bmg 0.21.0 → 0.21.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: ef6b320621feeff681bb76b804da274ae055573c
-  data.tar.gz: 2a119b1aa5838fbb2b657332de25fd22cc10dbb6
+  metadata.gz: 83f3d7ebfe5587af85b9823b7ab5a9d11d6e9b30
+  data.tar.gz: 34418e3d5ccfe156c3b0320e6821030a09ba1cdc
 SHA512:
-  metadata.gz: e33cda560cb1c3f119940745c63f45884a406d7b0bce113d08a43d2f374a9a658ab47667c63919d16d5863f6867f2d29808d23a049efe9e34487a282e29c9c82
-  data.tar.gz: ec437125ffbf4e41633b33c6ddc5eedeccef915dd2201f7d0f2a54e8ba348f272e67195ebc57cd8a8e839b07a0fa1aacb957eacf6723d34a42b03dac1d4783ec
+  metadata.gz: f1e68f36ad0cd1a5c9ba2d2dd5fdcd8f02220c3588e5ef995439db58c46dab2bb6fa91ff249f3541ae2096e9d12f3c0d9a9d424c99310c7e90fef7cd4f460994
+  data.tar.gz: 0c9c97791462f47799be4ae510d5be7b0aa3c30360592899fbfb336f31c25762d8fe5d35d87202413f85916ce8fc9199c48d40a26a22b480fc72d497edbe1f1d

data/README.md

@@ -1,6 +1,6 @@
 # Bmg, a relational algebra (Alf's successor)!
 
-Bmg is a relational algebra implemented as a ruby library. It implements the
+Bmg is a relational algebra implemented as a Ruby library. It implements the
 [Relation as First-Class Citizen](http://www.try-alf.org/blog/2013-10-21-relations-as-first-class-citizen)
 paradigm contributed with [Alf](http://www.try-alf.org/) a few years ago.
 
@@ -15,7 +15,7 @@ further down this README.
 * [Where are base relations coming from?](#where-are-base-relations-coming-from)
   * [Memory relations](#memory-relations)
   * [Connecting to SQL databases](#connecting-to-sql-databases)
-  * [Reading files (csv, excel, text)](#reading-files-csv-excel-text)
+  * [Reading files (csv, Excel, text)](#reading-files-csv-excel-text)
   * [Connecting to Redis databases](#connecting-to-redis-databases)
   * [Your own relations](#your-own-relations)
 * [List of supported operators](#supported-operators)
@@ -117,7 +117,7 @@ Bmg.sequel(:suppliers, sequel_db)
 #   {:array=>false})
 ```
 
-### Reading files (csv, excel, text)
+### Reading files (csv, Excel, text)
 
 Bmg provides simple adapters to read files and reach Relationland as soon as
 possible.
@@ -129,7 +129,7 @@ csv_options = { col_sep: ",", quote_char: '"' }
 r = Bmg.csv("path/to/a/file.csv", csv_options)
 ```
 
-Options are directly transmitted to `::CSV.new`, check ruby's standard
+Options are directly transmitted to `::CSV.new`, check Ruby's standard
 library.
 
 #### Excel files
@@ -302,7 +302,7 @@ r.where(predicate)                           # alias for restrict(predicate)
         .where(...)
       ```
 
-2. Bmg supports in memory relations, json relations, csv relations, SQL
+2. Bmg supports in-memory relations, JSON relations, csv relations, SQL
    relations and so on. It's not tight to SQL generation, and supports
    queries accross multiple data sources.
 
@@ -312,8 +312,8 @@ r.where(predicate)                           # alias for restrict(predicate)
 4. Bmg supports various *structuring* operators (group, image, autowrap,
    autosummarize, etc.) and allows building 'non flat' relations.
 
-5. Bmg can use full ruby power when that helps (e.g. regular expressions in
-   WHERE clauses or ruby code in EXTEND clauses). This may prevent Bmg from
+5. Bmg can use full Ruby power when that helps (e.g. regular expressions in
+   WHERE clauses or Ruby code in EXTEND clauses). This may prevent Bmg from
    delegating work to underlying data sources (e.g. SQL server) and should
    therefore be used with care though.
 
@@ -323,14 +323,14 @@ If you use Alf (or used it in the past), below are the main differences between
 Bmg and Alf. Bmg has NOT been written to be API-compatible with Alf and will
 probably never be.
 
-1. Bmg's implementation is much simpler than Alf and uses no ruby core
+1. Bmg's implementation is much simpler than Alf and uses no Ruby core
    extention.
 
 2. We are confident using Bmg in production. Systematic inspection of query
    plans is advised though. Alf was a bit too experimental to be used on
    (critical) production systems.
 
-3. Alf exposes a functional syntax, command line tool, restful tools and
+3. Alf exposes a functional syntax, command-line tool, restful tools and
    many more. Bmg is limited to the core algebra, main Relation abstraction
    and SQL generation.
 

data/lib/bmg/operator/join.rb

@@ -70,6 +70,29 @@ module Bmg
         end
       end
 
+      def _restrict(type, predicate)
+        return super if left_join?
+
+        # right_p makes no reference to attributes in left => full on right
+        # left_p makes no reference to attributes in right => full on left
+        up1, right_p = predicate.and_split(left.type.attrlist! - on)
+        up2, left_p = predicate.and_split(right.type.attrlist! - on)
+        if right_p.tautology? && left_p.tautology?
+          # no optimization can be done
+          super
+        else
+          # remains the full reduction, that up2 is not, since we got it
+          # from predicate, not from up1
+          remains, _ = up1.and_split(right.type.attrlist! - on)
+          left
+            .restrict(left_p)
+            .join(right.restrict(right_p), on)
+            .restrict(remains)
+        end
+      rescue UnknownAttributesError
+        super
+      end
+
       def _unautowrap(operand, options)
         return [operand, false] unless operand.is_a?(Operator::Autowrap)
         return [operand, false] unless operand.same_options?(options)

data/lib/bmg/operator/restrict.rb

@@ -51,7 +51,7 @@ module Bmg
     protected
 
       def _restrict(type, predicate)
-        Restrict.new(type, @operand, @predicate & predicate)
+        @operand.restrict(@predicate & predicate)
       end
 
     protected ### inspect

data/lib/bmg/reader/csv.rb

@@ -39,7 +39,9 @@ module Bmg
     private
 
       def tuple(row)
-        row.to_hash.each_with_object({}){|(k,v),h| h[k.to_sym] = v }
+        row.to_hash.each_with_object({}){|(k,v),h|
+          h[k.to_sym] = v
+        }
       end
 
       def handle_type(type)

data/lib/bmg/support/ordering.rb

@@ -2,7 +2,13 @@ module Bmg
   class Ordering
 
     def initialize(attrs)
-      @attrs = attrs
+      @attrs = if attrs.empty?
+        []
+      elsif attrs.first.is_a?(Symbol)
+        attrs.map{|a| [a, :asc] }
+      else
+        attrs
+      end
     end
     attr_reader :attrs
 
@@ -33,5 +39,9 @@ module Bmg
       0
     end
 
+    def to_a
+      attrs.to_a
+    end
+
   end # class Ordering
 end # module Bmg

data/lib/bmg/support/output_preferences.rb

@@ -3,6 +3,7 @@ module Bmg
 
     DEFAULT_PREFS = {
       attributes_ordering: nil,
+      tuple_ordering: nil,
       extra_attributes: :after
     }
 
@@ -17,6 +18,12 @@ module Bmg
       new(arg)
     end
 
+    def tuple_ordering
+      return nil unless to = options[:tuple_ordering]
+
+      @tuple_ordering = Ordering.new(to)
+    end
+
     def attributes_ordering
       options[:attributes_ordering]
     end
@@ -25,10 +32,29 @@ module Bmg
       options[:extra_attributes]
     end
 
+    def grouping_attributes
+      options[:grouping_attributes]
+    end
+
+    def erase_redundance_in_group(before, current)
+      return [nil, current] unless ga = grouping_attributes
+      return [current, current] unless before
+
+      new_before, new_current = current.dup, current.dup
+      ga.each do |attr|
+        return [new_before, new_current] unless before[attr] == current[attr]
+        new_current[attr] = nil
+      end
+      [new_before, new_current]
+    end
+
     def order_attrlist(attrlist)
       return attrlist if attributes_ordering.nil?
+
       index = Hash[attributes_ordering.each_with_index.to_a]
-      attrlist.sort{|a,b|
+      base  = attrlist
+      base  = attrlist & attributes_ordering if extra_attributes == :ignored
+      base.sort{|a,b|
         ai, bi = index[a], index[b]
         if ai && bi
           ai <=> bi

data/lib/bmg/version.rb

@@ -2,7 +2,7 @@ module Bmg
   module Version
     MAJOR = 0
     MINOR = 21
-    TINY  = 0
+    TINY  = 3
   end
   VERSION = "#{Version::MAJOR}.#{Version::MINOR}.#{Version::TINY}"
 end

data/lib/bmg/writer/csv.rb

@@ -16,12 +16,14 @@ module Bmg
         require 'csv'
         string_or_io, to_s = string_or_io.nil? ? [StringIO.new, true] : [string_or_io, false]
         headers, csv = infer_headers(relation.type), nil
-        relation.each do |tuple|
+        previous = nil
+        each_tuple(relation) do |tuple,i|
           if csv.nil?
             headers = infer_headers(tuple) if headers.nil?
             csv_opts = csv_options.merge(headers: headers)
             csv = CSV.new(string_or_io, **csv_opts)
           end
+          previous, tuple = output_preferences.erase_redundance_in_group(previous, tuple)
           csv << headers.map{|h| tuple[h] }
         end
         to_s ? string_or_io.string : string_or_io

data/lib/bmg/writer/xlsx.rb

@@ -6,11 +6,11 @@ module Bmg
       DEFAULT_OPTIONS = {
       }
 
-      def initialize(csv_options, output_preferences = nil)
-        @csv_options = DEFAULT_OPTIONS.merge(csv_options)
+      def initialize(xlsx_options, output_preferences = nil)
+        @xlsx_options = DEFAULT_OPTIONS.merge(xlsx_options)
         @output_preferences = OutputPreferences.dress(output_preferences)
       end
-      attr_reader :csv_options, :output_preferences
+      attr_reader :xlsx_options, :output_preferences
 
       def call(relation, path)
         require 'write_xlsx'
@@ -21,22 +21,24 @@ module Bmg
       attr_reader :workbook, :worksheet
 
       def _call(relation, path)
-        @workbook = WriteXLSX.new(path)
-        @worksheet = workbook.add_worksheet
+        @workbook = xlsx_options[:workbook] || WriteXLSX.new(path)
+        @worksheet = xlsx_options[:worksheet] || workbook.add_worksheet
 
         headers = infer_headers(relation.type)
-        relation.each_with_index do |tuple,i|
+        before = nil
+        each_tuple(relation) 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
+          before, tuple = output_preferences.erase_redundance_in_group(before, tuple)
           headers.each_with_index do |h,j|
             meth, *args = write_pair(tuple[h])
             worksheet.send(meth, 1+i, j, *args)
           end
         end
 
-        workbook.close
+        workbook.close unless xlsx_options[:workbook]
         path
       end
 

data/lib/bmg/writer.rb

@@ -12,6 +12,17 @@ module Bmg
       attrlist ? output_preferences.order_attrlist(attrlist) : nil
     end
 
+    def each_tuple(relation, &bl)
+      if ordering = output_preferences.tuple_ordering
+        relation
+          .to_a
+          .sort{|t1,t2| ordering.compare_attrs(t1, t2) }
+          .each_with_index(&bl)
+      else
+        relation.each_with_index(&bl)
+      end
+    end
+
   end # module Writer
 end # module Bmg
 require_relative 'writer/csv'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: bmg
 version: !ruby/object:Gem::Version
-  version: 0.21.0
+  version: 0.21.3
 platform: ruby
 authors:
 - Bernard Lambeau
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-05-18 00:00:00.000000000 Z
+date: 2024-01-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: predicate