rover-df 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: b003d311b623fdd38fee4c6fa76129ad4bba042e8193c1872928cb05085daad3
-  data.tar.gz: d0c8c04b2a8aec3ea5b7616cbcda61f03a12c96fc8f9a0c7aa29fc898948b759
+  metadata.gz: 2fb7c518e0c1d61e601012d0beff064fe04b6cfcfb852fd97d28b999f1173445
+  data.tar.gz: b1a3f177d05095799dc2f02082d921d87774375e3f91db468539540c0c5d9482
 SHA512:
-  metadata.gz: 8033d8ae9e5fb8c8e767ba68897d37745cc5d35a7a82cb2847e2b1d2c3adf8eeb406914cd47949d8f4c3f21307617ab550f435e7a5c257fe1187ed47dd943829
-  data.tar.gz: acfdca4ad081e2722c4b5269824de123d26aadab28532837d2bfc717c2ca263f73dc4335963beae654b24d4c17cebbefdcce761a946f391340a306c9ca2a8c9b
+  metadata.gz: 5a98bce5cdb1cd8ed2442dd54cdcc0bcb93f39ff0daa4c4386a5805b285bd898a10278aad2522947bc944e4f995c833ebbe96f08b29421e048fd850cb2cf7903
+  data.tar.gz: a64bdc8da3a2202f0b4583d02eb162e07d8ae387bacf4f434ab8b6377a8752da0919c24506616df0bb90667dc1eab9f7b4acd6076ef932d1fb69f57c395e6e00

data/CHANGELOG.md

@@ -1,3 +1,11 @@
+## 0.2.0 (2020-08-17)
+
+- Added `numeric?` and `zip` methods to vectors
+- Changed group calculations to return a data frame instead of a hash
+- Changed `each_row` to return enumerator
+- Improved inspect
+- Fixed `any?`, `all?`, and `uniq` for boolean vectors
+
 ## 0.1.1 (2020-06-10)
 
 - Added methods and options for types

data/README.md

@@ -141,12 +141,6 @@ Or
 df[df[:a] > 100 | df[:b] == "one"]
 ```
 
-Not
-
-```ruby
-df[df[:a] != 100]
-```
-
 ## Operations
 
 Basic operations
@@ -340,7 +334,7 @@ df.to_csv
 
 ## Types
 
-Pass column types when creating a data frame
+You can specify column types when creating a data frame
 
 ```ruby
 Rover::DataFrame.new(data, types: {"a" => :int, "b" => :float})

data/lib/rover/data_frame.rb

@@ -81,8 +81,9 @@ module Rover
       end
     end
 
-    # return each row as a hash
     def each_row
+      return enum_for(:each_row) unless block_given?
+
       size.times do |i|
         yield @vectors.map { |k, v| [k, v[i]] }.to_h
       end
@@ -247,18 +248,19 @@ module Rover
       line_start = 0
       spaces = 2
 
+      summarize = size >= 30
+
       @vectors.each do |k, v|
-        v = v.first(5).to_a
+        v = summarize ? v.first(5).to_a + ["..."] + v.last(5).to_a : v.to_a
         width = ([k] + v).map(&:to_s).map(&:size).max
         width = 3 if width < 3
 
         if lines.empty? || lines[-2].map { |l| l.size + spaces }.sum + width > 120
           line_start = lines.size
           lines << []
-          [size, 5].min.times do |i|
+          v.size.times do |i|
             lines << []
           end
-          lines << [] if size > 5
           lines << []
         end
 
@@ -266,7 +268,6 @@ module Rover
         v.each_with_index do |v2, i|
           lines[line_start + 1 + i] << "%#{width}s" % v2.to_s
         end
-        lines[line_start + 6] << "%#{width}s" % "..." if size > 5
       end
 
       lines.pop

data/lib/rover/group.rb

@@ -7,11 +7,14 @@ module Rover
 
     [:count, :max, :min, :mean, :median, :percentile, :sum].each do |name|
       define_method(name) do |*args|
-        result = {}
+        n = [name, args.first].compact.join("_")
+
+        rows = []
         grouped_dfs.each do |k, df|
-          result[k] = df.send(name, *args)
+          rows << k.merge(n => df.send(name, *args))
         end
-        result
+
+        DataFrame.new(rows)
       end
     end
 
@@ -26,21 +29,15 @@ module Rover
         raise ArgumentError, "Missing keys: #{missing_keys.join(", ")}" if missing_keys.any?
 
         groups = Hash.new { |hash, key| hash[key] = [] }
-        if @columns.size == 1
-          @df[@columns.first].each_with_index do |v, i|
-            groups[v] << i
-          end
-        else
-          i = 0
-          @df.each_row do |row|
-            groups[@columns.map { |c| row[c] }] << i
-            i += 1
-          end
+        i = 0
+        @df.each_row do |row|
+          groups[row.slice(*@columns)] << i
+          i += 1
         end
 
         result = {}
-        groups.each do |k, indexes|
-          result[k] = @df[indexes]
+        groups.keys.each do |k|
+          result[k] = @df[groups[k]]
         end
         result
       end

data/lib/rover/vector.rb

@@ -44,6 +44,10 @@ module Rover
       a
     end
 
+    def numeric?
+      ![:object, :bool].include?(type)
+    end
+
     def size
       @data.size
     end
@@ -51,7 +55,7 @@ module Rover
     alias_method :count, :size
 
     def uniq
-      Vector.new(@data.to_a.uniq)
+      Vector.new(to_a.uniq)
     end
 
     def missing
@@ -226,11 +230,15 @@ module Rover
     end
 
     def all?(&block)
-      @data.to_a.all?(&block)
+      to_a.all?(&block)
     end
 
     def any?(&block)
-      @data.to_a.any?(&block)
+      to_a.any?(&block)
+    end
+
+    def zip(other, &block)
+      to_a.zip(other.to_a, &block)
     end
 
     def first(n = 1)

data/lib/rover/version.rb

@@ -1,3 +1,3 @@
 module Rover
-  VERSION = "0.1.1"
+  VERSION = "0.2.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rover-df
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - Andrew Kane
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-06-10 00:00:00.000000000 Z
+date: 2020-08-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray