do
+    it "takes a parser object that responds to #call(row) and returns an array of rows" do
       class GradeRangeParser
-        def parse(row)
+        def call(row)
           row['range'].split('-').map do |subrange|
             virtual_row = row.dup
             virtual_row.delete 'range'
@@ -20,5 +20,23 @@ describe RemoteTable do
       t[3].must_equal 'description' => 'bad', 'grade' => 'D'
       t[4].must_equal 'description' => 'bad', 'grade' => 'F'
     end
+
+    it "takes a proc and returns an array of rows" do
+      grade_range_parser_proc = lambda do |row|
+        row['range'].split('-').map do |subrange|
+          virtual_row = row.dup
+          virtual_row.delete 'range'
+          virtual_row['grade'] = subrange
+          virtual_row
+        end
+      end
+      t = RemoteTable.new "file://#{File.expand_path('../data/ranges.csv', __FILE__)}", parser: grade_range_parser_proc
+      t[0].must_equal 'description' => 'great', 'grade' => 'A'
+      t[1].must_equal 'description' => 'great', 'grade' => 'B'
+      t[2].must_equal 'description' => 'ok', 'grade' => 'C'
+      t[3].must_equal 'description' => 'bad', 'grade' => 'D'
+      t[4].must_equal 'description' => 'bad', 'grade' => 'F'
+    end
+
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: remote_table
 version: !ruby/object:Gem::Version
-  version: 3.0.0.beta
+  version: 3.0.0.rc1
   prerelease: 6
 platform: ruby
 authors:
@@ -10,7 +10,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-07-31 00:00:00.000000000 Z
+date: 2013-08-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activesupport