yomise 0.1.1 → 0.1.3.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8544346549ebbc5186a149be6d0f777b8f004f2d078314b16a4d9a1ad8467bd3
-  data.tar.gz: cafc602a5d8e25f149cdb966033d95f4833ac061435ac165f6ffa8865baf28c7
+  metadata.gz: 28234d39ab9fcef06703c79054cde5a96f183bff244a7a217b5f2aa2c1672f0d
+  data.tar.gz: 216c59fb5b5e60f0e636e8891a32bc987e378073b7609231728f9375779c412a
 SHA512:
-  metadata.gz: a153bb6e926a94dca2634b7c69e99a40a6d393871ce9f915f8a77f7f9cecebb23e1e20dc22b785d49dbd59c5c55a4f2f69c7faf89e39093346281d1bd37162fe
-  data.tar.gz: fc357a3c2ad6f836d47bfaa9cf628149d191b15797a52f3bc19c1e2d59b54ec620e52585514a7b9f36922a23ae87e7989f1c455a4fd3e718b39087b22205ecd9
+  metadata.gz: 6355f13b441d32c5f73def97805c155a343add2f343f55c0a0c40ea483e6536937529c793afdcb592b80c550a5ccf2551980f2f6462c1fa5aa54be8abc9bdfa0
+  data.tar.gz: '08edbe65b56af24bb6d6a06096128a04edd6fdc37e061209593096fe11e657dde41081204b2625f3e2844e0f53d1a8b765ebf4b11f94ed08ffe561d5676aad83'

data/lib/to_csv.rb

@@ -8,15 +8,23 @@ class Daru::DataFrame
 		ans = self.map(&:name).join ","
 		self.to_a[0].each do |item|
 			ans += "\n"
-			ans += item.map{|k, v| v}.join(",")
+			ans += item.map{|k, v| "\"#{v}\""}.join(",")
 		end
 		
 		return ans
 	end
 	
-	def write_csv(path, encoding: nil)
-		enc = encoding.nil? ? "" : ":#{encoding}"
-		open(path, "w#{enc}") { _1.write to_csv }
+	def write_csv(path, encoding: nil, alt: false)
+		if alt
+			# Experimental: faster algorithm
+			# ""付加を先にDaru上でやってあげる案
+			
+			#今はとりあえず
+			self.to_daru.write_csv path, encoding: encoding
+		else
+			enc = encoding.nil? ? "" : ":#{encoding}"
+			open(path, "w#{enc}") { _1.write to_csv }
+		end
 	end
 
 	# To avoid bug about adding column to Daru::DataFrame
@@ -58,6 +66,29 @@ class Daru::DataFrame
 		self.vectors = Daru::Index.new(vector_names_ary)
 		self.index = Daru::Vector.new(self.index.to_a.map{_1[0]})
 	end
+	
+	def simple_pivot(index, vectors, values, agg: :mean, index_name: nil)
+		
+		# index, vectors are Arrays. 'values' is String or Array.
+		## 文字列データなどで最初のデータだけ欲しければ agg: :first
+		piv = self.pivot_table index: index, vectors: vectors, agg: agg, values: values
+		piv.vectors = Daru::Index.new( piv.vectors.to_a.map { _1.join("-") } )
+		piv.index = Daru::Vector.new( piv.index.to_a.map { _1.join("-") } )
+		
+		# indexを新しく追加
+		index_name ||= "Pivot_Index"
+		piv[index_name] = piv.index
+		
+		# 順番変更
+		piv.order = [piv.vectors.to_a[-1]] + piv.vectors.to_a[0..-2]
+		
+		return piv
+		
+	end
+	
+	def to_rover
+		Rover::DataFrame.new(self.to_a[0])
+	end
 
 	alias_method :addvec, :add_vector
 end
@@ -69,4 +100,21 @@ class Rover::DataFrame
 		enc = encoding.nil? ? "" : ":#{encoding}"
 		open(path, "w#{enc}") {|f| f.write self.to_csv}
 	end
+	
+	def to_daru
+		Daru::DataFrame.new(self.to_a)
+	end
+	
+	def simple_pivot(index, vectors, ...)
+		ddr = self.to_daru
+		piv = ddr.simple_pivot(index, vectors, ...)
+		return piv.to_rover
+	end
+	
+	def outer_join
+		ddr = self.to_daru
+		# j = ddr.join  ## 外部結合 
+		return j.to_rover
+	end
+	
 end

data/lib/yomise/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Yomise
-  VERSION = "0.1.1"
+  VERSION = "0.1.3.1"
 end

data/lib/yomise.rb

@@ -14,23 +14,49 @@ module Yomise
 	module_function
 
 	def read(path, **opt)
-		return /csv$/ === path ? read_csv(path, **opt) : read_excel(path, **opt)
+		return /csv$/i === path ? read_csv(path, **opt) : read_excel(path, **opt)
 	end
 	
 	# ##Generate Array from CSV File, and convert it to Hash or DataFrame.
 	# **opt candidate= line_from: 1, header: 0
 	# ver. 0.3.8~ default format=:daru
-	def read_csv(path, format: :daru, encoding: "utf-8", col_sep: ",", index: nil, **opt)
+	def read_csv(path, format: :daru, encoding: "utf-8", liberal_parsing: true, col_sep: ",", index: nil, **opt)
 		## TODO.. index: option that designate column number to generate DF index.
 		## That is, revicing set_index method.
 
 		# Get 2D Array
 		begin
-			csv = CSV.parse(File.open(path, encoding: encoding, &:read), col_sep: col_sep)
+			if liberal_parsing
+				csvd = CSV.read(path, encoding: encoding, liberal_parsing: true)
+				if encoding.to_s.downcase != "utf-8" 
+					csv = csvd.to_a.map {|l| l.map {|cell| cell.nil? ? nil : cell.encode("utf-8", invalid: :replace, replace: '') }}
+				else
+					csv = csvd
+				end
+				
+				encoding = "utf-8"
+			else
+				# Old style (Not Recommended)
+				# This "&:read" is not Yomise's function(defined avobe here).. parhaps File's method.
+				csv = CSV.parse(File.open(path, encoding: encoding, &:read), col_sep: col_sep)
+			end
 		rescue
 			# Try Another Encoding
 			## puts "Fail Encoding #{encoding}. Trying cp932..."
-			csv = CSV.parse(File.open(path, encoding: "cp932", &:read), col_sep: col_sep)
+			if liberal_parsing
+				csvd = CSV.read(path, encoding: "cp932", liberal_parsing: true)
+				if encoding.to_s.downcase != "utf-8" 
+					csv = csvd.to_a.map {|l| l.map {|cell| cell.nil? ? nil : cell.encode("utf-8", invalid: :replace, replace: '') }}
+				else
+					csv = csvd
+				end
+				
+				encoding = "UTF-8"
+			else
+				# Old style (Not Recommended)
+				# This "&:read" is not Yomise's function(defined avobe here).. parhaps File's method.
+				csv = CSV.parse(File.open(path, encoding: "cp932", &:read), col_sep: col_sep)
+			end
 			encoding = "cp932"
 		end
 		
@@ -50,7 +76,7 @@ module Yomise
 			
 			# Converting Encode and Setting index.. rover not supported yet
 			if format.to_s == "daru" || format.nil?
-				ans.convert_enc!(from: encoding, to: "utf-8")
+				ans.convert_enc!(from: encoding, to: "utf-8") if encoding.to_s.downcase != "utf-8"
 				begin
 					ans.index = ind_orig if index
 				rescue

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: yomise
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.1.3.1
 platform: ruby
 authors:
 - showata
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-09-19 00:00:00.000000000 Z
+date: 2025-09-27 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: daru