fastcsv 0.0.5 → 0.0.7

data/ext/fastcsv/fastcsv.rl

@@ -104,7 +104,7 @@ typedef struct {
       if (p - mark_row_sep != len_row_sep || row_sep[0] != *mark_row_sep || (len_row_sep == 2 && row_sep[1] != *(mark_row_sep + 1))) {
         FREE;
 
-        rb_raise(eError, "Unquoted fields do not allow \\r or \\n (line %d).", curline - 1);
+        rb_raise(eError, "Unquoted fields do not allow \\r or \\n (line %d).", curline);
       }
     }
     else {
@@ -112,11 +112,12 @@ typedef struct {
       row_sep = ALLOC_N(char, len_row_sep);
       memcpy(row_sep, mark_row_sep, len_row_sep);
     }
+
+    curline++;
   }
 
   action new_row {
     mark_row_sep = p;
-    curline++;
 
     if (d->start == 0 || p == d->start) {
       rb_ivar_set(self, s_row, rb_str_new2(""));
@@ -154,12 +155,14 @@ typedef struct {
   EOF = 0;
   quote_char = any when { fc == quote_char };
   col_sep = any when { fc == col_sep } >new_field;
-  row_sep = ('\r' '\n'? | '\n');
-  unquoted = (any* -- quote_char -- col_sep -- row_sep - EOF) %read_unquoted;
-  quoted = quote_char >open_quote (any - quote_char - EOF | quote_char quote_char | row_sep)* %read_quoted quote_char >close_quote;
+  row_sep = '\r' '\n'? | '\n';
+  unquoted = (any* -- quote_char -- col_sep -- row_sep -- '\r' -- '\n' - EOF) %read_unquoted;
+  quoted = quote_char >open_quote (any - quote_char - EOF | quote_char quote_char)* %read_quoted quote_char >close_quote;
   field = unquoted | quoted;
 
   # @see Ragel Guide: 6.3 Scanners
+  # > Entering
+  # % Leaving
   # An unquoted field can be zero-length.
   main := |*
     field col_sep;
@@ -250,7 +253,7 @@ static VALUE raw_parse(int argc, VALUE *argv, VALUE self) {
     rb_raise(rb_eArgError, ":col_sep has to be a single character String");
   }
 
-  // @see rb_io_extract_modeenc
+  // @see rb_io_extract_modeenc parse_mode_enc
   /* Set to defaults */
   rb_io_ext_int_to_encs(NULL, NULL, &enc, &enc2, 0);
 
@@ -466,6 +469,13 @@ static VALUE raw_parse(int argc, VALUE *argv, VALUE self) {
     else if (io) {
       have = pe - ts;
       memmove(buf, ts, have);
+      // @see https://github.com/hpricot/hpricot/blob/master/ext/hpricot_scan/hpricot_scan.rl#L92
+      if (d->start > ts) {
+        d->start = buf + (d->start - ts);
+      }
+      if (mark_row_sep >= ts) {
+        mark_row_sep = buf + (mark_row_sep - ts);
+      }
       te = buf + (te - ts);
       ts = buf;
     }

data/fastcsv.gemspec

@@ -2,7 +2,7 @@
 
 Gem::Specification.new do |s|
   s.name        = "fastcsv"
-  s.version     = '0.0.5'
+  s.version     = '0.0.7'
   s.platform    = Gem::Platform::RUBY
   s.authors     = ["James McKinney"]
   s.homepage    = "https://github.com/jpmckinney/fastcsv"
@@ -16,8 +16,8 @@ Gem::Specification.new do |s|
   s.extensions    = ["ext/fastcsv/extconf.rb"]
 
   s.add_development_dependency('coveralls')
-  s.add_development_dependency('json', '~> 1.8') # to silence coveralls warning
   s.add_development_dependency('rake')
-  s.add_development_dependency('rake-compiler')
   s.add_development_dependency('rspec', '~> 3.1')
+
+  s.add_development_dependency('rake-compiler')
 end

data/lib/fastcsv.rb

@@ -123,7 +123,7 @@ private
           encoding = enc
         end
       end
-      parser.raw_parse(@io, encoding: encoding, quote_char: quote_char, col_sep: col_sep) do |row|
+      parser.raw_parse(@io, encoding: encoding, quote_char: quote_char, col_sep: col_sep, row_sep: row_sep) do |row|
         Fiber.yield(row)
       end
     end

data/spec/fastcsv_spec.rb

@@ -136,7 +136,14 @@ RSpec.shared_examples 'a CSV parser' do
     end
   end
 
-  it "should parse an encoded string" do
+  it 'should raise the row number' do
+    csv = %("\n\n\n"\n"x)
+    csv_error = 'Unclosed quoted field on line 2.'
+    expect{CSV.parse(csv)}.to raise_error(CSV::MalformedCSVError, csv_error)
+    expect{parse(csv)}.to raise_error(FastCSV::MalformedCSVError, csv_error)
+  end
+
+  it 'should parse an encoded string' do
     csv = "ß"
     actual = parse(csv)
     expected = CSV.parse(csv)
@@ -181,6 +188,14 @@ RSpec.shared_examples 'a CSV parser' do
     it 'should allow zero' do
       expect(parse_with_buffer_size(simple, 0)).to eq(CSV.parse(simple))
     end
+
+    it 'should parse a consecutive field-sep and row-sep at a buffer boundary' do
+      # This is a further minimized reproduction of
+      # https://github.com/jpmckinney/fastcsv/issues/14
+      csv = "A0,B0\nA1,B1,\nA2,B2,\n"
+      buffer_size = 4
+      expect(parse_with_buffer_size(csv, buffer_size)).to eq(CSV.parse(csv))
+    end
   end
 end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: fastcsv
 version: !ruby/object:Gem::Version
-  version: 0.0.5
+  version: 0.0.7
 platform: ruby
 authors:
 - James McKinney
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-02-05 00:00:00.000000000 Z
+date: 2023-01-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: coveralls
@@ -24,20 +24,6 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-- !ruby/object:Gem::Dependency
-  name: json
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.8'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '1.8'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -53,33 +39,33 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: rake-compiler
+  name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '3.1'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '0'
+        version: '3.1'
 - !ruby/object:Gem::Dependency
-  name: rspec
+  name: rake-compiler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '3.1'
+        version: '0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '3.1'
+        version: '0'
 description: 
 email: 
 executables: []
@@ -90,12 +76,12 @@ files:
 - ".gitignore"
 - ".rspec"
 - ".travis.yml"
+- ".yardopts"
 - Gemfile
 - LICENSE
 - README.md
 - Rakefile
 - TESTS.md
-- USAGE
 - ext/fastcsv/extconf.rb
 - ext/fastcsv/fastcsv.c
 - ext/fastcsv/fastcsv.rl
@@ -141,8 +127,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.2.0
+rubygems_version: 3.0.3.1
 signing_key: 
 specification_version: 4
 summary: A fast Ragel-based CSV parser, compatible with Ruby's CSV

data/USAGE

@@ -1 +0,0 @@
-See README.md for full usage details.