godot/core/io/xml_parser.cpp

532 lines
12 KiB
C++
Raw Normal View History

2014-02-10 01:10:30 +00:00
/*************************************************************************/
/* xml_parser.cpp */
/*************************************************************************/
/* This file is part of: */
/* GODOT ENGINE */
/* https://godotengine.org */
2014-02-10 01:10:30 +00:00
/*************************************************************************/
/* Copyright (c) 2007-2021 Juan Linietsky, Ariel Manzur. */
/* Copyright (c) 2014-2021 Godot Engine contributors (cf. AUTHORS.md). */
2014-02-10 01:10:30 +00:00
/* */
/* Permission is hereby granted, free of charge, to any person obtaining */
/* a copy of this software and associated documentation files (the */
/* "Software"), to deal in the Software without restriction, including */
/* without limitation the rights to use, copy, modify, merge, publish, */
/* distribute, sublicense, and/or sell copies of the Software, and to */
/* permit persons to whom the Software is furnished to do so, subject to */
/* the following conditions: */
/* */
/* The above copyright notice and this permission notice shall be */
/* included in all copies or substantial portions of the Software. */
/* */
/* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, */
/* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF */
/* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.*/
/* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY */
/* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, */
/* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE */
/* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. */
/*************************************************************************/
2014-02-10 01:10:30 +00:00
#include "xml_parser.h"
#include "core/print_string.h"
2014-02-10 01:10:30 +00:00
//#define DEBUG_XML
VARIANT_ENUM_CAST(XMLParser::NodeType);
static inline bool _is_white_space(char c) {
return (c == ' ' || c == '\t' || c == '\n' || c == '\r');
2014-02-10 01:10:30 +00:00
}
//! sets the state that text was found. Returns true if set should be set
bool XMLParser::_set_text(char *start, char *end) {
2014-02-10 01:10:30 +00:00
// check if text is more than 2 characters, and if not, check if there is
// only white space, so that this text won't be reported
if (end - start < 3) {
char *p = start;
for (; p != end; ++p) {
if (!_is_white_space(*p)) {
2014-02-10 01:10:30 +00:00
break;
}
}
2014-02-10 01:10:30 +00:00
if (p == end) {
2014-02-10 01:10:30 +00:00
return false;
}
2014-02-10 01:10:30 +00:00
}
// set current text to the parsed text, and replace xml special characters
String s = String::utf8(start, (int)(end - start));
node_name = s.xml_unescape();
2014-02-10 01:10:30 +00:00
// current XML node type is text
node_type = NODE_TEXT;
return true;
}
void XMLParser::_parse_closing_xml_element() {
node_type = NODE_ELEMENT_END;
node_empty = false;
attributes.clear();
++P;
const char *pBeginClose = P;
2014-02-10 01:10:30 +00:00
while (*P && *P != '>') {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
node_name = String::utf8(pBeginClose, (int)(P - pBeginClose));
#ifdef DEBUG_XML
print_line("XML CLOSE: " + node_name);
2014-02-10 01:10:30 +00:00
#endif
if (*P) {
++P;
}
2014-02-10 01:10:30 +00:00
}
void XMLParser::_ignore_definition() {
node_type = NODE_UNKNOWN;
char *F = P;
2014-02-10 01:10:30 +00:00
// move until end marked with '>' reached
while (*P && *P != '>') {
2014-02-10 01:10:30 +00:00
++P;
}
node_name.parse_utf8(F, P - F);
if (*P) {
++P;
}
2014-02-10 01:10:30 +00:00
}
bool XMLParser::_parse_cdata() {
if (*(P + 1) != '[') {
2014-02-10 01:10:30 +00:00
return false;
}
2014-02-10 01:10:30 +00:00
node_type = NODE_CDATA;
// skip '<![CDATA['
int count = 0;
while (*P && count < 8) {
2014-02-10 01:10:30 +00:00
++P;
++count;
}
if (!*P) {
node_name = "";
2014-02-10 01:10:30 +00:00
return true;
}
2014-02-10 01:10:30 +00:00
char *cDataBegin = P;
char *cDataEnd = nullptr;
2014-02-10 01:10:30 +00:00
// find end of CDATA
while (*P && !cDataEnd) {
2014-02-10 01:10:30 +00:00
if (*P == '>' &&
(*(P - 1) == ']') &&
(*(P - 2) == ']')) {
2014-02-10 01:10:30 +00:00
cDataEnd = P - 2;
}
++P;
}
if (cDataEnd) {
cDataEnd = P;
}
node_name = String::utf8(cDataBegin, (int)(cDataEnd - cDataBegin));
2014-02-10 01:10:30 +00:00
#ifdef DEBUG_XML
print_line("XML CDATA: " + node_name);
2014-02-10 01:10:30 +00:00
#endif
return true;
}
void XMLParser::_parse_comment() {
node_type = NODE_COMMENT;
P += 1;
char *pEndOfInput = data + length;
char *pCommentBegin;
char *pCommentEnd;
2014-02-10 01:10:30 +00:00
if (P + 1 < pEndOfInput && P[0] == '-' && P[1] == '-') {
// Comment, use '-->' as end.
pCommentBegin = P + 2;
for (pCommentEnd = pCommentBegin; pCommentEnd + 2 < pEndOfInput; pCommentEnd++) {
if (pCommentEnd[0] == '-' && pCommentEnd[1] == '-' && pCommentEnd[2] == '>') {
break;
}
}
if (pCommentEnd + 2 < pEndOfInput) {
P = pCommentEnd + 3;
} else {
P = pCommentEnd = pEndOfInput;
}
} else {
// Like document type definition, match angle brackets.
pCommentBegin = P;
int count = 1;
while (*P && count) {
if (*P == '>') {
--count;
} else if (*P == '<') {
++count;
}
++P;
}
2014-02-10 01:10:30 +00:00
if (count) {
pCommentEnd = P;
} else {
pCommentEnd = P - 1;
}
2014-02-10 01:10:30 +00:00
}
node_name = String::utf8(pCommentBegin, (int)(pCommentEnd - pCommentBegin));
2014-02-10 01:10:30 +00:00
#ifdef DEBUG_XML
print_line("XML COMMENT: " + node_name);
2014-02-10 01:10:30 +00:00
#endif
}
void XMLParser::_parse_opening_xml_element() {
node_type = NODE_ELEMENT;
node_empty = false;
attributes.clear();
// find name
const char *startName = P;
2014-02-10 01:10:30 +00:00
// find end of element
while (*P && *P != '>' && !_is_white_space(*P)) {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
const char *endName = P;
2014-02-10 01:10:30 +00:00
// find attributes
while (*P && *P != '>') {
if (_is_white_space(*P)) {
2014-02-10 01:10:30 +00:00
++P;
} else {
if (*P != '/') {
2014-02-10 01:10:30 +00:00
// we've got an attribute
// read the attribute names
const char *attributeNameBegin = P;
2014-02-10 01:10:30 +00:00
while (*P && !_is_white_space(*P) && *P != '=') {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
if (!*P) {
break;
}
const char *attributeNameEnd = P;
2014-02-10 01:10:30 +00:00
++P;
// read the attribute value
// check for quotes and single quotes, thx to murphy
while ((*P != '\"') && (*P != '\'') && *P) {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
if (!*P) { // malformatted xml file
break;
}
2014-02-10 01:10:30 +00:00
const char attributeQuoteChar = *P;
++P;
const char *attributeValueBegin = P;
2014-02-10 01:10:30 +00:00
while (*P != attributeQuoteChar && *P) {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
const char *attributeValueEnd = P;
if (*P) {
++P;
}
2014-02-10 01:10:30 +00:00
Attribute attr;
attr.name = String::utf8(attributeNameBegin,
(int)(attributeNameEnd - attributeNameBegin));
2014-02-10 01:10:30 +00:00
String s = String::utf8(attributeValueBegin,
(int)(attributeValueEnd - attributeValueBegin));
2014-02-10 01:10:30 +00:00
attr.value = s.xml_unescape();
2014-02-10 01:10:30 +00:00
attributes.push_back(attr);
} else {
2014-02-10 01:10:30 +00:00
// tag is closed directly
++P;
node_empty = true;
break;
}
}
}
// check if this tag is closing directly
if (endName > startName && *(endName - 1) == '/') {
2014-02-10 01:10:30 +00:00
// directly closing tag
node_empty = true;
endName--;
}
node_name = String::utf8(startName, (int)(endName - startName));
#ifdef DEBUG_XML
print_line("XML OPEN: " + node_name);
2014-02-10 01:10:30 +00:00
#endif
if (*P) {
++P;
}
2014-02-10 01:10:30 +00:00
}
void XMLParser::_parse_current_node() {
char *start = P;
2014-02-10 01:10:30 +00:00
node_offset = P - data;
// more forward until '<' found
while (*P != '<' && *P) {
2014-02-10 01:10:30 +00:00
++P;
}
2014-02-10 01:10:30 +00:00
if (P - start > 0) {
2014-02-10 01:10:30 +00:00
// we found some text, store it
if (_set_text(start, P)) {
2014-02-10 01:10:30 +00:00
return;
}
2014-02-10 01:10:30 +00:00
}
if (!*P) {
return;
}
2014-02-10 01:10:30 +00:00
++P;
// based on current token, parse and report next element
switch (*P) {
case '/':
_parse_closing_xml_element();
break;
case '?':
_ignore_definition();
break;
case '!':
if (!_parse_cdata()) {
_parse_comment();
}
break;
default:
_parse_opening_xml_element();
break;
2014-02-10 01:10:30 +00:00
}
}
uint64_t XMLParser::get_node_offset() const {
return node_offset;
};
Error XMLParser::seek(uint64_t p_pos) {
ERR_FAIL_COND_V(!data, ERR_FILE_EOF);
2014-02-10 01:10:30 +00:00
ERR_FAIL_COND_V(p_pos >= length, ERR_FILE_EOF);
P = data + p_pos;
return read();
};
void XMLParser::_bind_methods() {
ClassDB::bind_method(D_METHOD("read"), &XMLParser::read);
ClassDB::bind_method(D_METHOD("get_node_type"), &XMLParser::get_node_type);
ClassDB::bind_method(D_METHOD("get_node_name"), &XMLParser::get_node_name);
ClassDB::bind_method(D_METHOD("get_node_data"), &XMLParser::get_node_data);
ClassDB::bind_method(D_METHOD("get_node_offset"), &XMLParser::get_node_offset);
ClassDB::bind_method(D_METHOD("get_attribute_count"), &XMLParser::get_attribute_count);
ClassDB::bind_method(D_METHOD("get_attribute_name", "idx"), &XMLParser::get_attribute_name);
ClassDB::bind_method(D_METHOD("get_attribute_value", "idx"), (String(XMLParser::*)(int) const) & XMLParser::get_attribute_value);
ClassDB::bind_method(D_METHOD("has_attribute", "name"), &XMLParser::has_attribute);
ClassDB::bind_method(D_METHOD("get_named_attribute_value", "name"), (String(XMLParser::*)(const String &) const) & XMLParser::get_attribute_value);
ClassDB::bind_method(D_METHOD("get_named_attribute_value_safe", "name"), &XMLParser::get_attribute_value_safe);
ClassDB::bind_method(D_METHOD("is_empty"), &XMLParser::is_empty);
ClassDB::bind_method(D_METHOD("get_current_line"), &XMLParser::get_current_line);
ClassDB::bind_method(D_METHOD("skip_section"), &XMLParser::skip_section);
ClassDB::bind_method(D_METHOD("seek", "position"), &XMLParser::seek);
ClassDB::bind_method(D_METHOD("open", "file"), &XMLParser::open);
ClassDB::bind_method(D_METHOD("open_buffer", "buffer"), &XMLParser::open_buffer);
BIND_ENUM_CONSTANT(NODE_NONE);
BIND_ENUM_CONSTANT(NODE_ELEMENT);
BIND_ENUM_CONSTANT(NODE_ELEMENT_END);
BIND_ENUM_CONSTANT(NODE_TEXT);
BIND_ENUM_CONSTANT(NODE_COMMENT);
BIND_ENUM_CONSTANT(NODE_CDATA);
BIND_ENUM_CONSTANT(NODE_UNKNOWN);
2014-02-10 01:10:30 +00:00
};
Error XMLParser::read() {
// if not end reached, parse the node
if (P && (P - data) < (int64_t)length - 1 && *P != 0) {
2014-02-10 01:10:30 +00:00
_parse_current_node();
return OK;
}
return ERR_FILE_EOF;
}
XMLParser::NodeType XMLParser::get_node_type() {
return node_type;
}
String XMLParser::get_node_data() const {
ERR_FAIL_COND_V(node_type != NODE_TEXT, "");
2014-02-10 01:10:30 +00:00
return node_name;
}
String XMLParser::get_node_name() const {
ERR_FAIL_COND_V(node_type == NODE_TEXT, "");
2014-02-10 01:10:30 +00:00
return node_name;
}
int XMLParser::get_attribute_count() const {
return attributes.size();
}
String XMLParser::get_attribute_name(int p_idx) const {
ERR_FAIL_INDEX_V(p_idx, attributes.size(), "");
2014-02-10 01:10:30 +00:00
return attributes[p_idx].name;
}
String XMLParser::get_attribute_value(int p_idx) const {
ERR_FAIL_INDEX_V(p_idx, attributes.size(), "");
2014-02-10 01:10:30 +00:00
return attributes[p_idx].value;
}
bool XMLParser::has_attribute(const String &p_name) const {
for (int i = 0; i < attributes.size(); i++) {
if (attributes[i].name == p_name) {
2014-02-10 01:10:30 +00:00
return true;
}
2014-02-10 01:10:30 +00:00
}
return false;
}
String XMLParser::get_attribute_value(const String &p_name) const {
int idx = -1;
for (int i = 0; i < attributes.size(); i++) {
if (attributes[i].name == p_name) {
idx = i;
2014-02-10 01:10:30 +00:00
break;
}
}
ERR_FAIL_COND_V_MSG(idx < 0, "", "Attribute not found: " + p_name + ".");
2014-02-10 01:10:30 +00:00
return attributes[idx].value;
}
String XMLParser::get_attribute_value_safe(const String &p_name) const {
int idx = -1;
for (int i = 0; i < attributes.size(); i++) {
if (attributes[i].name == p_name) {
idx = i;
2014-02-10 01:10:30 +00:00
break;
}
}
if (idx < 0) {
2014-02-10 01:10:30 +00:00
return "";
}
2014-02-10 01:10:30 +00:00
return attributes[idx].value;
}
bool XMLParser::is_empty() const {
return node_empty;
}
Error XMLParser::open_buffer(const Vector<uint8_t> &p_buffer) {
ERR_FAIL_COND_V(p_buffer.size() == 0, ERR_INVALID_DATA);
if (data) {
memdelete_arr(data);
}
length = p_buffer.size();
data = memnew_arr(char, length + 1);
memcpy(data, p_buffer.ptr(), length);
data[length] = 0;
P = data;
return OK;
}
Error XMLParser::open(const String &p_path) {
2014-02-10 01:10:30 +00:00
Error err;
FileAccess *file = FileAccess::open(p_path, FileAccess::READ, &err);
2014-02-10 01:10:30 +00:00
2019-09-25 08:28:50 +00:00
ERR_FAIL_COND_V_MSG(err != OK, err, "Cannot open file '" + p_path + "'.");
2014-02-10 01:10:30 +00:00
length = file->get_len();
ERR_FAIL_COND_V(length < 1, ERR_FILE_CORRUPT);
2014-02-10 01:10:30 +00:00
if (data) {
memdelete_arr(data);
}
data = memnew_arr(char, length + 1);
file->get_buffer((uint8_t *)data, length);
data[length] = 0;
P = data;
2014-02-10 01:10:30 +00:00
memdelete(file);
return OK;
}
void XMLParser::skip_section() {
// skip if this element is empty anyway.
if (is_empty()) {
2014-02-10 01:10:30 +00:00
return;
}
2014-02-10 01:10:30 +00:00
// read until we've reached the last element in this section
int tagcount = 1;
while (tagcount && read() == OK) {
2014-02-10 01:10:30 +00:00
if (get_node_type() == XMLParser::NODE_ELEMENT &&
!is_empty()) {
2014-02-10 01:10:30 +00:00
++tagcount;
} else if (get_node_type() == XMLParser::NODE_ELEMENT_END) {
2014-02-10 01:10:30 +00:00
--tagcount;
}
2014-02-10 01:10:30 +00:00
}
}
void XMLParser::close() {
if (data) {
2014-02-10 01:10:30 +00:00
memdelete_arr(data);
}
data = nullptr;
length = 0;
P = nullptr;
node_empty = false;
node_type = NODE_NONE;
2014-02-10 01:10:30 +00:00
node_offset = 0;
}
int XMLParser::get_current_line() const {
return 0;
}
XMLParser::XMLParser() {
}
2014-02-10 01:10:30 +00:00
XMLParser::~XMLParser() {
if (data) {
2014-02-10 01:10:30 +00:00
memdelete_arr(data);
}
2014-02-10 01:10:30 +00:00
}