eco-helpers 2.0.18 → 2.0.24

data/lib/eco/cli/config/default/workflow.rb

@@ -28,7 +28,10 @@ ASSETS.cli.config do |config|
           cases_with_input =  config.usecases.active(io: io).select do |usecase, data|
             io.class.input_required?(usecase.type)
           end
-          next io unless (!io.input || io.input.empty?) && !cases_with_input.empty?
+
+          input_is_required = !cases_with_input.empty? || io.options.dig(:input, :entries_from)
+          missing_input     = !io.input || io.input.empty?
+          next io unless missing_input && input_is_required
 
           if io.options.dig(:input, :entries_from)
             io = io.new(input: config.input.get(io: io))
@@ -50,8 +53,7 @@ ASSETS.cli.config do |config|
           cases_with_people = config.usecases.active(io: io).select do |usecase, data|
             io.class.people_required?(usecase.type)
           end
-          get_people = io.options.dig(:people, :get, :from) == :remote
-          next io unless !cases_with_people.empty? || get_people
+          next io if cases_with_people.empty? && !io.options.dig(:people, :get)
           io = io.new(people:  config.people(io: io))
         end
 
@@ -64,7 +66,8 @@ ASSETS.cli.config do |config|
 
     wf.before(:usecases) do |wf_cases, io|
       # save partial entries -> should be native to session.workflow
-      partial_update = io.options.dig(:people, :get, :type) == :partial
+      get_people     = io.options.dig(:people, :get)
+      partial_update = get_people && get_people.dig(:type) == :partial
       if !io.options[:dry_run] && partial_update
         partial_file = io.session.config.people.partial_cache
         io.session.file_manager.save_json(io.people, partial_file, :timestamp)
@@ -95,11 +98,12 @@ ASSETS.cli.config do |config|
       if io.session.post_launch.empty?
         wf_post.skip!
       else
-        partial_update = io.options.dig(:people, :get, :type) == :partial
+        get_people     = io.options.dig(:people, :get)
+        partial_update = get_people && get_people.dig(:type) == :partial
         if !io.options[:dry_run] && partial_update
           # get target people afresh
           people = io.session.micro.people_refresh(people: io.people, include_created: true)
-          io     = io.new(people: people)
+          io     = io.base.new(people: people)
         else
           wf_post.skip!
           msg = "Although there are post_launch cases, they will NOT be RUN"
@@ -136,7 +140,8 @@ ASSETS.cli.config do |config|
     end
 
     wf.on(:end) do |wf_end, io|
-      partial_update = io.options.dig(:people, :get, :type) == :partial
+      get_people     = io.options.dig(:people, :get)
+      partial_update = get_people && get_people.dig(:type) == :partial
       unless !io.options[:end_get] || io.options[:dry_run] || partial_update
         people = io.session.micro.people_cache
         io     = io.new(people: people)

data/lib/eco/cli/scripting/args_helpers.rb

@@ -75,10 +75,10 @@ module Eco
         def get_file(key, required: false, should_exist: true)
           filename = get_arg(key, with_param: true)
           if !filename && required
-            puts "You need to specify a file '#{key} file'"
+            puts "You need to specify a file or folder '#{key} file_or_folder'"
             exit(1)
           elsif !file_exists?(filename) && should_exist && required
-            puts "This file doesn't exist '#{filename}'"
+            puts "This file/folder doesn't exist '#{filename}'"
             exit(1)
           end
 

data/lib/eco/csv.rb

@@ -18,8 +18,10 @@ module Eco
         kargs = {headers: true,  skip_blanks: true}.merge(kargs)
 
         args = [file].tap do |arg|
-          coding = Eco::API::Common::Session::FileManager.encoding(file)
-          arg.push("rb:bom|utf-8") if coding == "bom"
+          encoding = Eco::API::Common::Session::FileManager.encoding(file)
+          #encoding = (encoding != "utf-8")? "#{encoding}|utf-8": encoding
+          #arg.push(encoding)
+          arg.push("rb:bom|utf-8") if encoding == "bom"
         end
 
         out = super(*args, **kargs).reject do |row|

data/lib/eco/csv/table.rb

@@ -1,4 +1,3 @@
-
 module Eco
   class CSV
     class Table < ::CSV::Table
@@ -9,6 +8,70 @@ module Eco
         super(to_rows_array(input))
       end
 
+      # @return [Hash] where keys are the groups and the values a `Eco::CSV::Table`
+      def group_by(&block)
+        rows.group_by(&block).transform_values do |rows|
+          self.class.new(rows)
+        end
+      end
+
+      # @return [Eco::CSV::Table]
+      def transform_values
+        transformed_rows = rows.map do |row|
+          res = yield(row)
+          case res
+          when Array
+            ::CSV::Row.new(row.headers, res)
+          when ::CSV::Row
+            res
+          end
+        end
+        self.class.new(transformed_rows)
+      end
+
+      # Slices the selected rows
+      # @return [Eco::CSV::Table]
+      def slice(*index)
+        case index.first
+        when Range, Numeric
+          self.class.new(rows.slice(index.first))
+        else
+          self
+        end
+      end
+
+      # @return [Eco::CSV::Table]
+      def slice_columns(*index)
+        case index.first
+        when Range, Numeric
+          columns_to_table(columns.slice(index.first))
+        when String
+          csv_cols = columns
+          csv_cols = index.each_with_object([]) do |name, cols|
+            col = csv_cols.find {|col| col.first == name}
+            cols << col if col
+          end
+          columns_to_table(csv_cols)
+        else
+          self
+        end
+      end
+
+      # @return [Eco::CSV::Table]
+      def delete_column(i)
+        csv_cols = columns
+        csv_cols.delete(i)
+        columns_to_table(csv_cols)
+      end
+
+      # Adds a new column at the end
+      # @param header_name [String] header of the new column
+      # @return [Eco::CSV::Table] with a new empty column
+      def add_column(header_name)
+        new_col = Array.new(length).unshift(header_name)
+        columns_to_table(columns.push(new_col))
+      end
+
       # @return [Array<::CSV::Row>]
       def rows
         [].tap do |out|
@@ -16,24 +79,40 @@ module Eco
         end
       end
 
+      # It removes all rows where all columns' values are the same
+      def delete_duplicates!
+        unique_rows = []
+        self.by_row!.delete_if do |row|
+          unique_rows.any? {|done| equal_rows?(row, done)}.tap do |found|
+            unique_rows << row unless found
+          end
+        end
+      end
+
+      # @param row1 [CSV:Row] row to be compared
+      # @param row2 [CSV:Row] row to be compared
+      # @param [Boolean] `true` if all values of `row1` are as of `row2`
+      def equal_rows?(row1, row2)
+        row1.fields.zip(row2.fields).all? do |(v1, v2)|
+          v1 == v2
+        end
+      end
+
       # @return [Integer] total number of rows not including the header
       def length
         to_a.length - 1
       end
 
+      def empty?
+        length < 1
+      end
+
       # @return [Array<Array>] each array is the column header followed by its values
       def columns
         to_a.transpose
       end
 
-      # Adds a new column at the end
-      # @param header_name [String] header of the new column
-      # @return [Eco::CSV::Table] with a new empty column
-      def add_column(header_name)
-        new_col = Array.new(length).unshift(header_name)
-        columns_to_table(columns.push(new_col))
-      end
-
+      # Creates a single `Hash` where each key, value is a column (header + values)
       # @note it will override columns with same header name
       # @return [Hash] keys are headers, values are arrays
       def columns_hash
@@ -42,6 +121,17 @@ module Eco
         end.to_h
       end
 
+      # Returns an array of row hashes
+      # @note it will override columns with same header
+      def to_a_h
+        rows.map(&:to_h)
+      end
+
+      # @see #to_a_h
+      def to_array_of_hashes
+        to_a_h
+      end
+
       private
 
       def columns_to_table(columns_array)
@@ -51,24 +141,34 @@ module Eco
 
       def to_rows_array(data)
         case data
-        when Array
-          return data unless data.length > 0
-          if data.first.is_a?(::CSV::Row)
-            data
-          elsif data.first.is_a?(Array)
-            headers  = data.shift
-            data.map do |arr_row|
-              CSV::Row.new(headers, arr_row)
-            end.compact
-          else
-            raise "Expected data that can be transformed into Array<Array>"
-          end
         when ::CSV::Table
           to_rows_array(data.to_a)
         when Hash
           # hash of columns header as key and column array as value
           rows_arrays = [a.keys].concat(a.values.first.zip(*a.values[1..-1]))
           to_rows_array(data.keys)
+        when Enumerable
+          data = data.dup.compact
+          return data unless data.count > 0
+          sample = data.first
+
+          case sample
+          when ::CSV::Row
+            data
+          when Array
+            headers  = data.shift
+            data.map do |arr_row|
+              ::CSV::Row.new(headers, arr_row)
+            end.compact
+          when Hash
+            headers     = sample.keys
+            headers_str = headers.map(&:to_s)
+            data.map do |hash|
+              ::CSV::Row.new(headers_str, hash.values_at(*headers))
+            end.compact
+          else
+            raise "Expected data that can be transformed into Array<::CSV::Row>. Given 'Enumerable' of '#{sample.class}'"
+          end
         else
           raise "Input type not supported. Given: #{data.class}"
         end

data/lib/eco/data/fuzzy_match.rb

@@ -27,17 +27,29 @@ module Eco
         include CharsPositionScore
         include NGramsScore
 
-        def jaro_winkler(str1, str2)
+        def jaro_winkler(str1, str2, **options)
+          return 0 if !str1 || !str2
           options = {
             ignore_case: true,
             weight:      0.25
-          }
+          }.merge(options)
           JaroWinkler.distance(str1, str2, **options)
         end
 
       end
 
       module InstanceMethods
+        FUZZY_MATCH_OPTIONS = [
+          :identities, :groupings, :stop_words, :read,
+          :must_match_grouping, :must_match_at_least_one_word,
+          :gather_last_result, :threshold
+        ]
+
+        JARO_OPTIONS     = [:ignore_case, :weight]
+        NGRAMS_OPTIONS   = [:range]
+        POSITION_OPTIONS = [:max_distance]
+        RESULTS_OPTIONS  = [:order, :threshold]
+
         include StopWords
 
         attr_accessor :fuzzy_options
@@ -46,62 +58,132 @@ module Eco
           @fuzzy_options ||= {}
         end
 
-        def fuzzy_match(haystack = nil, **options)
-          return @fuzzy_match if instance_variable_defined?(:@fuzzy_match)
-          @fuzzy_options = options.merge({
-            stop_words: PREPOSITIONS + PRONOUNS + ARTICLES
-          })
+        def fuzzy_match(haystack_data = nil, **options)
+          if instance_variable_defined?(:@fuzzy_match) && !haystack_data
+            return @fuzzy_match if fuzzy_match_options == fuzzy_match_options(options)
+          end
+          @fuzzy_options = options
           # make it run with a native C extension (for better performance: ~130 % increase of performance)
           ::FuzzyMatch.engine = :amatch
-          haystack = obtain_haystack(haystack).tap do |items|
-            if !fuzzy_read_method && found = items.find {|item| !item.is_a?(String)}
-              raise "To use non String objects as 'haystack' you should provide `read:` or `options[:read]`. Given element: #{found.class}"
-            end
-          end
-          @fuzzy_match = ::FuzzyMatch.new(haystack, fuzzy_options)
+          @fuzzy_match = ::FuzzyMatch.new(haystack(haystack_data), fuzzy_match_options)
         end
 
+        # TODO: integration for options[:unique_words] => to ensure repeated words do not bring down the score are cut by threshold
         # @note
         #   - When the `haystack` elements are **non** `String` objects, it excludes the needle itself from the results
-        # @param needle [String, Object] object is allowed when `fuzzy_options` includes `read:` key
+        # @param needle [String, Object] object is allowed when `fuzzy_options` includes `read:` key.
+        # @param needle_str [String, nil] the actual value of needle_str to be used.
+        # @param haystack [Enumerable] the items to find `needle` among.
         # @return [Eco::Data::FuzzyMatch::Results]
-        def find_all_with_score(needle, **options)
-          results = fuzzy_match(**options).find_all_with_score(needle).each_with_object([]) do |fuzzy_results, results|
+        def find_all_with_score(needle, needle_str: nil, haystack: nil, **options)
+          base_match    = fuzzy_match(haystack, **options)
+          match_results = base_match.find_all_with_score(needle_str || needle)
+          needle_str  ||= item_string(needle)
+          results       = match_results.each_with_object([]) do |fuzzy_results, results|
             item, dice, lev = fuzzy_results
             unless item == needle
-              needle_str = item_string(needle)
-              item_str   = item_string(item)
-              jaro_res   = self.class.jaro_winkler(needle_str, item_str)
-              ngram_res  = self.class.ngrams_score(needle_str, item_str, range: 3..5).ratio
-              wngram_res = self.class.words_ngrams_score(needle_str, item_str, range: 3..7).ratio
-              pos_res    = self.class.chars_position_score(needle_str, item_str).ratio
-              results << Result.new(item, item_str, dice, lev, jaro_res, ngram_res, wngram_res, pos_res)
+              item_str     = item_string(item)
+
+              if item_str.to_s.strip.empty? || needle_str.to_s.strip.empty?
+                dice = lev = jaro_res = ngram_res = ngram_res = wngram_res = pos_res  = 0
+              end
+
+              jaro_res     ||= jaro(needle_str, item_str)
+              ngram_res    ||= ngram(needle_str, item_str)
+              wngram_res   ||= words_ngram(needle_str, item_str)
+              pos_res      ||= position(needle_str, item_str)
+
+              results << Result.new(item, item_str, needle_str, dice, lev, jaro_res, ngram_res, wngram_res, pos_res)
+            end
+          end
+          Results.new(needle, needle_str, results).tap do |res|
+            res.order     = fuzzy_options[:order]     if fuzzy_options[:order]
+            res.threshold = fuzzy_options[:threshold] if fuzzy_options[:threshold]
+          end.relevant_results
+        end
+
+        def recalculate_results(results, needle_str: nil, **options)
+          raise "You should provide a block |needle_str, item_str, needle, item|" unless block_given?
+          new_results = results.each_with_object([]) do |result, new_results|
+            nstr, istr = yield(needle_str || results.value, result.value, results.needle, result.match)
+
+            if istr.to_s.strip.empty?
+              dice = lev = jaro_res = ngram_res = ngram_res = wngram_res = pos_res  = 1
+            elsif nstr.to_s.strip.empty?
+              unless istr = needle_str
+                dice = lev = jaro_res = ngram_res = ngram_res = wngram_res = pos_res  = 0
+              end
             end
+
+            res          = ::FuzzyMatch.score_class.new(nstr, istr) unless dice && lev
+            dice       ||= res&.dices_coefficient_similar || 0
+            lev        ||= res&.levenshtein_similar       || 0
+            jaro_res   ||= jaro(nstr, istr)
+            ngram_res  ||= ngram(nstr, istr)
+            wngram_res ||= words_ngram(nstr, istr)
+            pos_res    ||= position(nstr, istr)
+
+            new_results << Result.new(*result.values_at(:match, :value, :needle_str), dice, lev, jaro_res, ngram_res, wngram_res, pos_res)
           end
-          Results.new(needle, item_string(needle), results)
+          Results.new(results.needle, results.value, new_results).tap do |res|
+            res.order     = options[:order]     if options[:order]
+            res.threshold = options[:threshold] if options[:threshold]
+          end.relevant_results
         end
 
         private
 
+        def jaro(str1, str2)
+          options = fuzzy_options.slice(*JARO_OPTIONS)
+          self.class.jaro_winkler(str1, str2, **options)
+        end
+
+        def ngram(str1, str2)
+          options = { range: 3..5 }.merge(fuzzy_options.slice(*NGRAMS_OPTIONS))
+          self.class.ngrams_score(str1, str2, **options).ratio
+        end
+
+        def words_ngram(str1, str2)
+          options = { range: 3..7 }.merge(fuzzy_options.slice(*NGRAMS_OPTIONS))
+          self.class.words_ngrams_score(str1, str2, **options).ratio
+        end
+
+        def position(str1, str2)
+          options = fuzzy_options.slice(*POSITION_OPTIONS)
+          self.class.chars_position_score(str1, str2, **options).ratio
+        end
+
         # @note
         #   - When used in an `Enumerable` it will use `to_a`, or `values` if it's a `Hash`
         # @param data [Enumerable, nil]
         # @return [Array<Object>] the non-repeated values of `data`
-        def obtain_haystack(data = nil)
+        def haystack(data = nil)
           data = self if self.is_a?(Enumerable) && !data
           raise "'data' should be an Enumerable. Given: #{data.class}" unless data.is_a?(Enumerable)
           data = self.is_a?(Hash) ? self.values.flatten : to_a.flatten
-          data.uniq.compact
+          data.uniq.compact.tap do |items|
+            if !fuzzy_read_method && found = items.find {|item| !item.is_a?(String)}
+              raise "To use non String objects as 'haystack' you should provide `read:` or `options[:read]`. Given element: #{found.class}"
+            end
+          end
         end
 
         def item_string(item, attr = fuzzy_read_method)
           return item if !item || item.is_a?(String) || !attr
+          return attr.call(item) if attr.is_a?(Proc)
           attr = attr.to_sym
           return item.send(attr) if item.respond_to?(attr)
         end
 
+        def fuzzy_match_options(options = nil)
+          options = fuzzy_options unless options
+          options.slice(*FUZZY_MATCH_OPTIONS).merge({
+            stop_words: PREPOSITIONS + PRONOUNS + ARTICLES
+          })
+        end
+
         def fuzzy_read_method
-          fuzzy_options[:read]
+          fuzzy_match_options[:read]
         end
 
       end