Get rwarray extension working with new API.

8 files changed, 293 insertions, 258 deletions

diff --git a/ChangeLog b/ChangeLog
index dbfc8efb..9f6cad4d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,8 @@
+2012-06-24         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* ext.c (load_ext): Don't retun a value from a void function.
+	* gawkapi.c (api_set_array_element): Set up vname and parent_array.
+
 2012-06-21         Arnold D. Robbins     <arnold@skeeve.com>
 
 	More API and cleanup:
diff --git a/ext.c b/ext.c
index 911754bf..0b87def9 100644
--- a/ext.c
+++ b/ext.c
@@ -69,9 +69,9 @@ load_ext(const char *lib_name)
 	if (install_func(& api_impl, NULL /* ext_id */) == 0) {
 		warning(_("load_ext: library `%s' initialization routine `%s' failed\n"),
 				lib_name, INIT_FUNC);
-		return make_number(-1);
+		return;
 	}
-	return make_number(0);
+	return;
 }
 
 
diff --git a/extension/ChangeLog b/extension/ChangeLog
index b482d6f1..9d4305c1 100644
--- a/extension/ChangeLog
+++ b/extension/ChangeLog
@@ -1,3 +1,9 @@
+2012-06-24         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* Makefile.am: Enable rwarray extension.
+	* rwarray.c: Redone to use new API.
+	* rwarray.awk: Revamped for new version.
+
 2012-06-21         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* testext.c (test_array_elem): Add a subarray.
diff --git a/extension/Makefile.am b/extension/Makefile.am
index bdb1bd19..767c2abd 100644
--- a/extension/Makefile.am
+++ b/extension/Makefile.am
@@ -1,7 +1,7 @@
 #
 # extension/Makefile.am --- automake input file for gawk
 #
-# Copyright (C) 1995-2006 the Free Software Foundation, Inc.
+# Copyright (C) 1995-2006, 2012 the Free Software Foundation, Inc.
 #
 # This file is part of GAWK, the GNU implementation of the
 # AWK Programming Language.
@@ -36,6 +36,7 @@ pkgextension_LTLIBRARIES =	\
 	fork.la		\
 	ordchr.la	\
 	readfile.la	\
+	rwarray.la	\
 	testext.la	\
 	time.la
 
@@ -49,12 +50,12 @@ ordchr_la_SOURCES     = ordchr.c
 ordchr_la_LDFLAGS     = $(MY_MODULE_FLAGS)
 readfile_la_SOURCES   = readfile.c
 readfile_la_LDFLAGS   = $(MY_MODULE_FLAGS)
+rwarray_la_SOURCES    = rwarray.c
+rwarray_la_LDFLAGS    = $(MY_MODULE_FLAGS)
 time_la_SOURCES       = time.c
 time_la_LDFLAGS       = $(MY_MODULE_FLAGS)
-testext_la_SOURCES       = testext.c
-testext_la_LDFLAGS       = $(MY_MODULE_FLAGS)
-#rwarray_la_SOURCES    = rwarray.c
-#rwarray_la_LDFLAGS    = $(MY_MODULE_FLAGS)
+testext_la_SOURCES    = testext.c
+testext_la_LDFLAGS    = $(MY_MODULE_FLAGS)
 
 EXTRA_DIST = \
 	ChangeLog \
diff --git a/extension/Makefile.in b/extension/Makefile.in
index 202989a9..5666c9d4 100644
--- a/extension/Makefile.in
+++ b/extension/Makefile.in
@@ -18,7 +18,7 @@
 #
 # extension/Makefile.am --- automake input file for gawk
 #
-# Copyright (C) 1995-2006 the Free Software Foundation, Inc.
+# Copyright (C) 1995-2006, 2012 the Free Software Foundation, Inc.
 #
 # This file is part of GAWK, the GNU implementation of the
 # AWK Programming Language.
@@ -154,6 +154,12 @@ readfile_la_OBJECTS = $(am_readfile_la_OBJECTS)
 readfile_la_LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) \
 	$(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \
 	$(readfile_la_LDFLAGS) $(LDFLAGS) -o $@
+rwarray_la_LIBADD =
+am_rwarray_la_OBJECTS = rwarray.lo
+rwarray_la_OBJECTS = $(am_rwarray_la_OBJECTS)
+rwarray_la_LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) \
+	$(LIBTOOLFLAGS) --mode=link $(CCLD) $(AM_CFLAGS) $(CFLAGS) \
+	$(rwarray_la_LDFLAGS) $(LDFLAGS) -o $@
 testext_la_LIBADD =
 am_testext_la_OBJECTS = testext.lo
 testext_la_OBJECTS = $(am_testext_la_OBJECTS)
@@ -181,10 +187,10 @@ LINK = $(LIBTOOL) --tag=CC $(AM_LIBTOOLFLAGS) $(LIBTOOLFLAGS) \
 	$(LDFLAGS) -o $@
 SOURCES = $(filefuncs_la_SOURCES) $(fork_la_SOURCES) \
 	$(ordchr_la_SOURCES) $(readfile_la_SOURCES) \
-	$(testext_la_SOURCES) $(time_la_SOURCES)
+	$(rwarray_la_SOURCES) $(testext_la_SOURCES) $(time_la_SOURCES)
 DIST_SOURCES = $(filefuncs_la_SOURCES) $(fork_la_SOURCES) \
 	$(ordchr_la_SOURCES) $(readfile_la_SOURCES) \
-	$(testext_la_SOURCES) $(time_la_SOURCES)
+	$(rwarray_la_SOURCES) $(testext_la_SOURCES) $(time_la_SOURCES)
 am__can_run_installinfo = \
   case $$AM_UPDATE_INFO_DIR in \
     n|no|NO) false;; \
@@ -337,6 +343,7 @@ pkgextension_LTLIBRARIES = \
 	fork.la		\
 	ordchr.la	\
 	readfile.la	\
+	rwarray.la	\
 	testext.la	\
 	time.la
 
@@ -349,12 +356,12 @@ ordchr_la_SOURCES = ordchr.c
 ordchr_la_LDFLAGS = $(MY_MODULE_FLAGS)
 readfile_la_SOURCES = readfile.c
 readfile_la_LDFLAGS = $(MY_MODULE_FLAGS)
+rwarray_la_SOURCES = rwarray.c
+rwarray_la_LDFLAGS = $(MY_MODULE_FLAGS)
 time_la_SOURCES = time.c
 time_la_LDFLAGS = $(MY_MODULE_FLAGS)
 testext_la_SOURCES = testext.c
 testext_la_LDFLAGS = $(MY_MODULE_FLAGS)
-#rwarray_la_SOURCES    = rwarray.c
-#rwarray_la_LDFLAGS    = $(MY_MODULE_FLAGS)
 EXTRA_DIST = \
 	ChangeLog \
 	ChangeLog.0 \
@@ -456,6 +463,8 @@ ordchr.la: $(ordchr_la_OBJECTS) $(ordchr_la_DEPENDENCIES) $(EXTRA_ordchr_la_DEPE
 	$(ordchr_la_LINK) -rpath $(pkgextensiondir) $(ordchr_la_OBJECTS) $(ordchr_la_LIBADD) $(LIBS)
 readfile.la: $(readfile_la_OBJECTS) $(readfile_la_DEPENDENCIES) $(EXTRA_readfile_la_DEPENDENCIES) 
 	$(readfile_la_LINK) -rpath $(pkgextensiondir) $(readfile_la_OBJECTS) $(readfile_la_LIBADD) $(LIBS)
+rwarray.la: $(rwarray_la_OBJECTS) $(rwarray_la_DEPENDENCIES) $(EXTRA_rwarray_la_DEPENDENCIES) 
+	$(rwarray_la_LINK) -rpath $(pkgextensiondir) $(rwarray_la_OBJECTS) $(rwarray_la_LIBADD) $(LIBS)
 testext.la: $(testext_la_OBJECTS) $(testext_la_DEPENDENCIES) $(EXTRA_testext_la_DEPENDENCIES) 
 	$(testext_la_LINK) -rpath $(pkgextensiondir) $(testext_la_OBJECTS) $(testext_la_LIBADD) $(LIBS)
 time.la: $(time_la_OBJECTS) $(time_la_DEPENDENCIES) $(EXTRA_time_la_DEPENDENCIES) 
@@ -471,6 +480,7 @@ distclean-compile:
 @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/fork.Plo@am__quote@
 @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/ordchr.Plo@am__quote@
 @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/readfile.Plo@am__quote@
+@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/rwarray.Plo@am__quote@
 @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/testext.Plo@am__quote@
 @AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/time.Plo@am__quote@
 
diff --git a/extension/rwarray.awk b/extension/rwarray.awk
index 1057b396..b34a42a7 100644
--- a/extension/rwarray.awk
+++ b/extension/rwarray.awk
@@ -1,19 +1,24 @@
-BEGIN {
-	extension("./rwarray.so","dlload")
+@load "rwarray"
 
+BEGIN {
 	while ((getline word < "/usr/share/dict/words") > 0)
 		dict[word] = word word
 
-	for (i in dict)
-		printf("dict[%s] = %s\n", i, dict[i]) > "orig.out"
+	n = asorti(dict, dictindices)
+	for (i = 1; i <= n; i++)
+		printf("dict[%s] = %s\n", dictindices[i], dict[dictindices[i]]) > "orig.out"
 	close("orig.out");
 
-	writea("orig.bin", dict)
+	ret = writea("orig.bin", dict)
+	printf "writea() returned %d, expecting 1\n", ret
 
-	reada("orig.bin", dict)
+ 
+	ret = reada("orig.bin", dict)
+	printf "reada() returned %d, expecting 1\n", ret
 
-	for (i in dict)
-		printf("dict[%s] = %s\n", i, dict[i]) > "new.out"
+	n = asorti(dict, dictindices)
+	for (i = 1; i <= n; i++)
+		printf("dict[%s] = %s\n", dictindices[i], dict[dictindices[i]]) > "new.out"
 	close("new.out");
 
 	ret = system("cmp orig.out new.out")
@@ -23,6 +28,6 @@ BEGIN {
 	else
 		print "old and new are not equal - BAD"
 
-	if (ret == 0 && !("keepit" in ENVIRON))
-		system("rm orig.bin orig.out new.out")
+	if (ret == 0 && !("KEEPIT" in ENVIRON))
+		system("rm -f orig.bin orig.out new.out")
 }
diff --git a/extension/rwarray.c b/extension/rwarray.c
index fed040fc..e4ddde3b 100644
--- a/extension/rwarray.c
+++ b/extension/rwarray.c
@@ -3,10 +3,11 @@
  *
  * Arnold Robbins
  * May 2009
+ * Redone June 2012
  */
 
 /*
- * Copyright (C) 2009, 2010, 2011 the Free Software Foundation, Inc.
+ * Copyright (C) 2009, 2010, 2011, 2012 the Free Software Foundation, Inc.
  * 
  * This file is part of GAWK, the GNU implementation of the
  * AWK Programming Language.
@@ -26,76 +27,98 @@
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA
  */
 
-#include "awk.h"
+#include <stdio.h>
+#include <assert.h>
+#include <errno.h>
+#include <fcntl.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+
 #include <arpa/inet.h>
 #include <sys/types.h>
 #include <sys/stat.h>
-#include <fcntl.h>
 
+#include "gawkapi.h"
 
 #define MAGIC "awkrulz\n"
-#define MAJOR 1
+#define MAJOR 3
 #define MINOR 0
 
+static const gawk_api_t *api;	/* for convenience macros to work */
+static awk_ext_id_t *ext_id;
+
 int plugin_is_GPL_compatible;
 
-static int write_array(int fd, NODE *array);
-static int write_elem(int fd, int index, NODE *item);
-static int write_chain(int fd, int index, NODE *item);
-static int write_value(int fd, NODE *val);
+static awk_bool_t write_array(int fd, awk_array_t array);
+static awk_bool_t write_elem(int fd, awk_element_t *element);
+static awk_bool_t write_value(int fd, awk_value_t *val);
 
-static int read_array(int fd, NODE *array);
-static NODE *read_elem(int fd, int *index, NODE *array);
-static NODE *read_value(int fd);
+static awk_bool_t read_array(int fd, awk_array_t array);
+static awk_bool_t read_elem(int fd, awk_element_t *element);
+static awk_bool_t read_value(int fd, awk_value_t *value);
 
 /*
  * Format of array info:
  *
- * MAGIC	8 bytes
+ * MAGIC		8 bytes
  * Major version	4 bytes - network order
  * Minor version	4 bytes - network order
  * Element count	4 bytes - network order
- * Array size		4 bytes - network order
  * Elements
  * 
  * For each element:
- * Bucket number:	4 bytes - network order
- * Hash of index val:	4 bytes - network order
  * Length of index val:	4 bytes - network order
  * Index val as characters (N bytes)
- * Value type		1 byte (0 = string, 1 = number, 2 = array)
+ * Value type		4 bytes (0 = string, 1 = number, 2 = array)
  * IF string:
  * 	Length of value	4 bytes
  * 	Value as characters (N bytes)
- * ELSE
+ * ELSE IF number:
  * 	8 bytes as native double
+ * ELSE
+ * 	Element count
+ * 	Elements
+ * END IF
  */
 
 /* do_writea --- write an array */
 
-static NODE *
-do_writea(int nargs)
+static awk_value_t *
+do_writea(int nargs, awk_value_t *result)
 {
-	NODE *file, *array;
-	int ret;
-	int fd;
+	awk_value_t filename, array;
+	int fd = -1;
 	uint32_t major = MAJOR;
 	uint32_t minor = MINOR;
 
+	assert(result != NULL);
+	make_number(0.0, result);
+
 	if (do_lint && nargs > 2)
-		lintwarn("writea: called with too many arguments");
+		lintwarn(ext_id, "writea: called with too many arguments");
+
+	if (nargs < 2)
+		goto out;
 
 	/* directory is first arg, array to dump is second */
-	file = get_scalar_argument(0, false);
-	array = get_array_argument(1, false);
+	if (! get_argument(0, AWK_STRING, & filename)) {
+		fprintf(stderr, "do_writea: argument 0 is not a string\n");
+		errno = EINVAL;
+		goto done1;
+	}
 
-	/* open the file, if error, set ERRNO and return */
-	(void) force_string(file);
-	fd = creat(file->stptr, 0600);
-	if (fd < 0) {
+	if (! get_argument(1, AWK_ARRAY, & array)) {
+		fprintf(stderr, "do_writea: argument 1 is not an array\n");
+		errno = EINVAL;
 		goto done1;
 	}
 
+	/* open the file, if error, set ERRNO and return */
+	fd = creat(filename.str_value.str, 0600);
+	if (fd < 0)
+		goto done1;
+
 	if (write(fd, MAGIC, strlen(MAGIC)) != strlen(MAGIC))
 		goto done1;
 
@@ -107,356 +130,336 @@ do_writea(int nargs)
 	if (write(fd, & minor, sizeof(minor)) != sizeof(minor))
 		goto done1;
 
-	ret = write_array(fd, array);
-	if (ret != 0)
-		goto done1;
-	ret = 0;
-	goto done0;
+	if (write_array(fd, array.array_cookie)) {
+		make_number(1.0, result);
+		goto done0;
+	}
 
 done1:
-	ret = -1;
 	update_ERRNO_int(errno);
-	unlink(file->stptr);
+	unlink(filename.str_value.str);
 
 done0:
 	close(fd);
-
-	/* Set the return value */
-	return make_number((AWKNUM) ret);
+out:
+	return result;
 }
 
 
 /* write_array --- write out an array or a sub-array */
 
-static int
-write_array(int fd, NODE *array)
+static awk_bool_t
+write_array(int fd, awk_array_t array)
 {
-	int ret;
+	uint32_t i;
 	uint32_t count;
-	uint32_t array_sz;
-	int i;
-
-	count = htonl(array->table_size);
-	if (write(fd, & count, sizeof(count)) != sizeof(count))
-		return -1;
+	awk_flat_array_t *flat_array;
 
-	array_sz = htonl(array->array_size);
-	if (write(fd, & array_sz, sizeof(array_sz)) != sizeof(array_sz))
-		return -1;
-
-	for (i = 0; i < array->array_size; i++) {
-		ret = write_chain(fd, i, array->var_array[i]);
-		if (ret != 0)
-			return ret;
+	if (! flatten_array(array, & flat_array)) {
+		printf("write_array: could not flatten array\n");
+		return 0;
 	}
-	return 0;
-}
 
+	count = htonl(flat_array->count);
+	if (write(fd, & count, sizeof(count)) != sizeof(count))
+		return 0;
 
-/* write_chain --- write out a whole hash chain */
-
-/*
- * Write elements in the chain in reverse order so that
- * when we read the elements back in we can just push them
- * onto the front and thus recreate the array as it was.
- */
-
-static int
-write_chain(int fd, int index, NODE *bucket)
-{
-	int ret;
+	for (i = 0; i < flat_array->count; i++) {
+		if (! write_elem(fd, & flat_array->elements[i]))
+			return 0;
+	}
 
-	if (bucket == NULL)
+	if (! release_flattened_array(array, flat_array)) {
+		printf("write_array: could not release flattened array\n");
 		return 0;
+	}
 
-	ret = write_chain(fd, index, bucket->ahnext);
-	if (ret != 0)
-		return ret;
-
-	return write_elem(fd, index, bucket);
+	return 1;
 }
 
 /* write_elem --- write out a single element */
 
-static int
-write_elem(int fd, int index, NODE *item)
+static awk_bool_t
+write_elem(int fd, awk_element_t *element)
 {
-	uint32_t hashval, indexval_len;
-
-	index = htonl(index);
-	if (write(fd, & index, sizeof(index)) != sizeof(index))
-		return -1;
+	uint32_t indexval_len;
+	ssize_t write_count;
 
-	hashval = htonl(item->ahcode);
-	if (write(fd, & hashval, sizeof(hashval)) != sizeof(hashval))
-		return -1;
-
-	indexval_len = htonl(item->ahname_len);
+	indexval_len = htonl(element->index.str_value.len);
 	if (write(fd, & indexval_len, sizeof(indexval_len)) != sizeof(indexval_len))
-		return -1;
+		return 0;
 
-	if (write(fd, item->ahname_str, item->ahname_len) != item->ahname_len)
-		return -1;
+	if (element->index.str_value.len > 0) {
+		write_count = write(fd, element->index.str_value.str,
+				element->index.str_value.len);
+		if (write_count != (ssize_t) element->index.str_value.len)
+			return 0;
+	}
 
-	return write_value(fd, item->ahvalue);
+	return write_value(fd, & element->value);
 }
 
 /* write_value --- write a number or a string or a array */
 
 static int
-write_value(int fd, NODE *val)
+write_value(int fd, awk_value_t *val)
 {
-	int code, len;
+	uint32_t code, len;
 
-	if (val->type == Node_var_array) {
+	if (val->val_type == AWK_ARRAY) {
 		code = htonl(2);
 		if (write(fd, & code, sizeof(code)) != sizeof(code))
-			return -1;
-		return write_array(fd, val);
+			return 0;
+		return write_array(fd, val->array_cookie);
 	}
 
-	if ((val->flags & NUMBER) != 0) {
+	if (val->val_type == AWK_NUMBER) {
 		code = htonl(1);
 		if (write(fd, & code, sizeof(code)) != sizeof(code))
-			return -1;
+			return 0;
 
-		if (write(fd, & val->numbr, sizeof(val->numbr)) != sizeof(val->numbr))
-			return -1;
+		if (write(fd, & val->num_value, sizeof(val->num_value)) != sizeof(val->num_value))
+			return 0;
 	} else {
 		code = 0;
 		if (write(fd, & code, sizeof(code)) != sizeof(code))
-			return -1;
+			return 0;
 
-		len = htonl(val->stlen);
+		len = htonl(val->str_value.len);
 		if (write(fd, & len, sizeof(len)) != sizeof(len))
-			return -1;
+			return 0;
 
-		if (write(fd, val->stptr, val->stlen) != val->stlen)
-			return -1;
+		if (write(fd, val->str_value.str, val->str_value.len)
+				!= (ssize_t) val->str_value.len)
+			return 0;
 	}
 
-	return 0;
+	return 1;
 }
 
 /* do_reada --- read an array */
 
-static NODE *
-do_reada(int nargs)
+static awk_value_t *
+do_reada(int nargs, awk_value_t *result)
 {
-	NODE *file, *array;
-	int ret;
-	int fd;
+	awk_value_t filename, array;
+	int fd = -1;
 	uint32_t major;
 	uint32_t minor;
 	char magic_buf[30];
 
+	assert(result != NULL);
+	make_number(0.0, result);
+
 	if (do_lint && nargs > 2)
-		lintwarn("reada: called with too many arguments");
+		lintwarn(ext_id, "reada: called with too many arguments");
 
-	/* directory is first arg, array to dump is second */
-	file = get_scalar_argument(0, false);
-	array = get_array_argument(1, false);
+	if (nargs < 2)
+		goto out;
 
-	(void) force_string(file);
-	fd = open(file->stptr, O_RDONLY);
-	if (fd < 0) {
+	/* directory is first arg, array to read is second */
+	if (! get_argument(0, AWK_STRING, & filename)) {
+		fprintf(stderr, "do_reada: argument 0 is not a string\n");
+		errno = EINVAL;
 		goto done1;
 	}
 
+	if (! get_argument(1, AWK_ARRAY, & array)) {
+		fprintf(stderr, "do_reada: argument 1 is not an array\n");
+		errno = EINVAL;
+		goto done1;
+	}
+
+	fd = open(filename.str_value.str, O_RDONLY);
+	if (fd < 0)
+		goto done1;
+
 	memset(magic_buf, '\0', sizeof(magic_buf));
 	if (read(fd, magic_buf, strlen(MAGIC)) != strlen(MAGIC)) {
+		errno = EBADF;
 		goto done1;
 	}
 
 	if (strcmp(magic_buf, MAGIC) != 0) {
+		errno = EBADF;
 		goto done1;
 	}
 
 	if (read(fd, & major, sizeof(major)) != sizeof(major)) {
+		errno = EBADF;
 		goto done1;
 	}
 	major = ntohl(major);
 
 	if (major != MAJOR) {
+		errno = EBADF;
 		goto done1;
 	}
 
 	if (read(fd, & minor, sizeof(minor)) != sizeof(minor)) {
+		/* read() sets errno */
 		goto done1;
 	}
+
 	minor = ntohl(minor);
 	if (minor != MINOR) {
+		errno = EBADF;
 		goto done1;
 	}
 
-	assoc_clear(array, NULL);
+	if (! clear_array(array.array_cookie)) {
+		errno = ENOMEM;
+		printf("do_reada: clear_array failed\n");
+		goto done1;
+	}
 
-	ret = read_array(fd, array);
-	if (ret == 0)
+	if (read_array(fd, array.array_cookie)) {
+		make_number(1.0, result);
 		goto done0;
+	}
 
 done1:
-	ret = -1;
 	update_ERRNO_int(errno);
-
 done0:
 	close(fd);
-
-	/* Set the return value */
-	return make_number((AWKNUM) ret);
+out:
+	return result;
 }
 
 
 /* read_array --- read in an array or sub-array */
 
-static int
-read_array(int fd, NODE *array)
+static awk_bool_t
+read_array(int fd, awk_array_t array)
 {
-	int i;
+	uint32_t i;
 	uint32_t count;
-	uint32_t array_sz;
-	int index;
-	NODE *new_elem;
+	awk_element_t new_elem;
 
 	if (read(fd, & count, sizeof(count)) != sizeof(count)) {
-		return -1;
-	}
-	array->table_size = ntohl(count);
-
-	if (read(fd, & array_sz, sizeof(array_sz)) != sizeof(array_sz)) {
-		return -1;
+		return 0;
 	}
-	array->array_size = ntohl(array_sz);
-
-	/* malloc var_array */
-	array->var_array = (NODE **) malloc(array->array_size * sizeof(NODE *));
-	memset(array->var_array, '\0', array->array_size * sizeof(NODE *));
-
-	for (i = 0; i < array->table_size; i++) {
-		if ((new_elem = read_elem(fd, & index, array)) != NULL) {
-			new_elem->ahnext = array->var_array[index];
-			array->var_array[index] = new_elem;
+	count = ntohl(count);
+
+	for (i = 0; i < count; i++) {
+		if (read_elem(fd, & new_elem)) {
+			/* add to array */
+			if (! set_array_element(array, & new_elem)) {
+				printf("read_array: set_array_element failed\n");
+				return 0;
+			}
 		} else
 			break;
 	}
-	if (i != array->table_size)
-		return -1; 
-	return 0;
-}
 
+	if (i != count)
+		return 0;
+
+	return 1;
+}
 
 /* read_elem --- read in a single element */
 
-static NODE *
-read_elem(int fd, int *the_index, NODE *array)
+static awk_bool_t
+read_elem(int fd, awk_element_t *element)
 {
-	uint32_t hashval, indexval_len, index;
-	NODE *item;
-	NODE *val;
-	int ret;
+	uint32_t index_len;
+	static char *buffer;
+	static uint32_t buflen;
+	ssize_t ret;
 
-	*the_index = 0;
-
-	if ((ret = read(fd, & index, sizeof(index))) != sizeof(index)) {
-		return NULL;
+	if ((ret = read(fd, & index_len, sizeof(index_len))) != sizeof(index_len)) {
+		return 0;
 	}
-	*the_index = index = ntohl(index);
+	index_len = ntohl(index_len);
 
-	getnode(item);
-	memset(item, 0, sizeof(*item));
-	item->type = Node_ahash;
-	item->flags = MALLOC;
+	memset(element, 0, sizeof(*element));
 
-	if (read(fd, & hashval, sizeof(hashval)) != sizeof(hashval)) {
-		return NULL;
-	}
+	if (index_len > 0) {
+		if (buffer == NULL) {
+			// allocate buffer
+			emalloc(buffer, char *, index_len, "read_elem");
+			buflen = index_len;
+		} else if (buflen < index_len) {
+			// reallocate buffer
+			char *cp = realloc(buffer, index_len);
 
-	item->ahcode = ntohl(hashval);
+			if (cp == NULL)
+				return 0;
 
-	if (read(fd, & indexval_len, sizeof(indexval_len)) != sizeof(indexval_len)) {
-		return NULL;
-	}
-	item->ahname_len = ntohl(indexval_len);
+			buffer = cp;
+			buflen = index_len;
+		}
 
-	item->ahname_str = malloc(item->ahname_len + 2);
-	if (read(fd, item->ahname_str, item->ahname_len) != item->ahname_len) {
-		return NULL;
+		if (read(fd, buffer, index_len) != (ssize_t) index_len) {
+			return 0;
+		}
+		make_string(buffer, index_len, & element->index);
+	} else {
+		make_string("", 0, & element->index);
 	}
-	item->ahname_str[item->ahname_len] = '\0';
-	item->ahname_ref = 1;
 
-	item->ahvalue = val = read_value(fd);
-	if (val == NULL) {
-		return NULL;
-	}
-	if (val->type == Node_var_array) {
-		char *aname;
-		size_t aname_len;
-
-		/* construct the sub-array name */
-		aname_len = strlen(array->vname) + item->ahname_len + 4;
-		emalloc(aname, char *, aname_len + 2, "read_elem");
-		sprintf(aname, "%s[\"%.*s\"]", array->vname, (int) item->ahname_len, item->ahname_str);
-		val->vname = aname;
-	}
+	if (! read_value(fd, & element->value))
+		return 0;
 
-	return item;
+	return 1;
 }
 
 /* read_value --- read a number or a string */
 
-static NODE *
-read_value(int fd)
+static awk_bool_t
+read_value(int fd, awk_value_t *value)
 {
-	NODE *val;
-	int code, len;
+	uint32_t code, len;
 
-	getnode(val);
-	memset(val, 0, sizeof(*val));
-	val->type = Node_val;
+	if (read(fd, & code, sizeof(code)) != sizeof(code))
+		return 0;
 
-	if (read(fd, & code, sizeof(code)) != sizeof(code)) {
-		return NULL;
-	}
 	code = ntohl(code);
 
 	if (code == 2) {
-		val->type = Node_var_array;
-		if (read_array(fd, val) != 0)
-			return NULL; 
+		awk_array_t array = create_array();
+
+		if (read_array(fd, array) != 0)
+			return 0; 
+
+		/* hook into value */
+		value->val_type = AWK_ARRAY;
+		value->array_cookie = array;
 	} else if (code == 1) {
-		if (read(fd, & val->numbr, sizeof(val->numbr)) != sizeof(val->numbr)) {
-			return NULL;
-		}
+		double d;
+
+		if (read(fd, & d, sizeof(d)) != sizeof(d))
+			return 0;
 
-		val->flags = NUMBER|NUMCUR|MALLOC;
+		/* hook into value */
+		value->val_type = AWK_NUMBER;
+		value->num_value = d;
 	} else {
 		if (read(fd, & len, sizeof(len)) != sizeof(len)) {
-			return NULL;
+			return 0;
 		}
-		val->stlen = ntohl(len);
-		val->stptr = malloc(val->stlen + 2);
-		memset(val->stptr, '\0', val->stlen + 2);
-
-		if (read(fd, val->stptr, val->stlen) != val->stlen) {
-			return NULL;
+		len = ntohl(len);
+		value->val_type = AWK_STRING;
+		value->str_value.len = len;
+		value->str_value.str = malloc(len + 2);
+		memset(value->str_value.str, '\0', len + 2);
+
+		if (read(fd, value->str_value.str, len) != len) {
+			free(value->str_value.str);
+			return 0;
 		}
-
-		val->flags = STRING|STRCUR|MALLOC;
 	}
 
-	return val;
+	return 1;
 }
 
-/* dlload --- load new builtins in this library */
+static awk_ext_func_t func_table[] = {
+	{ "writea", do_writea, 2 },
+	{ "reada", do_reada, 2 },
+};
 
-NODE *
-dlload(tree, dl)
-NODE *tree;
-void *dl;
-{
-	make_builtin("writea", do_writea, 2);
-	make_builtin("reada", do_reada, 2);
 
-	return make_number((AWKNUM) 0);
-}
+/* define the dl_load function using the boilerplate macro */
+
+dl_load_func(func_table, rwarray, "")
diff --git a/gawkapi.c b/gawkapi.c
index 139d77b8..f87c7759 100644
--- a/gawkapi.c
+++ b/gawkapi.c
@@ -528,6 +528,7 @@ api_set_array_element(awk_ext_id_t id, awk_array_t a_cookie,
 {
 	NODE *array = (NODE *)a_cookie;
 	NODE *tmp;
+	NODE *elem;
 	NODE **aptr;
 
 	/* don't check for index len zero, null str is ok as index */
@@ -542,9 +543,13 @@ api_set_array_element(awk_ext_id_t id, awk_array_t a_cookie,
 	aptr = assoc_lookup(array, tmp);
 	unref(tmp);
 	unref(*aptr);
-	*aptr = awk_value_to_node(& element->value);
-	if ((*aptr)->type == Node_var_array)
-		(*aptr)->parent_array = array;
+	elem = *aptr = awk_value_to_node(& element->value);
+	if (elem->type == Node_var_array) {
+		elem->parent_array = array;
+		elem->vname = estrdup(element->index.str_value.str,
+					element->index.str_value.len);
+		make_aname(elem);
+	}
 
 	return true;
 }