smarter_csv 1.16.6 → 1.17.0.pre5

data/ext/smarter_csv/Makefile

@@ -0,0 +1,270 @@
+
+SHELL = /bin/sh
+
+# V=0 quiet, V=1 verbose.  other values don't work.
+V = 0
+V0 = $(V:0=)
+Q1 = $(V:1=)
+Q = $(Q1:0=@)
+ECHO1 = $(V:1=@ :)
+ECHO = $(ECHO1:0=@ echo)
+NULLCMD = :
+
+#### Start of system configuration section. ####
+
+srcdir = .
+topdir = /Users/tilo/.rvm/rubies/ruby-3.2.2/include/ruby-3.2.0
+hdrdir = $(topdir)
+arch_hdrdir = /Users/tilo/.rvm/rubies/ruby-3.2.2/include/ruby-3.2.0/arm64-darwin23
+PATH_SEPARATOR = :
+VPATH = $(srcdir):$(arch_hdrdir)/ruby:$(hdrdir)/ruby
+prefix = $(DESTDIR)/Users/tilo/.rvm/rubies/ruby-3.2.2
+rubysitearchprefix = $(rubylibprefix)/$(sitearch)
+rubyarchprefix = $(rubylibprefix)/$(arch)
+rubylibprefix = $(libdir)/$(RUBY_BASE_NAME)
+exec_prefix = $(prefix)
+vendorarchhdrdir = $(vendorhdrdir)/$(sitearch)
+sitearchhdrdir = $(sitehdrdir)/$(sitearch)
+rubyarchhdrdir = $(rubyhdrdir)/$(arch)
+vendorhdrdir = $(rubyhdrdir)/vendor_ruby
+sitehdrdir = $(rubyhdrdir)/site_ruby
+rubyhdrdir = $(includedir)/$(RUBY_VERSION_NAME)
+vendorarchdir = $(vendorlibdir)/$(sitearch)
+vendorlibdir = $(vendordir)/$(ruby_version)
+vendordir = $(rubylibprefix)/vendor_ruby
+sitearchdir = $(sitelibdir)/$(sitearch)
+sitelibdir = $(sitedir)/$(ruby_version)
+sitedir = $(rubylibprefix)/site_ruby
+rubyarchdir = $(rubylibdir)/$(arch)
+rubylibdir = $(rubylibprefix)/$(ruby_version)
+sitearchincludedir = $(includedir)/$(sitearch)
+archincludedir = $(includedir)/$(arch)
+sitearchlibdir = $(libdir)/$(sitearch)
+archlibdir = $(libdir)/$(arch)
+ridir = $(datarootdir)/$(RI_BASE_NAME)
+mandir = $(datarootdir)/man
+localedir = $(datarootdir)/locale
+libdir = $(exec_prefix)/lib
+psdir = $(docdir)
+pdfdir = $(docdir)
+dvidir = $(docdir)
+htmldir = $(docdir)
+infodir = $(datarootdir)/info
+docdir = $(datarootdir)/doc/$(PACKAGE)
+oldincludedir = $(DESTDIR)/usr/include
+includedir = $(SDKROOT)$(prefix)/include
+runstatedir = $(localstatedir)/run
+localstatedir = $(prefix)/var
+sharedstatedir = $(prefix)/com
+sysconfdir = $(prefix)/etc
+datadir = $(datarootdir)
+datarootdir = $(prefix)/share
+libexecdir = $(exec_prefix)/libexec
+sbindir = $(exec_prefix)/sbin
+bindir = $(exec_prefix)/bin
+archdir = $(rubyarchdir)
+
+
+CC_WRAPPER = 
+CC = gcc
+CXX = g++
+LIBRUBY = $(LIBRUBY_SO)
+LIBRUBY_A = lib$(RUBY_SO_NAME)-static.a
+LIBRUBYARG_SHARED = -l$(RUBY_SO_NAME)
+LIBRUBYARG_STATIC = -l$(RUBY_SO_NAME)-static -framework CoreFoundation $(MAINLIBS)
+empty =
+OUTFLAG = -o $(empty)
+COUTFLAG = -o $(empty)
+CSRCFLAG = $(empty)
+
+RUBY_EXTCONF_H = 
+cflags   = -fdeclspec $(optflags) $(debugflags) $(warnflags)
+cxxflags = 
+optflags = -O3
+debugflags = -ggdb3
+warnflags = -Wall -Wextra -Wextra-tokens -Wdeprecated-declarations -Wdivision-by-zero -Wdiv-by-zero -Wimplicit-function-declaration -Wimplicit-int -Wmisleading-indentation -Wpointer-arith -Wshorten-64-to-32 -Wwrite-strings -Wold-style-definition -Wmissing-noreturn -Wno-cast-function-type -Wno-constant-logical-operand -Wno-long-long -Wno-missing-field-initializers -Wno-overlength-strings -Wno-parentheses-equality -Wno-self-assign -Wno-tautological-compare -Wno-unused-parameter -Wno-unused-value -Wunused-variable -Wundef
+cppflags = 
+CCDLFLAGS = -fno-common
+CFLAGS   = $(CCDLFLAGS) -O3 -I/opt/homebrew/opt/libyaml/include -I/opt/homebrew/opt/libksba/include -I/opt/homebrew/opt/readline/include -I/opt/homebrew/opt/zlib/include -I/opt/homebrew/opt/openssl@1.1/include $(cflags) -fno-common -pipe $(ARCH_FLAG)
+INCFLAGS = -I. -I$(arch_hdrdir) -I$(hdrdir)/ruby/backward -I$(hdrdir) -I$(srcdir)
+DEFS     = 
+CPPFLAGS =  -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -D_DARWIN_UNLIMITED_SELECT -D_REENTRANT $(DEFS) $(cppflags)
+CXXFLAGS = $(CCDLFLAGS) -fdeclspec $(ARCH_FLAG)
+ldflags  = -L. -L/opt/homebrew/opt/libyaml/lib -L/opt/homebrew/opt/libksba/lib -L/opt/homebrew/opt/readline/lib -L/opt/homebrew/opt/zlib/lib -L/opt/homebrew/opt/openssl@1.1/lib -fstack-protector-strong
+dldflags = -L/opt/homebrew/opt/libyaml/lib -L/opt/homebrew/opt/libksba/lib -L/opt/homebrew/opt/readline/lib -L/opt/homebrew/opt/zlib/lib -L/opt/homebrew/opt/openssl@1.1/lib -Wl,-undefined,dynamic_lookup $(LIBRUBYARG_SHARED)
+ARCH_FLAG = 
+DLDFLAGS = $(ldflags) $(dldflags) $(ARCH_FLAG)
+LDSHARED = $(CC) -dynamic -bundle
+LDSHAREDXX = $(CXX) -dynamic -bundle
+AR = ar
+EXEEXT = 
+
+RUBY_INSTALL_NAME = $(RUBY_BASE_NAME)
+RUBY_SO_NAME = ruby.3.2
+RUBYW_INSTALL_NAME = 
+RUBY_VERSION_NAME = $(RUBY_BASE_NAME)-$(ruby_version)
+RUBYW_BASE_NAME = rubyw
+RUBY_BASE_NAME = ruby
+
+arch = arm64-darwin23
+sitearch = $(arch)
+ruby_version = 3.2.0
+ruby = $(bindir)/$(RUBY_BASE_NAME)
+RUBY = $(ruby)
+BUILTRUBY = $(bindir)/$(RUBY_BASE_NAME)
+ruby_headers = $(hdrdir)/ruby.h $(hdrdir)/ruby/backward.h $(hdrdir)/ruby/ruby.h $(hdrdir)/ruby/defines.h $(hdrdir)/ruby/missing.h $(hdrdir)/ruby/intern.h $(hdrdir)/ruby/st.h $(hdrdir)/ruby/subst.h $(arch_hdrdir)/ruby/config.h
+
+RM = rm -f
+RM_RF = rm -fr
+RMDIRS = rmdir -p
+MAKEDIRS = /opt/homebrew/opt/coreutils/bin/gmkdir -p
+INSTALL = /opt/homebrew/opt/coreutils/bin/ginstall -c
+INSTALL_PROG = $(INSTALL) -m 0755
+INSTALL_DATA = $(INSTALL) -m 644
+COPY = cp
+TOUCH = exit >
+
+#### End of system configuration section. ####
+
+preload = 
+libpath = . $(libdir)
+LIBPATH =  -L. -L$(libdir)
+DEFFILE = 
+
+CLEANFILES = mkmf.log
+DISTCLEANFILES = 
+DISTCLEANDIRS = 
+
+extout = 
+extout_prefix = 
+target_prefix = /smarter_csv
+LOCAL_LIBS = 
+LIBS = $(LIBRUBYARG_SHARED)  -lpthread  
+ORIG_SRCS = smarter_csv.c
+SRCS = $(ORIG_SRCS) 
+OBJS = smarter_csv.o
+HDRS = 
+LOCAL_HDRS = 
+TARGET = smarter_csv
+TARGET_NAME = smarter_csv
+TARGET_ENTRY = Init_$(TARGET_NAME)
+DLLIB = $(TARGET).bundle
+EXTSTATIC = 
+STATIC_LIB = 
+
+TIMESTAMP_DIR = .
+BINDIR        = $(bindir)
+RUBYCOMMONDIR = $(sitedir)$(target_prefix)
+RUBYLIBDIR    = $(sitelibdir)$(target_prefix)
+RUBYARCHDIR   = $(sitearchdir)$(target_prefix)
+HDRDIR        = $(sitehdrdir)$(target_prefix)
+ARCHHDRDIR    = $(sitearchhdrdir)$(target_prefix)
+TARGET_SO_DIR =
+TARGET_SO     = $(TARGET_SO_DIR)$(DLLIB)
+CLEANLIBS     = $(TARGET_SO) $(TARGET_SO).dSYM
+CLEANOBJS     = $(OBJS) *.bak
+TARGET_SO_DIR_TIMESTAMP = $(TIMESTAMP_DIR)/.sitearchdir.-.smarter_csv.time
+
+all:    $(DLLIB)
+static: $(STATIC_LIB)
+.PHONY: all install static install-so install-rb
+.PHONY: clean clean-so clean-static clean-rb
+
+clean-static::
+clean-rb-default::
+clean-rb::
+clean-so::
+clean: clean-so clean-static clean-rb-default clean-rb
+		-$(Q)$(RM_RF) $(CLEANLIBS) $(CLEANOBJS) $(CLEANFILES) .*.time
+
+distclean-rb-default::
+distclean-rb::
+distclean-so::
+distclean-static::
+distclean: clean distclean-so distclean-static distclean-rb-default distclean-rb
+		-$(Q)$(RM) Makefile $(RUBY_EXTCONF_H) conftest.* mkmf.log
+		-$(Q)$(RM) core ruby$(EXEEXT) *~ $(DISTCLEANFILES)
+		-$(Q)$(RMDIRS) $(DISTCLEANDIRS) 2> /dev/null || true
+
+realclean: distclean
+install: install-so install-rb
+
+install-so: $(DLLIB) $(TARGET_SO_DIR_TIMESTAMP)
+	$(INSTALL_PROG) $(DLLIB) $(RUBYARCHDIR)
+clean-static::
+	-$(Q)$(RM) $(STATIC_LIB)
+install-rb: pre-install-rb do-install-rb install-rb-default
+install-rb-default: pre-install-rb-default do-install-rb-default
+pre-install-rb: Makefile
+pre-install-rb-default: Makefile
+do-install-rb:
+do-install-rb-default:
+pre-install-rb-default:
+	@$(NULLCMD)
+$(TARGET_SO_DIR_TIMESTAMP):
+	$(Q) $(MAKEDIRS) $(@D) $(RUBYARCHDIR)
+	$(Q) $(TOUCH) $@
+
+site-install: site-install-so site-install-rb
+site-install-so: install-so
+site-install-rb: install-rb
+
+.SUFFIXES: .c .m .cc .mm .cxx .cpp .o .S
+
+.cc.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.cc.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+.mm.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.mm.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+.cxx.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.cxx.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+.cpp.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.cpp.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CXX) $(INCFLAGS) $(CPPFLAGS) $(CXXFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+.c.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CC) $(INCFLAGS) $(CPPFLAGS) $(CFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.c.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CC) $(INCFLAGS) $(CPPFLAGS) $(CFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+.m.o:
+	$(ECHO) compiling $(<)
+	$(Q) $(CC) $(INCFLAGS) $(CPPFLAGS) $(CFLAGS) $(COUTFLAG)$@ -c $(CSRCFLAG)$<
+
+.m.S:
+	$(ECHO) translating $(<)
+	$(Q) $(CC) $(INCFLAGS) $(CPPFLAGS) $(CFLAGS) $(COUTFLAG)$@ -S $(CSRCFLAG)$<
+
+$(TARGET_SO): $(OBJS) Makefile
+	$(ECHO) linking shared-object smarter_csv/$(DLLIB)
+	-$(Q)$(RM) $(@)
+	$(Q) $(LDSHARED) -o $@ $(OBJS) $(LIBPATH) $(DLDFLAGS) $(LOCAL_LIBS) $(LIBS)
+	$(Q) $(POSTLINK)
+
+
+
+$(OBJS): $(HDRS) $(ruby_headers)

data/ext/smarter_csv/extconf.rb

@@ -3,15 +3,6 @@
 require 'mkmf'
 require "rbconfig"
 
-# On non-MRI Rubies (JRuby, TruffleRuby, ...) there is no C extension to build, and trying to build
-# it breaks `gem install` for anything that depends on smarter_csv. Write a no-op Makefile so install
-# succeeds, then stop. At runtime SmarterCSV falls back to its pure-Ruby parser (it checks whether the
-# C functions actually loaded via respond_to?(:parse_csv_line_c)).
-if RUBY_ENGINE != 'ruby'
-  File.write('Makefile', dummy_makefile($srcdir).join)
-  exit 0
-end
-
 if RbConfig::MAKEFILE_CONFIG["CFLAGS"].include?("-g -O3")
   fixed_CFLAGS = RbConfig::MAKEFILE_CONFIG["CFLAGS"].sub("-g -O3", "-O3 $(cflags)")
   puts("Fix CFLAGS: #{RbConfig::MAKEFILE_CONFIG["CFLAGS"]} -> #{fixed_CFLAGS}")

data/ext/smarter_csv/smarter_csv.c

@@ -304,40 +304,25 @@ static VALUE rb_parse_csv_line(VALUE self, VALUE line, VALUE col_sep, VALUE quot
           if (!allow_escaped_quotes || backslash_count % 2 == 0) {
             if (__builtin_expect(quote_boundary_standard, 1)) {
               if (in_quotes) {
-                if (p + 2 < endP && *(p + 1) == quote_char_val) {
-                  /* RFC doubled quote inside a quoted field ("" → ").
-                   * Give this precedence over the closing-quote check, but only
-                   * when another byte follows the doubled pair.
-                   *
-                   * Compatibility note: we intentionally do NOT force terminal
-                   * "" to be consumed here. SmarterCSV has a long-standing lenient
-                   * behavior for malformed tails like ...\"" in :double_quotes mode:
-                   * the final quote may still close the field instead of turning the
-                   * row into an unclosed-quote error. Issue #334 needs doubled-quote
-                   * precedence for ..."",... (more content follows), but we keep the
-                   * historical leniency for terminal ..."". */
-                  p++;
-                } else {
-                  // closing quote: only valid if followed by col_sep, row_sep, or end of line
-                  bool valid_close = (p + 1 >= endP);
-                  if (!valid_close) {
-                    valid_close = true;
-                    for (long j = 0; j < col_sep_len; j++) {
-                      if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
-                    }
-                  }
-                  if (!valid_close && row_sep_len > 0) {
-                    valid_close = true;
-                    for (long j = 0; j < row_sep_len; j++) {
-                      if (*(p + 1 + j) != *(row_sepP + j)) { valid_close = false; break; }
-                    }
+                // closing quote: only valid if followed by col_sep, row_sep, or end of line
+                bool valid_close = (p + 1 >= endP);
+                if (!valid_close) {
+                  valid_close = true;
+                  for (long j = 0; j < col_sep_len; j++) {
+                    if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
                   }
-                  if (valid_close) {
-                    in_quotes = false;
-                    field_started = true;
+                }
+                if (!valid_close && row_sep_len > 0) {
+                  valid_close = true;
+                  for (long j = 0; j < row_sep_len; j++) {
+                    if (*(p + 1 + j) != *(row_sepP + j)) { valid_close = false; break; }
                   }
-                  // else: quote inside quoted field → literal
                 }
+                if (valid_close) {
+                  in_quotes = false;
+                  field_started = true;
+                }
+                // else: quote inside quoted field → literal (handles "" doubling)
               } else if (!field_started) {
                 in_quotes = true;     // opening quote at field boundary
                 field_started = true;
@@ -778,11 +763,6 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash(VALUE self, VALUE line,
    * the frame stays well below 4 KB and ___chkstk_darwin never fires on ARM64 macOS.
    */
   bool *keep_bitmap = NULL;
-  /* In THIS (non-ctx) function the bitmap is alloca'd to headers_len on every call (see the alloca
-   * sites below), so keep_bitmap[] is exactly headers_len long and headers_len is the correct bound
-   * at all access sites. Do NOT mirror rb_parse_line_to_hash_ctx's keep_bitmap_len here: that variant
-   * caches its bitmap across rows (where @headers can grow), so it must use the captured length; this
-   * one rebuilds per call and does not. */
   bool keep_extra_columns = true; /* extra cols (> headers_len): keep by default */
   bool has_only = false;          /* true when only_headers: filtering is active */
   long early_exit_after = -1;     /* column index after which we stop; -1 = no early exit */
@@ -1101,40 +1081,25 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash(VALUE self, VALUE line,
             if (!allow_escaped_quotes || backslash_count % 2 == 0) {
               if (__builtin_expect(quote_boundary_standard, 1)) {
                 if (in_quotes) {
-                  if (p + 2 < endP && *(p + 1) == quote_char_val) {
-                    /* RFC doubled quote inside a quoted field ("" → ").
-                     * Give this precedence over the closing-quote check, but only
-                     * when another byte follows the doubled pair.
-                     *
-                     * Compatibility note: we intentionally do NOT force terminal
-                     * "" to be consumed here. SmarterCSV has a long-standing lenient
-                     * behavior for malformed tails like ...\"" in :double_quotes mode:
-                     * the final quote may still close the field instead of turning the
-                     * row into an unclosed-quote error. Issue #334 needs doubled-quote
-                     * precedence for ..."",... (more content follows), but we keep the
-                     * historical leniency for terminal ..."". */
-                    p++;
-                  } else {
-                    // closing quote: only valid if followed by col_sep, row_sep, or end of line
-                    bool valid_close = (p + 1 >= endP);
-                    if (!valid_close) {
-                      valid_close = true;
-                      for (long j = 0; j < col_sep_len; j++) {
-                        if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
-                      }
-                    }
-                    if (!valid_close && row_sep_len2 > 0) {
-                      valid_close = true;
-                      for (long j = 0; j < row_sep_len2; j++) {
-                        if (*(p + 1 + j) != *(row_sepP2 + j)) { valid_close = false; break; }
-                      }
+                  // closing quote: only valid if followed by col_sep, row_sep, or end of line
+                  bool valid_close = (p + 1 >= endP);
+                  if (!valid_close) {
+                    valid_close = true;
+                    for (long j = 0; j < col_sep_len; j++) {
+                      if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
                     }
-                    if (valid_close) {
-                      in_quotes = false;
-                      field_started = true;
+                  }
+                  if (!valid_close && row_sep_len2 > 0) {
+                    valid_close = true;
+                    for (long j = 0; j < row_sep_len2; j++) {
+                      if (*(p + 1 + j) != *(row_sepP2 + j)) { valid_close = false; break; }
                     }
-                    // else: quote inside quoted field → literal
                   }
+                  if (valid_close) {
+                    in_quotes = false;
+                    field_started = true;
+                  }
+                  // else: quote inside quoted field → literal (handles "" doubling)
                 } else if (!field_started) {
                   in_quotes = true;     // opening quote at field boundary
                   field_started = true;
@@ -1211,20 +1176,12 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash(VALUE self, VALUE line,
    * return nil instead of the hash so the row can be skipped.
    * With lazy allocation, if all_blank is true, xform.hash is still Qnil —
    * no hash was ever allocated.
-   *
-   * If remove_empty_hashes is disabled, preserve the row as an empty hash.
-   * This keeps parity with the Ruby path without adding any cost to the
-   * normal non-blank hot path.
    */
-  if (all_blank) {
-    if (remove_empty) {
-      VALUE result = rb_ary_new_capa(2);
-      rb_ary_push(result, Qnil);
-      rb_ary_push(result, LONG2FIX(element_count));
-      return result;
-    }
-
-    ensure_hash_allocated(&xform);
+  if (remove_empty && all_blank) {
+    VALUE result = rb_ary_new_capa(2);
+    rb_ary_push(result, Qnil);
+    rb_ary_push(result, LONG2FIX(element_count));
+    return result;
   }
 
   /* ----------------------------------------
@@ -1464,14 +1421,6 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
   int  numeric_mode        = ctx->numeric_mode;
   VALUE numeric_keys       = ctx->numeric_keys;
   bool *keep_bitmap         = ctx->keep_bitmap;
-  /* keep_bitmap is cached in the context (xmalloc'd once at construction, sized to the header count
-   * THEN). @headers can grow in place as undeclared extra columns appear, so the live headers_len
-   * (re-read each call below) may exceed the bitmap's length. Every keep_bitmap[] access in this
-   * function MUST be bounded by keep_bitmap_len, never headers_len — indices past the bitmap are
-   * extra columns and follow keep_extra_columns. Bounding by the grown headers_len was an
-   * out-of-bounds heap read (the bug). The sibling rb_parse_line_to_hash safely uses headers_len
-   * because it re-allocs its bitmap to headers_len on every call. */
-  long  keep_bitmap_len     = ctx->keep_bitmap_len;
   bool  keep_extra_columns  = ctx->keep_extra_columns;
   long  early_exit_after    = ctx->early_exit_after;
 
@@ -1573,7 +1522,7 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
           while (trim_end >= trim_start && (*trim_end == ' ' || *trim_end == '\t')) trim_end--;
         }
         long trimmed_len = (trim_end >= trim_start) ? (trim_end - trim_start + 1) : 0;
-        if (!keep_bitmap || (element_count < keep_bitmap_len ? keep_bitmap[element_count] : keep_extra_columns)) {
+        if (!keep_bitmap || (element_count < headers_len ? keep_bitmap[element_count] : keep_extra_columns)) {
           if (insert_field_into_hash(&xform, trim_start, trimmed_len, element_count, false, quote_char_val, encoding))
             all_blank = false;
         }
@@ -1594,7 +1543,7 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
           while (trim_end >= trim_start && (*trim_end == ' ' || *trim_end == '\t')) trim_end--;
         }
         long trimmed_len = (trim_end >= trim_start) ? (trim_end - trim_start + 1) : 0;
-        if (!keep_bitmap || (element_count < keep_bitmap_len ? keep_bitmap[element_count] : keep_extra_columns)) {
+        if (!keep_bitmap || (element_count < headers_len ? keep_bitmap[element_count] : keep_extra_columns)) {
           if (insert_field_into_hash(&xform, trim_start, trimmed_len, element_count, false, quote_char_val, encoding))
             all_blank = false;
         }
@@ -1657,7 +1606,7 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
 
         bool has_embedded_quotes = quoted || (trimmed_len > 0 && memchr(trim_start, quote_char_val, trimmed_len));
 
-        if (!keep_bitmap || (element_count < keep_bitmap_len ? keep_bitmap[element_count] : keep_extra_columns)) {
+        if (!keep_bitmap || (element_count < headers_len ? keep_bitmap[element_count] : keep_extra_columns)) {
           if (insert_field_into_hash(&xform, trim_start, trimmed_len, element_count, has_embedded_quotes, quote_char_val, encoding))
             all_blank = false;
         }
@@ -1691,40 +1640,25 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
             if (!allow_escaped_quotes || backslash_count % 2 == 0) {
               if (__builtin_expect(quote_boundary_standard, 1)) {
                 if (in_quotes) {
-                  if (p + 2 < endP && *(p + 1) == quote_char_val) {
-                    /* RFC doubled quote inside a quoted field ("" → ").
-                     * Give this precedence over the closing-quote check, but only
-                     * when another byte follows the doubled pair.
-                     *
-                     * Compatibility note: we intentionally do NOT force terminal
-                     * "" to be consumed here. SmarterCSV has a long-standing lenient
-                     * behavior for malformed tails like ...\"" in :double_quotes mode:
-                     * the final quote may still close the field instead of turning the
-                     * row into an unclosed-quote error. Issue #334 needs doubled-quote
-                     * precedence for ..."",... (more content follows), but we keep the
-                     * historical leniency for terminal ..."". */
-                    p++;
-                  } else {
-                    /* closing quote: only valid if followed by col_sep, row_sep, or end */
-                    bool valid_close = (p + 1 >= endP);
-                    if (!valid_close) {
-                      valid_close = true;
-                      for (long j = 0; j < col_sep_len; j++) {
-                        if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
-                      }
-                    }
-                    if (!valid_close && row_sep_len2 > 0) {
-                      valid_close = true;
-                      for (long j = 0; j < row_sep_len2; j++) {
-                        if (*(p + 1 + j) != *(row_sepP2 + j)) { valid_close = false; break; }
-                      }
+                  /* closing quote: only valid if followed by col_sep, row_sep, or end */
+                  bool valid_close = (p + 1 >= endP);
+                  if (!valid_close) {
+                    valid_close = true;
+                    for (long j = 0; j < col_sep_len; j++) {
+                      if (*(p + 1 + j) != *(col_sepP + j)) { valid_close = false; break; }
                     }
-                    if (valid_close) {
-                      in_quotes     = false;
-                      field_started = true;
+                  }
+                  if (!valid_close && row_sep_len2 > 0) {
+                    valid_close = true;
+                    for (long j = 0; j < row_sep_len2; j++) {
+                      if (*(p + 1 + j) != *(row_sepP2 + j)) { valid_close = false; break; }
                     }
-                    /* else: quote inside quoted field → literal */
                   }
+                  if (valid_close) {
+                    in_quotes     = false;
+                    field_started = true;
+                  }
+                  /* else: quote inside quoted field → literal (handles "" doubling) */
                 } else if (!field_started) {
                   in_quotes     = true;   /* opening quote at field boundary */
                   field_started = true;
@@ -1783,7 +1717,7 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
 
       bool has_embedded_quotes = quoted || (trimmed_len > 0 && memchr(trim_start, quote_char_val, trimmed_len));
 
-      if (!keep_bitmap || (element_count < keep_bitmap_len ? keep_bitmap[element_count] : keep_extra_columns)) {
+      if (!keep_bitmap || (element_count < headers_len ? keep_bitmap[element_count] : keep_extra_columns)) {
         if (insert_field_into_hash(&xform, trim_start, trimmed_len, element_count, has_embedded_quotes, quote_char_val, encoding))
           all_blank = false;
       }
@@ -1794,15 +1728,11 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
   /* ----------------------------------------
    * SECTION 6: Handle blank rows
    * ---------------------------------------- */
-  if (all_blank) {
-    if (remove_empty) {
-      VALUE result = rb_ary_new_capa(2);
-      rb_ary_push(result, Qnil);
-      rb_ary_push(result, LONG2FIX(element_count));
-      return result;
-    }
-
-    ensure_hash_allocated(&xform);
+  if (remove_empty && all_blank) {
+    VALUE result = rb_ary_new_capa(2);
+    rb_ary_push(result, Qnil);
+    rb_ary_push(result, LONG2FIX(element_count));
+    return result;
   }
 
   /* ----------------------------------------
@@ -1811,7 +1741,7 @@ __attribute__((hot)) static VALUE rb_parse_line_to_hash_ctx(VALUE self, VALUE li
   if (!remove_empty_values) {
     ensure_hash_allocated(&xform);
     for (long i = element_count; i < headers_len; i++) {
-      if (!keep_bitmap || (i < keep_bitmap_len ? keep_bitmap[i] : keep_extra_columns)) {
+      if (!keep_bitmap || keep_bitmap[i]) {
         rb_hash_aset(xform.hash, rb_ary_entry(headers, i), Qnil);
       }
     }