hippie_csv 0.0.6 → 0.0.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: f53a562b556d85f2c893855cf0aa34ea6126afab
-  data.tar.gz: 9c1a2954a38b2c01b1a9dcbc13c25fb21e9799be
+  metadata.gz: 0c3685fbff376b677274444f9b9211a80165cabd
+  data.tar.gz: 36322262f11c8ff9e28e90fa17ebd9be07cc9bad
 SHA512:
-  metadata.gz: 7db7d04670863b9bffc7c66d8c1410b57892b4729819c6755816bd3e5deffcf0acc2af75870d6e26b41dc5643f1ef7eb8ec9cce86aa9868fff49129c2872e204
-  data.tar.gz: 2a587740e0ee24c021a373c71e4c3038447de70595d7ff1e89d86dc51aa696083a238cef221bdb2e6970a53eca11d407b6a0c5ee11df1449e32305ce25113055
+  metadata.gz: 8d050fb4b1d7836f1167bacae1884126c92797026e1ac9177b1df316dcf146ffb34d46fc062e1a0bf5440cc9c6863fb750bc14ea76a08969a9969a90b70840b9
+  data.tar.gz: 24bcf0eb9c671a7116a311b74e3b381ae7557067f1e5d9174c835ac58344b5f4b0e8eaa263443d2ffd0f7b73ba7474fcb21e7c3be2b948a7d4add2507da4b14b

data/lib/hippie_csv/constants.rb

@@ -1,7 +1,12 @@
 require "csv"
 
 module HippieCSV
-  QUOTE_CHARACTERS = %w(" ' |).freeze
+  QUOTE_CHARACTERS = %w(" ' | ∑ ⦿ ◉).freeze
+  # The latter three characters are not expected to intentionally used as
+  # quotes. Rather, when usual quote characters are badly misused, we want
+  # to fall back to a character _unlikely_ to be in the file, such that
+  # we can at least parse.
+
   DELIMETERS = %W(, ; \t).freeze
   ENCODING = "utf-8".freeze
   ALTERNATE_ENCODING = "utf-16".freeze

data/lib/hippie_csv/support.rb

@@ -27,15 +27,17 @@ module HippieCSV
 
       def maybe_parse(string)
         QUOTE_CHARACTERS.find do |quote_character|
-          Support.rescuing_malformed do
-            return Support.parse_csv(string.strip, quote_character)
+          [string, tolerate_escaping(string, quote_character)].find do |string_to_parse|
+            rescuing_malformed do
+              return parse_csv(string_to_parse.strip, quote_character)
+            end
           end
         end
       end
 
       def parse_csv(string, quote_character)
         CSV.parse(
-          tolerate_escaping(string, quote_character),
+          string,
           quote_char: quote_character,
           col_sep: guess_delimeter(string, quote_character)
         )

data/lib/hippie_csv/version.rb

@@ -1,3 +1,3 @@
 module HippieCSV
-  VERSION = "0.0.6"
+  VERSION = "0.0.7"
 end

data/spec/fixtures/bad_quoting.csv

@@ -0,0 +1,8 @@
+id,full_name,email,support,admin_of,user_payment_count,user_payment_total,click_count,registration_date,user_status,email_validation
+1,Tom Stonew,tom@example.com,Stone's Soccer Club,Stone's Soccer Club,5,£1.16,96,3/31/2015 17:28:39,active,Y
+262,Mark J Collins,info@example.co,|Run N Hide Things Get Cancelled!|,Run N Hide | Things Get Cancelled!,0,NULL,0,7/2/2015 18:41:05,active,N
+474,Francis Daly,francisdaly@live.com,"\A Junction\"" Project Association""",NULL,0,NULL,1,8/13/2015 14:50:05,active,N
+640,mitchell miller,test@example.co.uk,Berkely ⦿ Club,NULL,0,NULL,1,9/9/2015 10:35:00,active,N
+646,Ash ,hi@example.com,"\NSAAA A" """,NULL,0,NULL,0,9/9/2015 14:21:21,active,N
+649,Ken Kenneth,ken@example.com,Lusk ∑ Club,Lusk ∑ Club,0,NULL,1,9/9/2015 21:45:24,active,Y
+650,Bam Gouldstone,bam@example.com,$ Shave Club,NULL,0,NULL,0,9/10/2015 6:43:07,active,N

data/spec/hippie_csv/constants_spec.rb

@@ -5,8 +5,8 @@ describe HippieCSV do
   describe "::QUOTE_CHARACTERS" do
     let(:supported_quote_characters) { ['"', "'", "|"] }
 
-    it "defines supported quote characters" do
-      expect(HippieCSV::QUOTE_CHARACTERS).to match(supported_quote_characters)
+    it "supports common quote characters" do
+      expect(HippieCSV::QUOTE_CHARACTERS).to include(*supported_quote_characters)
     end
 
     it "is not modifiable" do

data/spec/hippie_csv/support_spec.rb

@@ -68,7 +68,6 @@ describe HippieCSV::Support do
 
     it "works" do
       expect(subject).to receive(:guess_delimeter).with(string, quote_character).and_return(',')
-      expect(subject).to receive(:tolerate_escaping).with(string, quote_character).and_return(string)
 
       expect(subject.parse_csv(string, quote_character)).to eq(
         [["name", "email"], ["stephen", "test@example.com"]]

data/spec/hippie_csv/version_spec.rb

@@ -3,7 +3,7 @@ require "spec_helper"
 describe HippieCSV do
 
   it "defines a version" do
-    expect(HippieCSV::VERSION).to eq("0.0.6")
+    expect(HippieCSV::VERSION).to eq("0.0.7")
   end
 
 end

data/spec/hippie_csv_spec.rb

@@ -121,6 +121,17 @@ describe HippieCSV do
       expect(Time.now).to be_within(5).of(start_time)
     end
 
+    it "works when many invalid quote types contained" do
+      path = fixture_path(:bad_quoting)
+
+      expect { CSV.read(path) }.to raise_error
+      expect {
+        import = subject.read(path)
+        expect(import.map(&:count).uniq).to eq([11])
+        expect(import.count).to eq(8)
+      }.not_to raise_error
+    end
+
     it "strips leading/trailing blank lines" do
       path = fixture_path(:trailing_leading_lines)
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: hippie_csv
 version: !ruby/object:Gem::Version
-  version: 0.0.6
+  version: 0.0.7
 platform: ruby
 authors:
 - Stephen O'Brien
@@ -100,6 +100,7 @@ files:
 - lib/hippie_csv/support.rb
 - lib/hippie_csv/version.rb
 - spec/fixtures/accents_semicolon_windows_1252.csv
+- spec/fixtures/bad_quoting.csv
 - spec/fixtures/dos_line_ending.csv
 - spec/fixtures/encoding.csv
 - spec/fixtures/escaped_quotes.csv
@@ -141,6 +142,7 @@ specification_version: 4
 summary: Tolerant, liberal CSV parsing
 test_files:
 - spec/fixtures/accents_semicolon_windows_1252.csv
+- spec/fixtures/bad_quoting.csv
 - spec/fixtures/dos_line_ending.csv
 - spec/fixtures/encoding.csv
 - spec/fixtures/escaped_quotes.csv