Make SimpleXMLElement a RecursiveIterator

Context: https://externals.io/message/108789

This essentially moves the functionality of SimpleXMLIterator into
SimpleXMLElement, and makes SimpleXMLIterator a no-op extension.

Ideally SimpleXMLElement would be an IteratorAggregate, whose
getIterator() method returns SimpleXMLIterator. However, because
SimpleXMLIterator extends SimpleXMLElement (and code depends on
this in non-trivial ways), this is not possible.

The only way to not keep SimpleXMLElement as a magic Traversable
(that implements neither Iterator nor IteratorAggregate) is to
move the SimpleXMLIterator functionality into it.

Author: nikic

ext/simplexml/config.m4

@@ -12,7 +12,7 @@ if test "$PHP_SIMPLEXML" != "no"; then
 
   PHP_SETUP_LIBXML(SIMPLEXML_SHARED_LIBADD, [
     AC_DEFINE(HAVE_SIMPLEXML,1,[ ])
-    PHP_NEW_EXTENSION(simplexml, simplexml.c sxe.c, $ext_shared)
+    PHP_NEW_EXTENSION(simplexml, simplexml.c, $ext_shared)
     PHP_INSTALL_HEADERS([ext/simplexml/php_simplexml.h ext/simplexml/php_simplexml_exports.h])
     PHP_SUBST(SIMPLEXML_SHARED_LIBADD)
   ])

ext/simplexml/config.w32

@@ -7,7 +7,7 @@ if (PHP_SIMPLEXML == "yes") {
 		ADD_EXTENSION_DEP('simplexml', 'libxml') &&
 		CHECK_HEADER_ADD_INCLUDE("libxml/tree.h", "CFLAGS_SIMPLEXML", PHP_PHP_BUILD + "\\include\\libxml2")
 	) {
-		EXTENSION("simplexml", "simplexml.c sxe.c");
+		EXTENSION("simplexml", "simplexml.c");
 		AC_DEFINE("HAVE_SIMPLEXML", 1, "Simple XML support");
 		if (!PHP_SIMPLEXML_SHARED) {
 			ADD_FLAG("CFLAGS_SIMPLEXML", "/D LIBXML_STATIC");

ext/simplexml/php_simplexml.h

@@ -77,6 +77,9 @@ typedef struct {
 #	define PHP_SXE_API ZEND_API
 #endif
 
+extern PHP_SXE_API zend_class_entry *ce_SimpleXMLIterator;
+extern PHP_SXE_API zend_class_entry *ce_SimpleXMLElement;
+
 PHP_SXE_API zend_class_entry *sxe_get_element_class_entry();
 
 #endif

ext/simplexml/simplexml.c

@@ -31,9 +31,11 @@
 #include "simplexml_arginfo.h"
 #include "zend_exceptions.h"
 #include "zend_interfaces.h"
-#include "sxe.h"
+#include "ext/spl/spl_iterators.h"
 
 zend_class_entry *sxe_class_entry = NULL;
+PHP_SXE_API zend_class_entry *ce_SimpleXMLIterator;
+PHP_SXE_API zend_class_entry *ce_SimpleXMLElement;
 
 PHP_SXE_API zend_class_entry *sxe_get_element_class_entry() /* {{{ */
 {
@@ -2016,6 +2018,138 @@ SXE_METHOD(count)
 }
 /* }}} */
 
+
+/* {{{ proto void SimpleXMLElement::rewind()
+ Rewind to first element */
+SXE_METHOD(rewind)
+{
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	php_sxe_rewind_iterator(Z_SXEOBJ_P(ZEND_THIS));
+}
+/* }}} */
+
+/* {{{ proto bool SimpleXMLElement::valid()
+ Check whether iteration is valid */
+SXE_METHOD(valid)
+{
+	php_sxe_object *sxe = Z_SXEOBJ_P(ZEND_THIS);
+
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	RETURN_BOOL(!Z_ISUNDEF(sxe->iter.data));
+}
+/* }}} */
+
+/* {{{ proto SimpleXMLElement SimpleXMLElement::current()
+ Get current element */
+SXE_METHOD(current)
+{
+	php_sxe_object *sxe = Z_SXEOBJ_P(ZEND_THIS);
+	zval *data;
+
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	if (Z_ISUNDEF(sxe->iter.data)) {
+		return; /* return NULL */
+	}
+
+	data = &sxe->iter.data;
+	ZVAL_COPY_DEREF(return_value, data);
+}
+/* }}} */
+
+/* {{{ proto string SimpleXMLElement::key()
+ Get name of current child element */
+SXE_METHOD(key)
+{
+	xmlNodePtr curnode;
+	php_sxe_object *intern;
+	php_sxe_object *sxe = Z_SXEOBJ_P(ZEND_THIS);
+
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	if (Z_ISUNDEF(sxe->iter.data)) {
+		RETURN_FALSE;
+	}
+
+	intern = Z_SXEOBJ_P(&sxe->iter.data);
+	if (intern != NULL && intern->node != NULL) {
+		curnode = (xmlNodePtr)((php_libxml_node_ptr *)intern->node)->node;
+		RETURN_STRINGL((char*)curnode->name, xmlStrlen(curnode->name));
+	}
+
+	RETURN_FALSE;
+}
+/* }}} */
+
+/* {{{ proto void SimpleXMLElement::next()
+ Move to next element */
+SXE_METHOD(next)
+{
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	php_sxe_move_forward_iterator(Z_SXEOBJ_P(ZEND_THIS));
+}
+/* }}} */
+
+/* {{{ proto bool SimpleXMLElement::hasChildren()
+ Check whether element has children (elements) */
+SXE_METHOD(hasChildren)
+{
+	php_sxe_object *sxe = Z_SXEOBJ_P(ZEND_THIS);
+	php_sxe_object *child;
+	xmlNodePtr      node;
+
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	if (Z_ISUNDEF(sxe->iter.data) || sxe->iter.type == SXE_ITER_ATTRLIST) {
+		RETURN_FALSE;
+	}
+	child = Z_SXEOBJ_P(&sxe->iter.data);
+
+	GET_NODE(child, node);
+	if (node) {
+		node = node->children;
+	}
+	while (node && node->type != XML_ELEMENT_NODE) {
+		node = node->next;
+	}
+	RETURN_BOOL(node ? 1 : 0);
+}
+/* }}} */
+
+/* {{{ proto SimpleXMLElement SimpleXMLElement::getChildren()
+ Get child element iterator */
+SXE_METHOD(getChildren)
+{
+	php_sxe_object *sxe = Z_SXEOBJ_P(ZEND_THIS);
+	zval *data;
+
+	if (zend_parse_parameters_none() == FAILURE) {
+		RETURN_THROWS();
+	}
+
+	if (Z_ISUNDEF(sxe->iter.data) || sxe->iter.type == SXE_ITER_ATTRLIST) {
+		return; /* return NULL */
+	}
+
+	data = &sxe->iter.data;
+	ZVAL_COPY_DEREF(return_value, data);
+}
+
 static zend_object_handlers sxe_object_handlers;
 
 /* {{{ sxe_object_clone()
@@ -2619,6 +2753,13 @@ static const zend_function_entry sxe_functions[] = { /* {{{ */
 	SXE_ME(addAttribute,           arginfo_class_SimpleXMLElement_addAttribute, ZEND_ACC_PUBLIC)
 	SXE_ME(__toString,             arginfo_class_SimpleXMLElement___toString, ZEND_ACC_PUBLIC)
 	SXE_ME(count,                  arginfo_class_SimpleXMLElement_count, ZEND_ACC_PUBLIC)
+	SXE_ME(rewind,                 arginfo_class_SimpleXMLElement_rewind, ZEND_ACC_PUBLIC)
+	SXE_ME(valid,                  arginfo_class_SimpleXMLElement_valid, ZEND_ACC_PUBLIC)
+	SXE_ME(current,                arginfo_class_SimpleXMLElement_current, ZEND_ACC_PUBLIC)
+	SXE_ME(key,                    arginfo_class_SimpleXMLElement_key, ZEND_ACC_PUBLIC)
+	SXE_ME(next,                   arginfo_class_SimpleXMLElement_next, ZEND_ACC_PUBLIC)
+	SXE_ME(hasChildren,            arginfo_class_SimpleXMLElement_hasChildren, ZEND_ACC_PUBLIC)
+	SXE_ME(getChildren,            arginfo_class_SimpleXMLElement_getChildren, ZEND_ACC_PUBLIC)
 	PHP_FE_END
 };
 /* }}} */
@@ -2627,13 +2768,14 @@ static const zend_function_entry sxe_functions[] = { /* {{{ */
  */
 PHP_MINIT_FUNCTION(simplexml)
 {
-	zend_class_entry sxe;
+	zend_class_entry ce;
 
-	INIT_CLASS_ENTRY(sxe, "SimpleXMLElement", sxe_functions);
-	sxe.create_object = sxe_object_new;
-	sxe_class_entry = zend_register_internal_class(&sxe);
+	INIT_CLASS_ENTRY(ce, "SimpleXMLElement", sxe_functions);
+	sxe_class_entry = zend_register_internal_class(&ce);
+	sxe_class_entry->create_object = sxe_object_new;
 	sxe_class_entry->get_iterator = php_sxe_get_iterator;
-	zend_class_implements(sxe_class_entry, 3, zend_ce_traversable, zend_ce_countable, zend_ce_stringable);
+	zend_class_implements(sxe_class_entry, 3,
+		zend_ce_countable, zend_ce_stringable, spl_ce_RecursiveIterator);
 
 	memcpy(&sxe_object_handlers, &std_object_handlers, sizeof(zend_object_handlers));
 	sxe_object_handlers.offset = XtOffsetOf(php_sxe_object, zo);
@@ -2660,9 +2802,13 @@ PHP_MINIT_FUNCTION(simplexml)
 	sxe_class_entry->serialize = zend_class_serialize_deny;
 	sxe_class_entry->unserialize = zend_class_unserialize_deny;
 
-	php_libxml_register_export(sxe_class_entry, simplexml_export_node);
+	/* TODO: Why do we have two variables for this? */
+	ce_SimpleXMLElement = sxe_class_entry;
 
-	PHP_MINIT(sxe)(INIT_FUNC_ARGS_PASSTHRU);
+	INIT_CLASS_ENTRY(ce, "SimpleXMLIterator", NULL);
+	ce_SimpleXMLIterator = zend_register_internal_class_ex(&ce, ce_SimpleXMLElement);
+
+	php_libxml_register_export(sxe_class_entry, simplexml_export_node);
 
 	return SUCCESS;
 }

ext/simplexml/simplexml.stub.php

@@ -6,7 +6,7 @@ function simplexml_load_string(string $data, ?string $class_name = SimpleXMLElem
 
 function simplexml_import_dom(DOMNode $node, ?string $class_name = SimpleXMLElement::class): ?SimpleXMLElement {}
 
-class SimpleXMLElement implements Stringable
+class SimpleXMLElement implements Stringable, Countable, RecursiveIterator
 {
     /** @return array|false */
     public function xpath(string $path) {}
@@ -47,4 +47,29 @@ public function __toString(): string {}
 
     /** @return int */
     public function count() {}
+
+    /** @return void */
+    public function rewind() {}
+
+    /** @return bool */
+    public function valid() {}
+
+    /** @return ?SimpleXMLElement */
+    public function current() {}
+
+    /** @return string|false */
+    public function key() {}
+
+    /** @return void */
+    public function next() {}
+
+    /** @return bool */
+    public function hasChildren() {}
+
+    /** @return ?SimpleXMLElement */
+    public function getChildren() {}
+}
+
+class SimpleXMLIterator extends SimpleXMLElement
+{
 }

ext/simplexml/simplexml_arginfo.h

@@ -75,3 +75,17 @@ ZEND_BEGIN_ARG_WITH_RETURN_TYPE_INFO_EX(arginfo_class_SimpleXMLElement___toStrin
 ZEND_END_ARG_INFO()
 
 #define arginfo_class_SimpleXMLElement_count arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_rewind arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_valid arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_current arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_key arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_next arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_hasChildren arginfo_class_SimpleXMLElement_getName
+
+#define arginfo_class_SimpleXMLElement_getChildren arginfo_class_SimpleXMLElement_getName