bmg 0.21.2 → 0.21.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: c0ca5faec7c7529e032eca725661a0e480b0e527
-  data.tar.gz: b2891c92b61211b40048d8545988bc1de5fd5bfc
+  metadata.gz: 59da4b61f1d6c332de148552eab6277032df750c
+  data.tar.gz: b65877260b6a5eca59fb864f52a08dad26f917f4
 SHA512:
-  metadata.gz: 399c045c70776108049673574a0f7631dc6b3b70e879b180de00feab49bd4ec4553b991a04549c4a336741cbadd3195204d918d1de5b060ee3195748d7bd2446
-  data.tar.gz: 97304f0d053a2755bfc0a15504610e917686e86f5b8fa7871ff09f66446a6b7cef9aa7043f68637ee5d72ac394838f773522dcfb4a175b2cc7443bb723cb279c
+  metadata.gz: fd85e6e179eb6e1a4d9ffe2f074db3c0d856d239bc65b8d535d6d14cff987a0af1eed75046aed0538726ff87925b9a2a400153a1acec7d7ecc3cc7fdc50d0e87
+  data.tar.gz: 5954aad879e293d34eb175ff54055473e3a983e6a16ec816694eb0b16fd90897cbf9d31bcdf53bfdef664891abb42cb538c7962954d76331c291411e816c956b

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
@@ -245,6 +245,7 @@ r.allbut([:a, :b, ...])                      # remove specified attributes
 r.autowrap(split: '_')                       # structure a flat relation, split: '_' is the default
 r.autosummarize([:a, :b, ...], x: :sum)      # (experimental) usual summarizers supported
 r.constants(x: 12, ...)                      # add constant attributes (sometimes useful in unions)
+r.cross_product(right)                       # cross product, alias `cross_join`
 r.extend(x: ->(t){ ... }, ...)               # add computed attributes
 r.extend(x: :y)                              # shortcut for r.extend(x: ->(t){ t[:y] })
 r.exclude(predicate)                         # shortcut for restrict(!predicate)
@@ -302,7 +303,7 @@ 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 +313,8 @@ 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 +324,14 @@ 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/algebra/shortcuts.rb

@@ -57,6 +57,14 @@ module Bmg
         self.left_join(right.rename(renaming), on.keys, *args)
       end
 
+      def cross_product(right)
+        return join(right) unless self.type.typechecked? || right.type.typechecked?
+        return join(right) unless self.type.knows_attrlist? && right.type.knows_attrlist?
+
+        self.type.cross_join_compatible!(right)
+        return join(right)
+      end
+
       def matching(right, on = [])
         return super unless on.is_a?(Hash)
         renaming = Hash[on.map{|k,v| [v,k] }]

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/type.rb

@@ -310,7 +310,7 @@ module Bmg
       }
     end
 
-  private
+  public
 
     def known_attributes!(attrs)
       extra = attrs - self.attrlist
@@ -331,5 +331,12 @@ module Bmg
       end
     end
 
+    def cross_join_compatible!(right)
+      shared = self.attrlist & right.type.attrlist
+      unless shared.empty?
+        raise TypeError, "Cross product incompatible — duplicate attribute(s): #{shared.join(', ')}"
+      end
+    end
+
   end # class Type
 end # module Bmg

data/lib/bmg/version.rb

@@ -2,7 +2,7 @@ module Bmg
   module Version
     MAJOR = 0
     MINOR = 21
-    TINY  = 2
+    TINY  = 4
   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: 0.21.2
+  version: 0.21.4
 platform: ruby
 authors:
 - Bernard Lambeau
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-10-09 00:00:00.000000000 Z
+date: 2024-03-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: predicate