rover-df 0.4.1 → 0.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 843e2b3015dd7ad6a9d6a0bb604876a18f3a2f29c37359b50c86087bce64b257
-  data.tar.gz: 4317f3694ee5fe5c4f0a1dd1dcd55aefa51575ad141779e619b80b4f65790057
+  metadata.gz: 1073cc7597ae35e09e20efb34bc8f22026fb597783a2746a69dbc46b6f6ce0cb
+  data.tar.gz: c22191045efeb2892ffc0393ca37e19b700dcaf05ee1761840875e4c4eda8087
 SHA512:
-  metadata.gz: 72d33c6f391854f538271090663316e991baa23d2341d227f135802836891867b0bce0096fc6756a88c3cc121a32725d7c7705ad5fe9f9519587761bd86e3d87
-  data.tar.gz: f487cd44ed31a8a79bf23fbf06dde3b9e5f8a77bb9ef3d56b7f36e7d5df5c3ea4ac3e386907a183afa8c973448d420a827dd3354cc35ea86d4e0625287b0182e
+  metadata.gz: 23c85ed31b9c2f481860d1304fb229094afbe66eef81075e0caa46d8e42c4052081b14cef1120614b04e2a45999a790dc73682b6aa60bf525ac9dda0b2d897ea
+  data.tar.gz: 62ee77f8a3344751828711f0d0e73c3f2ae2ac9525d56c7bfbce114a715d0b1bf2f4e75ee475850e3dab6d606078606e19dfed3627bdb4b6a877928795c82d85

data/CHANGELOG.md

@@ -1,3 +1,12 @@
+## 0.5.0 (2025-06-07)
+
+- Strings and symbols are no longer treated as different keys
+- Changed methods that return column names to always use strings
+- Changed `[]` method to return vector instead of Numo array with range or array argument
+- Changed `first` and `last` methods to return element instead of vector when no arguments
+- Changed `round`, `ceil`, and `floor` methods to always return same type as original vector
+- Dropped support for Ruby < 3.2
+
 ## 0.4.1 (2024-10-07)
 
 - Fixed connection leasing for Active Record 7.2+

data/LICENSE.txt

@@ -1,4 +1,4 @@
-Copyright (c) 2020-2022 Andrew Kane
+Copyright (c) 2020-2025 Andrew Kane
 
 MIT License
 

data/README.md

@@ -97,8 +97,6 @@ Select a column
 df[:a]
 ```
 
-> Note that strings and symbols are different keys, just like hashes. Creating a data frame from Active Record, a CSV, or Parquet uses strings.
-
 Select multiple columns
 
 ```ruby

data/lib/rover/data_frame.rb

@@ -4,7 +4,7 @@ module Rover
       data, options = process_args(args)
 
       @vectors = {}
-      types = options[:types] || {}
+      types = (options[:types] || {}).transform_keys(&:to_s)
 
       if data.is_a?(DataFrame)
         data.vectors.each do |k, v|
@@ -14,7 +14,7 @@ module Rover
         data.to_h.each do |k, v|
           @vectors[k] =
             if v.respond_to?(:to_a)
-              Vector.new(v, type: types[k])
+              Vector.new(v, type: types[k.to_s])
             else
               v
             end
@@ -23,7 +23,7 @@ module Rover
         # handle scalars
         size = @vectors.values.find { |v| v.is_a?(Vector) }&.size || 1
         @vectors.each_key do |k|
-          @vectors[k] = to_vector(@vectors[k], size: size, type: types[k])
+          @vectors[k] = to_vector(@vectors[k], size: size, type: types[k.to_s])
         end
       elsif data.is_a?(Array)
         vectors = {}
@@ -38,12 +38,12 @@ module Rover
           end
         end
         vectors.each do |k, v|
-          @vectors[k] = to_vector(v, type: types[k])
+          @vectors[k] = to_vector(v, type: types[k.to_s])
         end
       elsif defined?(ActiveRecord) && (data.is_a?(ActiveRecord::Relation) || (data.is_a?(Class) && data < ActiveRecord::Base) || data.is_a?(ActiveRecord::Result))
         result = data.is_a?(ActiveRecord::Result) ? data : data.connection_pool.with_connection { |c| c.select_all(data.all.to_sql) }
         result.columns.each_with_index do |k, i|
-          @vectors[k] = to_vector(result.rows.map { |r| r[i] }, type: types[k])
+          @vectors[k] = to_vector(result.rows.map { |r| r[i] }, type: types[k.to_s])
         end
       else
         raise ArgumentError, "Cannot cast to data frame: #{data.class.name}"
@@ -54,6 +54,9 @@ module Rover
         check_key(k)
       end
 
+      # TODO check for duplicate keys
+      @vectors.transform_keys!(&:to_s)
+
       # check sizes
       sizes = @vectors.values.map(&:size).uniq
       if sizes.size > 1
@@ -73,12 +76,12 @@ module Rover
         df = DataFrame.new
         where.each do |k|
           check_column(k)
-          df[k] = @vectors[k]
+          df[k] = @vectors[k.to_s]
         end
         df
       else
         # single column
-        @vectors[where]
+        @vectors[where.to_s]
       end
     end
 
@@ -103,7 +106,7 @@ module Rover
       check_key(k)
       v = to_vector(v, size: size)
       raise ArgumentError, "Size mismatch (given #{v.size}, expected #{size})" if @vectors.any? && v.size != size
-      @vectors[k] = v
+      @vectors[k.to_s] = v
     end
 
     def size
@@ -140,6 +143,7 @@ module Rover
       mapping.each_key do |k|
         check_column(k)
       end
+      mapping = mapping.to_h { |k, v| [k.to_s, v.to_s] }
       # use transform_keys! to preserve order
       @vectors.transform_keys! do |k|
         mapping[k] || k
@@ -148,7 +152,7 @@ module Rover
     end
 
     def delete(key)
-      @vectors.delete(key)
+      @vectors.delete(key.to_s)
     end
 
     def except(*keys)
@@ -163,7 +167,7 @@ module Rover
     end
 
     def include?(key)
-      @vectors.include?(key)
+      @vectors.include?(key.to_s)
     end
 
     def head(n = 5)
@@ -237,6 +241,7 @@ module Rover
 
     def to_csv
       require "csv"
+
       CSV.generate do |csv|
         csv << keys
         numo = vectors.values.map(&:to_numo)
@@ -287,6 +292,7 @@ module Rover
     # for IRuby
     def to_html
       require "iruby"
+
       if size > 7
         # pass 8 rows so maxrows is applied
         IRuby::HTML.table((self[0..4] + self[-4..-1]).to_h, maxrows: 7)
@@ -347,7 +353,7 @@ module Rover
     end
 
     def group(*columns)
-      Group.new(self, columns.flatten)
+      Group.new(self, columns.flatten.map(&:to_s))
     end
 
     [:max, :min, :median, :mean, :percentile, :sum, :std, :var].each do |name|
@@ -549,6 +555,9 @@ module Rover
       check_join_keys(self, self_on)
       check_join_keys(other, other_on)
 
+      self_on.map!(&:to_s)
+      other_on.map!(&:to_s)
+
       indexed = other.to_a.group_by { |r| r.values_at(*other_on) }
       indexed.default = []
 

data/lib/rover/vector.rb

@@ -88,9 +88,10 @@ module Rover
     def [](v)
       if v.is_a?(Vector)
         Vector.new(v.to_numo.mask(@data))
-      else
-        # TODO return vector unless v is an integer in 0.4.0
+      elsif v.is_a?(Numeric)
         @data[v]
+      else
+        Vector.new(@data[v])
       end
     end
 
@@ -204,8 +205,7 @@ module Rover
       if ndigits == 0
         Vector.new(@data.round)
       else
-        # TODO pass type
-        Vector.new(@data.to_a.map { |v| v.round(ndigits) })
+        Vector.new(@data.map { |v| v.round(ndigits) })
       end
     end
 
@@ -213,8 +213,7 @@ module Rover
       if ndigits == 0
         Vector.new(@data.ceil)
       else
-        # TODO pass type
-        Vector.new(@data.to_a.map { |v| v.ceil(ndigits) })
+        Vector.new(@data.map { |v| v.ceil(ndigits) })
       end
     end
 
@@ -222,8 +221,7 @@ module Rover
       if ndigits == 0
         Vector.new(@data.floor)
       else
-        # TODO pass type
-        Vector.new(@data.to_a.map { |v| v.floor(ndigits) })
+        Vector.new(@data.map { |v| v.floor(ndigits) })
       end
     end
 
@@ -327,18 +325,20 @@ module Rover
       to_a.zip(other.to_a, &block)
     end
 
-    # TODO return element instead of vector if no argument in 0.4.0
-    def first(n = 1)
-      if n >= size
+    def first(n = NOT_SET)
+      if n == NOT_SET
+        @data[0]
+      elsif n >= size
         Vector.new(@data)
       else
         Vector.new(@data[0...n])
       end
     end
 
-    # TODO return element instead of vector if no argument in 0.4.0
-    def last(n = 1)
-      if n >= size
+    def last(n = NOT_SET)
+      if n == NOT_SET
+        @data[-1]
+      elsif n >= size
         Vector.new(@data)
       else
         Vector.new(@data[-n..-1])
@@ -398,6 +398,7 @@ module Rover
     # for IRuby
     def to_html
       require "iruby"
+
       if size > 7
         # pass 8 rows so maxrows is applied
         IRuby::HTML.table(first(4).to_a + last(4).to_a, maxrows: 7)

data/lib/rover/version.rb

@@ -1,3 +1,3 @@
 module Rover
-  VERSION = "0.4.1"
+  VERSION = "0.5.0"
 end

data/lib/rover.rb

@@ -40,7 +40,7 @@ module Rover
 
       raise ArgumentError, "Must specify headers" if headers == false
 
-      # TODO use date converter in 0.4.0 - need to test performance
+      # TODO use date converter in 0.5.0 - need to test performance
       table = yield({converters: :numeric}.merge(csv_options))
 
       headers = nil if headers == true
@@ -112,6 +112,7 @@ module Rover
       table = yield
       data = {}
       types ||= {}
+      types = types.transform_keys(&:to_s)
       table.each_column do |column|
         k = column.field.name
         if types[k]

metadata

@@ -1,14 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: rover-df
 version: !ruby/object:Gem::Version
-  version: 0.4.1
+  version: 0.5.0
 platform: ruby
 authors:
 - Andrew Kane
-autorequire:
 bindir: bin
 cert_chain: []
-date: 2024-10-08 00:00:00.000000000 Z
+date: 1980-01-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: numo-narray
@@ -24,7 +23,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 0.9.1.9
-description:
 email: andrew@ankane.org
 executables: []
 extensions: []
@@ -43,7 +41,6 @@ homepage: https://github.com/ankane/rover
 licenses:
 - MIT
 metadata: {}
-post_install_message:
 rdoc_options: []
 require_paths:
 - lib
@@ -51,15 +48,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '3.1'
+      version: '3.2'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.5.16
-signing_key:
+rubygems_version: 3.6.7
 specification_version: 4
 summary: Simple, powerful data frames for Ruby
 test_files: []