bmg 0.17.8 → 0.18.4

data/lib/bmg/summarizer/distinct.rb

@@ -0,0 +1,36 @@
+module Bmg
+  class Summarizer
+    #
+    # Collect the distinct values as an array.
+    #
+    # Example:
+    #
+    #   # direct ruby usage
+    #   Bmg::Summarizer.distinct(:qty).summarize(...)
+    #
+    class Distinct < Summarizer
+
+      # Returns [] as least value.
+      def least()
+        {}
+      end
+
+      # Adds val to the memo array
+      def _happens(memo, val)
+        memo[val] = true
+        memo
+      end
+
+      def finalize(memo)
+        memo.keys
+      end
+
+    end # class Distinct
+
+    # Factors a distinct summarizer
+    def self.distinct(*args, &bl)
+      Distinct.new(*args, &bl)
+    end
+
+  end # class Summarizer
+end # module Bmg

data/lib/bmg/summarizer/multiple.rb

@@ -0,0 +1,46 @@
+module Bmg
+  class Summarizer
+    #
+    # A summarizer that collects multiple summarization as a wrapped
+    # tuple.
+    #
+    # Example:
+    #
+    #   # direct ruby usage
+    #   Bmg::Summarizer.multiple(x: ..., y: ...).summarize(...)
+    #
+    class Multiple < Summarizer
+
+      def initialize(defs)
+        @summarization = Summarizer.summarization(defs)
+      end
+
+      # Returns [] as least value.
+      def least()
+        @summarization.each_pair.each_with_object({}){|(k,v),memo|
+          memo[k] = v.least
+        }
+      end
+
+      # Adds val to the memo array
+      def happens(memo, val)
+        @summarization.each_pair.each_with_object({}){|(k,v),memo2|
+          memo2[k] = v.happens(memo[k], val)
+        }
+      end
+
+      def finalize(memo)
+        @summarization.each_pair.each_with_object({}){|(k,v),memo2|
+          memo2[k] = v.finalize(memo[k])
+        }
+      end
+
+    end # class Multiple
+
+    # Factors a distinct summarizer
+    def self.multiple(defs)
+      Multiple.new(defs)
+    end
+
+  end # class Summarizer
+end # module Bmg

data/lib/bmg/summarizer/percentile.rb

@@ -0,0 +1,79 @@
+module Bmg
+  class Summarizer
+    #
+    # Percentile summarizer.
+    #
+    # Example:
+    #
+    #   # direct ruby usage
+    #   Bmg::Summarizer.percentile(:qty, 50).summarize(...)
+    #
+    class Percentile < Summarizer
+
+      DEFAULT_OPTIONS = {
+        :variant => :continuous
+      }
+
+      def initialize(*args, &bl)
+        @nth = args.find{|a| a.is_a?(Integer) } || 50
+        functor = args.find{|a| a.is_a?(Symbol) } || bl
+        options = args.select{|a| a.is_a?(Hash) }.inject(DEFAULT_OPTIONS){|memo,opts|
+          memo.merge(opts)
+        }.dup
+        super(functor, options)
+      end
+
+      # Returns [] as least value.
+      def least()
+        []
+      end
+
+      # Collects the value
+      def _happens(memo, val)
+        memo << val
+      end
+
+      # Finalizes the computation.
+      def finalize(memo)
+        return nil if memo.empty?
+        index = memo.size.to_f * (@nth.to_f / 100.0)
+        floor, ceil = index.floor, index.ceil
+        ceil +=1 if floor == ceil
+        below = [floor - 1, 0].max
+        above = [[ceil - 1, memo.size - 1].min, 0].max
+        sorted = memo.sort
+        if options[:variant] == :continuous
+          (sorted[above] + sorted[below]) / 2.0
+        else
+          sorted[below]
+        end
+      end
+
+    end # class Avg
+
+    def self.percentile(*args, &bl)
+      Percentile.new(*args, &bl)
+    end
+
+    def self.percentile_cont(*args, &bl)
+      Percentile.new(*(args + [{:variant => :continuous}]), &bl)
+    end
+
+    def self.percentile_disc(*args, &bl)
+      Percentile.new(*(args + [{:variant => :discrete}]), &bl)
+    end
+
+    def self.median(*args, &bl)
+      Percentile.new(*(args + [50]), &bl)
+    end
+
+    def self.median_cont(*args, &bl)
+      Percentile.new(*(args + [50, {:variant => :continuous}]), &bl)
+    end
+
+    def self.median_disc(*args, &bl)
+      Percentile.new(*(args + [50, {:variant => :discrete}]), &bl)
+    end
+
+  end # class Summarizer
+end # module Bmg

data/lib/bmg/support.rb

@@ -1,4 +1,5 @@
 require_relative 'support/tuple_algebra'
 require_relative 'support/tuple_transformer'
 require_relative 'support/keys'
+require_relative 'support/ordering'
 require_relative 'support/output_preferences'

data/lib/bmg/support/ordering.rb

@@ -0,0 +1,20 @@
+module Bmg
+  class Ordering
+
+    def initialize(attrs)
+      @attrs = attrs
+    end
+    attr_reader :attrs
+
+    def comparator
+      ->(t1, t2) {
+        attrs.each do |(attr,direction)|
+          c = t1[attr] <=> t2[attr]
+          return (direction == :desc ? -c : c) unless c==0
+        end
+        0
+      }
+    end
+
+  end # class Ordering
+end # module Bmg

data/lib/bmg/support/tuple_algebra.rb

@@ -19,5 +19,11 @@ module Bmg
     end
     module_function :rename
 
+    def symbolize_keys(h)
+      return h if h.empty?
+      h.each_with_object({}){|(k,v),h| h[k.to_sym] = v }
+    end
+    module_function :symbolize_keys
+
   end # module TupleAlgebra
 end # module Bmg

data/lib/bmg/support/tuple_transformer.rb

@@ -26,17 +26,22 @@ module Bmg
 
       def transform_tuple(tuple, with)
         case with
-        when Symbol
-          tuple.each_with_object({}){|(k,v),dup|
-            dup[k] = transform_attr(v, with)
-          }
-        when Proc
+        when Symbol, Proc, Regexp
           tuple.each_with_object({}){|(k,v),dup|
             dup[k] = transform_attr(v, with)
           }
         when Hash
           with.each_with_object(tuple.dup){|(k,v),dup|
-            dup[k] = transform_attr(dup[k], v)
+            case k
+            when Symbol
+              dup[k] = transform_attr(dup[k], v)
+            when Class
+              dup.keys.each do |attrname|
+                dup[attrname] = transform_attr(dup[attrname], v) if dup[attrname].is_a?(k)
+              end
+            else
+              raise ArgumentError, "Unexpected transformation `#{with.inspect}`"
+            end
           }
         when Array
           with.inject(tuple){|dup,on|
@@ -51,6 +56,9 @@ module Bmg
         case with
         when Symbol
           value.send(with)
+        when Regexp
+          m = with.match(value.to_s)
+          m.nil? ? m : m.to_s
         when Proc
           with.call(value)
         when Hash

data/lib/bmg/version.rb

@@ -1,8 +1,8 @@
 module Bmg
   module Version
     MAJOR = 0
-    MINOR = 17
-    TINY  = 8
+    MINOR = 18
+    TINY  = 4
   end
   VERSION = "#{Version::MAJOR}.#{Version::MINOR}.#{Version::TINY}"
 end

data/lib/bmg/writer.rb

@@ -1 +1,17 @@
+module Bmg
+  module Writer
+
+  protected
+
+    def infer_headers(from)
+      attrlist = if from.is_a?(Type) && from.knows_attrlist?
+        from.to_attrlist
+      elsif from.is_a?(Hash)
+        from.keys
+      end
+      attrlist ? output_preferences.order_attrlist(attrlist) : nil
+    end
+
+  end # module Writer
+end # module Bmg
 require_relative 'writer/csv'

data/lib/bmg/writer/csv.rb

@@ -26,17 +26,6 @@ module Bmg
         to_s ? string_or_io.string : string_or_io
       end
 
-    private
-
-      def infer_headers(from)
-        attrlist = if from.is_a?(Type) && from.knows_attrlist?
-          from.to_attrlist
-        elsif from.is_a?(Hash)
-          from.keys
-        end
-        attrlist ? output_preferences.order_attrlist(attrlist) : nil
-      end
-
     end # class Csv
   end # module Writer
 end # module Bmg

data/lib/bmg/writer/xlsx.rb

@@ -0,0 +1,68 @@
+module Bmg
+  module Writer
+    class Xlsx
+      include Writer
+
+      DEFAULT_OPTIONS = {
+      }
+
+      def initialize(csv_options, output_preferences = nil)
+        @csv_options = DEFAULT_OPTIONS.merge(csv_options)
+        @output_preferences = OutputPreferences.dress(output_preferences)
+      end
+      attr_reader :csv_options, :output_preferences
+
+      def call(relation, path)
+        require 'write_xlsx'
+        dup._call(relation, path)
+      end
+
+    protected
+      attr_reader :workbook, :worksheet
+
+      def _call(relation, path)
+        @workbook = WriteXLSX.new(path)
+        @worksheet = workbook.add_worksheet
+
+        headers = infer_headers(relation.type)
+        relation.each_with_index do |tuple,i|
+          headers = infer_headers(tuple) if headers.nil?
+          headers.each_with_index do |h,i|
+            worksheet.write_string(0, i, h)
+          end if i == 0
+          headers.each_with_index do |h,j|
+            meth, *args = write_pair(tuple[h])
+            worksheet.send(meth, 1+i, j, *args)
+          end
+        end
+
+        workbook.close
+        path
+      end
+
+      def write_pair(value)
+        case value
+        when Numeric
+          [:write_number, value]
+        when Date
+          [:write_date_time, value, date_format]
+        else
+          [:write_string, value.to_s]
+        end
+      end
+
+      def date_format
+        @date_format ||= workbook.add_format(:num_format => 'yyyy-mm-dd')
+      end
+
+    end # class Xlsx
+  end # module Writer
+  module Relation
+
+    def to_xlsx(options = {}, path = nil, preferences = nil)
+      options, path = {}, options unless options.is_a?(Hash)
+      Writer::Xlsx.new(options, preferences).call(self, path)
+    end
+
+  end # module Relation
+end # module Bmg

data/tasks/test.rake

@@ -6,17 +6,24 @@ namespace :test do
   desc "Runs unit tests"
   RSpec::Core::RakeTask.new(:unit) do |t|
     t.pattern = "spec/unit/**/test_*.rb"
-    t.rspec_opts = ["-Ilib", "-Ispec/unit", "--fail-fast", "--color", "--backtrace", "--format=progress"]
+    t.rspec_opts = ["-Ilib", "-Ispec/unit", "--color", "--backtrace", "--format=progress"]
   end
   tests << :unit
 
   desc "Runs integration tests"
   RSpec::Core::RakeTask.new(:integration) do |t|
     t.pattern = "spec/integration/**/test_*.rb"
-    t.rspec_opts = ["-Ilib", "-Ispec/integration", "--fail-fast", "--color", "--backtrace", "--format=progress"]
+    t.rspec_opts = ["-Ilib", "-Ispec/integration", "--color", "--backtrace", "--format=progress"]
   end
   tests << :integration
 
+  desc "Runs github regression tests"
+  RSpec::Core::RakeTask.new(:regression) do |t|
+    t.pattern = "spec/regression/**/test_*.rb"
+    t.rspec_opts = ["-Ilib", "-Ispec/regression", "--color", "--backtrace", "--format=progress"]
+  end
+  tests << :regression
+
   task :all => tests
 end
 

metadata

@@ -1,49 +1,49 @@
 --- !ruby/object:Gem::Specification
 name: bmg
 version: !ruby/object:Gem::Version
-  version: 0.17.8
+  version: 0.18.4
 platform: ruby
 authors:
 - Bernard Lambeau
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-09-10 00:00:00.000000000 Z
+date: 2021-05-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: predicate
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '2.4'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 2.4.0
+        version: 2.5.0
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.5'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '2.4'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 2.4.0
+        version: 2.5.0
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '2.5'
 - !ruby/object:Gem::Dependency
   name: path
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '2.0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '1.3'
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -78,14 +78,28 @@ dependencies:
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '2.7'
+        version: '2.8'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
-        version: '2.7'
+        version: '2.8'
+- !ruby/object:Gem::Dependency
+  name: write_xlsx
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.0'
 - !ruby/object:Gem::Dependency
   name: sequel
   requirement: !ruby/object:Gem::Requirement
@@ -154,6 +168,7 @@ files:
 - lib/bmg/reader.rb
 - lib/bmg/reader/csv.rb
 - lib/bmg/reader/excel.rb
+- lib/bmg/reader/text_file.rb
 - lib/bmg/relation.rb
 - lib/bmg/relation/empty.rb
 - lib/bmg/relation/in_memory.rb
@@ -250,16 +265,21 @@ files:
 - lib/bmg/sql/version.rb
 - lib/bmg/summarizer.rb
 - lib/bmg/summarizer/avg.rb
+- lib/bmg/summarizer/by_proc.rb
 - lib/bmg/summarizer/collect.rb
 - lib/bmg/summarizer/concat.rb
 - lib/bmg/summarizer/count.rb
+- lib/bmg/summarizer/distinct.rb
 - lib/bmg/summarizer/max.rb
 - lib/bmg/summarizer/min.rb
+- lib/bmg/summarizer/multiple.rb
+- lib/bmg/summarizer/percentile.rb
 - lib/bmg/summarizer/stddev.rb
 - lib/bmg/summarizer/sum.rb
 - lib/bmg/summarizer/variance.rb
 - lib/bmg/support.rb
 - lib/bmg/support/keys.rb
+- lib/bmg/support/ordering.rb
 - lib/bmg/support/output_preferences.rb
 - lib/bmg/support/tuple_algebra.rb
 - lib/bmg/support/tuple_transformer.rb
@@ -267,6 +287,7 @@ files:
 - lib/bmg/version.rb
 - lib/bmg/writer.rb
 - lib/bmg/writer/csv.rb
+- lib/bmg/writer/xlsx.rb
 - tasks/gem.rake
 - tasks/test.rake
 homepage: http://github.com/enspirit/bmg
@@ -288,7 +309,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.1.2
+rubygems_version: 3.0.8
 signing_key: 
 specification_version: 4
 summary: Bmg is Alf's successor.