fastcsv 0.0.7 → 0.0.9

data/ext/fastcsv/fastcsv.rl

@@ -1,5 +1,7 @@
 #include <ruby.h>
 #include <ruby/encoding.h>
+#include <stdbool.h>
+
 // CSV specifications.
 // http://tools.ietf.org/html/rfc4180
 // http://w3c.github.io/csvw/syntax/#ebnf
@@ -40,6 +42,7 @@ typedef struct {
 
   action open_quote {
     unclosed_line = curline;
+    in_quoted_field = true;
   }
 
   action close_quote {
@@ -58,41 +61,16 @@ typedef struct {
   }
 
   action read_quoted {
-    if (p == ts) {
-      field = rb_enc_str_new("", 0, encoding);
-      ENCODE;
-    }
-    // @note If we add an action on '""', we can skip some steps if no '""' is found.
-    else if (p > ts) {
-      // Operating on ts in-place produces odd behavior, FYI.
-      char *copy = ALLOC_N(char, p - ts);
-      memcpy(copy, ts, p - ts);
-
-      char *reader = ts, *writer = copy;
-      int escaped = 0;
-
-      while (p > reader) {
-        if (*reader == quote_char && !escaped) {
-          // Skip the escaping character.
-          escaped = 1;
-        }
-        else {
-          escaped = 0;
-          *writer++ = *reader;
-        }
-        reader++;
-      }
+    // intentionally blank - see parse_quoted_field
+  }
 
-      field = rb_enc_str_new(copy, writer - copy, encoding);
+  action new_field {
+    if (in_quoted_field) {
+      parse_quoted_field(&field, encoding, quote_char, ts + 1, p - 1);
       ENCODE;
-
-      if (copy != NULL) {
-        free(copy);
-      }
+      in_quoted_field = false;
     }
-  }
 
-  action new_field {
     rb_ary_push(row, field);
     field = Qnil;
   }
@@ -126,6 +104,12 @@ typedef struct {
       rb_ivar_set(self, s_row, rb_str_new(d->start, p - d->start));
     }
 
+    if (in_quoted_field) {
+      parse_quoted_field(&field, encoding, quote_char, ts + 1, p - 1);
+      ENCODE;
+      in_quoted_field = false;
+    }
+
     if (!NIL_P(field) || RARRAY_LEN(row)) { // same as new_field
       rb_ary_push(row, field);
       field = Qnil;
@@ -202,6 +186,38 @@ static void rb_io_ext_int_to_encs(rb_encoding *ext, rb_encoding *intern, rb_enco
   }
 }
 
+static void parse_quoted_field(VALUE* field, rb_encoding* encoding, char quote_char, char* quoted_field_start, char *quoted_field_end) {
+  // read the full quoted field, handling any escape sequences
+  if (quoted_field_end == quoted_field_start) {
+    // empty quoted field is an empty string
+    *field = rb_enc_str_new("", 0, encoding);
+  } else {
+    // largest possible buffer. if there's escaping, the resulting string will
+    // not use the entire buffer
+    char *copy = ALLOC_N(char, quoted_field_end - quoted_field_start);
+    char *reader = quoted_field_start, *writer = copy;
+    int escaped = 0;
+
+    while (quoted_field_end > reader) {
+      if (*reader == quote_char && !escaped) {
+        // Skip the escaping character.
+        escaped = 1;
+      }
+      else {
+        escaped = 0;
+        *writer++ = *reader;
+      }
+      reader++;
+    }
+
+    *field = rb_enc_str_new(copy, writer - copy, encoding);
+
+    if (copy != NULL) {
+      free(copy);
+    }
+  }
+}
+
 static VALUE raw_parse(int argc, VALUE *argv, VALUE self) {
   int cs, act, have = 0, curline = 1, io = 0;
   char *ts = 0, *te = 0, *buf = 0, *eof = 0, *mark_row_sep = 0, *row_sep = 0;
@@ -217,6 +233,8 @@ static VALUE raw_parse(int argc, VALUE *argv, VALUE self) {
   VALUE option;
   char quote_char = '"', col_sep = ',';
 
+  bool in_quoted_field = false;
+
   rb_scan_args(argc, argv, "11", &port, &opts);
   taint = OBJ_TAINTED(port);
   io = rb_respond_to(port, s_read);

data/fastcsv.gemspec

@@ -2,7 +2,7 @@
 
 Gem::Specification.new do |s|
   s.name        = "fastcsv"
-  s.version     = '0.0.7'
+  s.version     = '0.0.9'
   s.platform    = Gem::Platform::RUBY
   s.authors     = ["James McKinney"]
   s.homepage    = "https://github.com/jpmckinney/fastcsv"
@@ -14,6 +14,7 @@ Gem::Specification.new do |s|
   s.executables   = `git ls-files -- bin/*`.split("\n").map{ |f| File.basename(f) }
   s.require_paths = ["lib"]
   s.extensions    = ["ext/fastcsv/extconf.rb"]
+  s.required_ruby_version = '< 2.6'
 
   s.add_development_dependency('coveralls')
   s.add_development_dependency('rake')

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: fastcsv
 version: !ruby/object:Gem::Version
-  version: 0.0.7
+  version: 0.0.9
 platform: ruby
 authors:
 - James McKinney
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2023-01-03 00:00:00.000000000 Z
+date: 2025-06-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: coveralls
@@ -66,8 +66,8 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-description: 
-email: 
+description:
+email:
 executables: []
 extensions:
 - ext/fastcsv/extconf.rb
@@ -112,23 +112,23 @@ homepage: https://github.com/jpmckinney/fastcsv
 licenses:
 - MIT
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - ">="
+  - - "<"
     - !ruby/object:Gem::Version
-      version: '0'
+      version: '2.6'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.3.1
-signing_key: 
+rubygems_version: 3.5.21
+signing_key:
 specification_version: 4
 summary: A fast Ragel-based CSV parser, compatible with Ruby's CSV
 test_files: