2018-12-17 16:55:11 +00:00
|
|
|
#include <qpdf/JSON.hh>
|
2022-02-04 21:31:31 +00:00
|
|
|
|
2022-05-04 13:12:57 +00:00
|
|
|
#include <qpdf/BufferInputSource.hh>
|
2022-05-04 20:28:12 +00:00
|
|
|
#include <qpdf/Pl_Base64.hh>
|
|
|
|
#include <qpdf/Pl_Concatenate.hh>
|
2022-05-04 12:32:54 +00:00
|
|
|
#include <qpdf/Pl_String.hh>
|
2022-05-21 13:05:54 +00:00
|
|
|
#include <qpdf/QIntC.hh>
|
2018-12-17 16:55:11 +00:00
|
|
|
#include <qpdf/QTC.hh>
|
2022-04-02 21:14:10 +00:00
|
|
|
#include <qpdf/QUtil.hh>
|
2022-01-17 23:40:38 +00:00
|
|
|
#include <cstring>
|
2022-04-02 21:14:10 +00:00
|
|
|
#include <stdexcept>
|
2018-12-17 16:55:11 +00:00
|
|
|
|
2022-05-21 13:05:54 +00:00
|
|
|
template <typename T>
|
|
|
|
static qpdf_offset_t
|
|
|
|
toO(T const& i)
|
|
|
|
{
|
|
|
|
return QIntC::to_offset(i);
|
|
|
|
}
|
|
|
|
|
2022-02-04 15:52:37 +00:00
|
|
|
JSON::Members::Members(std::shared_ptr<JSON_value> value) :
|
2022-05-01 18:06:31 +00:00
|
|
|
value(value),
|
|
|
|
start(0),
|
|
|
|
end(0)
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-02-04 15:52:37 +00:00
|
|
|
JSON::JSON(std::shared_ptr<JSON_value> value) :
|
2018-12-17 16:55:11 +00:00
|
|
|
m(new Members(value))
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::writeClose(Pipeline* p, bool first, size_t depth, char const* delimiter)
|
|
|
|
{
|
2023-01-28 11:09:46 +00:00
|
|
|
if (first) {
|
|
|
|
*p << delimiter;
|
|
|
|
} else {
|
|
|
|
std::string s{"\n"};
|
|
|
|
s.append(2 * depth, ' ');
|
|
|
|
*p << s + delimiter;
|
2022-05-04 12:32:54 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeNext(Pipeline* p, bool& first, size_t depth)
|
|
|
|
{
|
|
|
|
if (first) {
|
|
|
|
first = false;
|
2023-01-28 11:15:27 +00:00
|
|
|
std::string s{"\n"};
|
|
|
|
s.append(2 * depth, ' ');
|
|
|
|
*p << s;
|
2022-05-04 12:32:54 +00:00
|
|
|
} else {
|
2023-01-28 11:15:27 +00:00
|
|
|
std::string s{",\n"};
|
|
|
|
s.append(2 * depth, ' ');
|
|
|
|
*p << s;
|
2022-05-04 12:32:54 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeDictionaryOpen(Pipeline* p, bool& first, size_t depth)
|
|
|
|
{
|
|
|
|
*p << "{";
|
|
|
|
first = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeArrayOpen(Pipeline* p, bool& first, size_t depth)
|
|
|
|
{
|
|
|
|
*p << "[";
|
|
|
|
first = true;
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeDictionaryClose(Pipeline* p, bool first, size_t depth)
|
|
|
|
{
|
|
|
|
writeClose(p, first, depth, "}");
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeArrayClose(Pipeline* p, bool first, size_t depth)
|
|
|
|
{
|
|
|
|
writeClose(p, first, depth, "]");
|
|
|
|
}
|
|
|
|
|
Objects json: write incrementally and in numeric order
The following script was used to adjust test data:
----------
#!/usr/bin/env python3
import json
import sys
import re
def json_dumps(data):
return json.dumps(data, ensure_ascii=False,
indent=2, separators=(',', ': '))
for filename in sys.argv[1:]:
with open(filename, 'r') as f:
data = json.loads(f.read())
if 'objects' not in data:
continue
trailer = None
to_sort = []
for k, v in data['objects'].items():
if k == 'trailer':
trailer = v
else:
m = re.match(r'^(\d+) \d+ R', k)
if m:
to_sort.append([int(m.group(1)), k, v])
newobjects = {x[1]: x[2] for x in sorted(to_sort)}
if trailer is not None:
newobjects['trailer'] = trailer
data['objects'] = newobjects
print(json_dumps(data))
----------
2022-05-06 23:04:20 +00:00
|
|
|
void
|
|
|
|
JSON::writeDictionaryKey(
|
|
|
|
Pipeline* p, bool& first, std::string const& key, size_t depth)
|
|
|
|
{
|
|
|
|
writeNext(p, first, depth);
|
2023-01-28 12:54:49 +00:00
|
|
|
*p << std::string("\"") + key + "\": ";
|
Objects json: write incrementally and in numeric order
The following script was used to adjust test data:
----------
#!/usr/bin/env python3
import json
import sys
import re
def json_dumps(data):
return json.dumps(data, ensure_ascii=False,
indent=2, separators=(',', ': '))
for filename in sys.argv[1:]:
with open(filename, 'r') as f:
data = json.loads(f.read())
if 'objects' not in data:
continue
trailer = None
to_sort = []
for k, v in data['objects'].items():
if k == 'trailer':
trailer = v
else:
m = re.match(r'^(\d+) \d+ R', k)
if m:
to_sort.append([int(m.group(1)), k, v])
newobjects = {x[1]: x[2] for x in sorted(to_sort)}
if trailer is not None:
newobjects['trailer'] = trailer
data['objects'] = newobjects
print(json_dumps(data))
----------
2022-05-06 23:04:20 +00:00
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::writeDictionaryItem(
|
|
|
|
Pipeline* p,
|
|
|
|
bool& first,
|
|
|
|
std::string const& key,
|
|
|
|
JSON const& value,
|
|
|
|
size_t depth)
|
|
|
|
{
|
Objects json: write incrementally and in numeric order
The following script was used to adjust test data:
----------
#!/usr/bin/env python3
import json
import sys
import re
def json_dumps(data):
return json.dumps(data, ensure_ascii=False,
indent=2, separators=(',', ': '))
for filename in sys.argv[1:]:
with open(filename, 'r') as f:
data = json.loads(f.read())
if 'objects' not in data:
continue
trailer = None
to_sort = []
for k, v in data['objects'].items():
if k == 'trailer':
trailer = v
else:
m = re.match(r'^(\d+) \d+ R', k)
if m:
to_sort.append([int(m.group(1)), k, v])
newobjects = {x[1]: x[2] for x in sorted(to_sort)}
if trailer is not None:
newobjects['trailer'] = trailer
data['objects'] = newobjects
print(json_dumps(data))
----------
2022-05-06 23:04:20 +00:00
|
|
|
writeDictionaryKey(p, first, key, depth);
|
2022-07-30 17:32:59 +00:00
|
|
|
value.write(p, depth);
|
2022-05-04 12:32:54 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::writeArrayItem(
|
|
|
|
Pipeline* p, bool& first, JSON const& element, size_t depth)
|
|
|
|
{
|
|
|
|
writeNext(p, first, depth);
|
2022-07-30 17:32:59 +00:00
|
|
|
element.write(p, depth);
|
2022-05-04 12:32:54 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::JSON_dictionary::write(Pipeline* p, size_t depth) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
|
|
|
bool first = true;
|
2022-05-04 12:32:54 +00:00
|
|
|
writeDictionaryOpen(p, first, depth);
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& iter: members) {
|
2022-07-30 17:32:59 +00:00
|
|
|
writeDictionaryItem(p, first, iter.first, iter.second, 1 + depth);
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
2022-05-04 12:32:54 +00:00
|
|
|
writeDictionaryClose(p, first, depth);
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::JSON_array::write(Pipeline* p, size_t depth) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
|
|
|
bool first = true;
|
2022-05-04 12:32:54 +00:00
|
|
|
writeArrayOpen(p, first, depth);
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& element: elements) {
|
2022-07-30 17:32:59 +00:00
|
|
|
writeArrayItem(p, first, element, 1 + depth);
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
2022-05-04 12:32:54 +00:00
|
|
|
writeArrayClose(p, first, depth);
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON::JSON_string::JSON_string(std::string const& utf8) :
|
2022-01-19 14:31:28 +00:00
|
|
|
utf8(utf8),
|
2018-12-17 16:55:11 +00:00
|
|
|
encoded(encode_string(utf8))
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::JSON_string::write(Pipeline* p, size_t) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
2023-01-28 12:59:26 +00:00
|
|
|
*p << std::string("\"") + encoded + "\"";
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON::JSON_number::JSON_number(long long value) :
|
2022-09-21 16:49:21 +00:00
|
|
|
encoded(std::to_string(value))
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
JSON::JSON_number::JSON_number(double value) :
|
|
|
|
encoded(QUtil::double_to_string(value, 6))
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
JSON::JSON_number::JSON_number(std::string const& value) :
|
|
|
|
encoded(value)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::JSON_number::write(Pipeline* p, size_t) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
2022-05-04 12:32:54 +00:00
|
|
|
*p << encoded;
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON::JSON_bool::JSON_bool(bool val) :
|
|
|
|
value(val)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::JSON_bool::write(Pipeline* p, size_t) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
2022-05-04 12:32:54 +00:00
|
|
|
*p << (value ? "true" : "false");
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::JSON_null::write(Pipeline* p, size_t) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
2022-05-04 12:32:54 +00:00
|
|
|
*p << "null";
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
2022-05-04 20:28:12 +00:00
|
|
|
JSON::JSON_blob::JSON_blob(std::function<void(Pipeline*)> fn) :
|
|
|
|
fn(fn)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSON::JSON_blob::write(Pipeline* p, size_t) const
|
|
|
|
{
|
|
|
|
*p << "\"";
|
|
|
|
Pl_Concatenate cat("blob concatenate", p);
|
|
|
|
Pl_Base64 base64("blob base64", &cat, Pl_Base64::a_encode);
|
|
|
|
fn(&base64);
|
|
|
|
base64.finish();
|
|
|
|
*p << "\"";
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
void
|
|
|
|
JSON::write(Pipeline* p, size_t depth) const
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
2022-08-07 09:33:25 +00:00
|
|
|
if (nullptr == this->m->value) {
|
2022-05-04 12:32:54 +00:00
|
|
|
*p << "null";
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-05-04 12:32:54 +00:00
|
|
|
this->m->value->write(p, depth);
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-05-04 12:32:54 +00:00
|
|
|
std::string
|
|
|
|
JSON::unparse() const
|
|
|
|
{
|
|
|
|
std::string s;
|
2022-05-06 21:37:25 +00:00
|
|
|
Pl_String p("unparse", nullptr, s);
|
2022-05-04 12:32:54 +00:00
|
|
|
write(&p, 0);
|
|
|
|
return s;
|
|
|
|
}
|
|
|
|
|
2018-12-17 16:55:11 +00:00
|
|
|
std::string
|
|
|
|
JSON::encode_string(std::string const& str)
|
|
|
|
{
|
|
|
|
std::string result;
|
|
|
|
size_t len = str.length();
|
2022-04-02 21:14:10 +00:00
|
|
|
for (size_t i = 0; i < len; ++i) {
|
2018-12-17 16:55:11 +00:00
|
|
|
unsigned char ch = static_cast<unsigned char>(str.at(i));
|
2022-04-02 21:14:10 +00:00
|
|
|
switch (ch) {
|
|
|
|
case '\\':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\\\";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\"':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\\"";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\b':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\b";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\f':
|
2022-01-17 23:38:12 +00:00
|
|
|
result += "\\f";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\n':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\n";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\r':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\r";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case '\t':
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\t";
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
default:
|
|
|
|
if (ch < 32) {
|
2018-12-17 16:55:11 +00:00
|
|
|
result += "\\u" + QUtil::int_to_string_base(ch, 16, 4);
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2019-06-21 03:35:23 +00:00
|
|
|
result.append(1, static_cast<char>(ch));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeDictionary()
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_dictionary>());
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::addDictionaryMember(std::string const& key, JSON const& val)
|
|
|
|
{
|
2023-01-28 14:32:01 +00:00
|
|
|
if (auto* obj = dynamic_cast<JSON_dictionary*>(this->m->value.get())) {
|
|
|
|
return obj->members[encode_string(key)] =
|
|
|
|
val.m->value ? val.m->value : std::make_shared<JSON_null>();
|
|
|
|
} else {
|
2018-12-17 16:55:11 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON::addDictionaryMember called on non-dictionary");
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-05-20 13:57:30 +00:00
|
|
|
bool
|
|
|
|
JSON::checkDictionaryKeySeen(std::string const& key)
|
|
|
|
{
|
|
|
|
JSON_dictionary* obj = dynamic_cast<JSON_dictionary*>(this->m->value.get());
|
2022-07-26 11:37:50 +00:00
|
|
|
if (nullptr == obj) {
|
2022-05-20 13:57:30 +00:00
|
|
|
throw std::logic_error(
|
|
|
|
"JSON::checkDictionaryKey called on non-dictionary");
|
|
|
|
}
|
|
|
|
if (obj->parsed_keys.count(key)) {
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
obj->parsed_keys.insert(key);
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2018-12-17 16:55:11 +00:00
|
|
|
JSON
|
|
|
|
JSON::makeArray()
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_array>());
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::addArrayElement(JSON const& val)
|
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
JSON_array* arr = dynamic_cast<JSON_array*>(this->m->value.get());
|
2022-07-26 11:37:50 +00:00
|
|
|
if (nullptr == arr) {
|
2018-12-17 16:55:11 +00:00
|
|
|
throw std::runtime_error("JSON::addArrayElement called on non-array");
|
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
if (val.m->value.get()) {
|
2018-12-17 16:55:11 +00:00
|
|
|
arr->elements.push_back(val.m->value);
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-02-04 15:52:37 +00:00
|
|
|
arr->elements.push_back(std::make_shared<JSON_null>());
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
return arr->elements.back();
|
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeString(std::string const& utf8)
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_string>(utf8));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeInt(long long int value)
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_number>(value));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeReal(double value)
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_number>(value));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeNumber(std::string const& encoded)
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_number>(encoded));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeBool(bool value)
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_bool>(value));
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::makeNull()
|
|
|
|
{
|
2022-02-04 15:52:37 +00:00
|
|
|
return JSON(std::make_shared<JSON_null>());
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
2022-05-04 20:28:12 +00:00
|
|
|
JSON
|
|
|
|
JSON::makeBlob(std::function<void(Pipeline*)> fn)
|
|
|
|
{
|
|
|
|
return JSON(std::make_shared<JSON_blob>(fn));
|
|
|
|
}
|
|
|
|
|
2022-01-17 23:38:05 +00:00
|
|
|
bool
|
|
|
|
JSON::isArray() const
|
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
return nullptr != dynamic_cast<JSON_array const*>(this->m->value.get());
|
2022-01-17 23:38:05 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::isDictionary() const
|
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
return nullptr !=
|
|
|
|
dynamic_cast<JSON_dictionary const*>(this->m->value.get());
|
2022-01-17 23:38:05 +00:00
|
|
|
}
|
|
|
|
|
2022-01-19 14:31:28 +00:00
|
|
|
bool
|
|
|
|
JSON::getString(std::string& utf8) const
|
|
|
|
{
|
2022-02-04 15:10:19 +00:00
|
|
|
auto v = dynamic_cast<JSON_string const*>(this->m->value.get());
|
2022-04-02 21:14:10 +00:00
|
|
|
if (v == nullptr) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
utf8 = v->utf8;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::getNumber(std::string& value) const
|
|
|
|
{
|
2022-02-04 15:10:19 +00:00
|
|
|
auto v = dynamic_cast<JSON_number const*>(this->m->value.get());
|
2022-04-02 21:14:10 +00:00
|
|
|
if (v == nullptr) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
value = v->encoded;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::getBool(bool& value) const
|
|
|
|
{
|
2022-02-04 15:10:19 +00:00
|
|
|
auto v = dynamic_cast<JSON_bool const*>(this->m->value.get());
|
2022-04-02 21:14:10 +00:00
|
|
|
if (v == nullptr) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return false;
|
|
|
|
}
|
|
|
|
value = v->value;
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::isNull() const
|
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
if (dynamic_cast<JSON_null const*>(this->m->value.get())) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return true;
|
|
|
|
}
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::forEachDictItem(
|
|
|
|
std::function<void(std::string const& key, JSON value)> fn) const
|
|
|
|
{
|
2022-02-04 15:10:19 +00:00
|
|
|
auto v = dynamic_cast<JSON_dictionary const*>(this->m->value.get());
|
2022-04-02 21:14:10 +00:00
|
|
|
if (v == nullptr) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return false;
|
|
|
|
}
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& k: v->members) {
|
2022-01-19 14:31:28 +00:00
|
|
|
fn(k.first, JSON(k.second));
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
|
|
|
bool
|
|
|
|
JSON::forEachArrayItem(std::function<void(JSON value)> fn) const
|
|
|
|
{
|
2022-02-04 15:10:19 +00:00
|
|
|
auto v = dynamic_cast<JSON_array const*>(this->m->value.get());
|
2022-04-02 21:14:10 +00:00
|
|
|
if (v == nullptr) {
|
2022-01-19 14:31:28 +00:00
|
|
|
return false;
|
|
|
|
}
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& i: v->elements) {
|
2022-01-19 14:31:28 +00:00
|
|
|
fn(JSON(i));
|
|
|
|
}
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2018-12-17 16:55:11 +00:00
|
|
|
bool
|
|
|
|
JSON::checkSchema(JSON schema, std::list<std::string>& errors)
|
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
return checkSchemaInternal(
|
|
|
|
this->m->value.get(), schema.m->value.get(), 0, errors, "");
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
|
2022-01-22 19:33:26 +00:00
|
|
|
bool
|
2022-04-02 21:14:10 +00:00
|
|
|
JSON::checkSchema(
|
|
|
|
JSON schema, unsigned long flags, std::list<std::string>& errors)
|
2022-01-22 19:33:26 +00:00
|
|
|
{
|
2022-04-02 21:14:10 +00:00
|
|
|
return checkSchemaInternal(
|
|
|
|
this->m->value.get(), schema.m->value.get(), flags, errors, "");
|
2022-01-22 19:33:26 +00:00
|
|
|
}
|
2018-12-17 16:55:11 +00:00
|
|
|
|
|
|
|
bool
|
2022-04-02 21:14:10 +00:00
|
|
|
JSON::checkSchemaInternal(
|
|
|
|
JSON_value* this_v,
|
|
|
|
JSON_value* sch_v,
|
|
|
|
unsigned long flags,
|
|
|
|
std::list<std::string>& errors,
|
|
|
|
std::string prefix)
|
2018-12-17 16:55:11 +00:00
|
|
|
{
|
|
|
|
JSON_array* this_arr = dynamic_cast<JSON_array*>(this_v);
|
|
|
|
JSON_dictionary* this_dict = dynamic_cast<JSON_dictionary*>(this_v);
|
|
|
|
|
|
|
|
JSON_array* sch_arr = dynamic_cast<JSON_array*>(sch_v);
|
|
|
|
JSON_dictionary* sch_dict = dynamic_cast<JSON_dictionary*>(sch_v);
|
|
|
|
|
2022-01-22 19:33:26 +00:00
|
|
|
JSON_string* sch_str = dynamic_cast<JSON_string*>(sch_v);
|
|
|
|
|
2018-12-17 16:55:11 +00:00
|
|
|
std::string err_prefix;
|
2022-04-02 21:14:10 +00:00
|
|
|
if (prefix.empty()) {
|
2018-12-17 16:55:11 +00:00
|
|
|
err_prefix = "top-level object";
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2018-12-17 16:55:11 +00:00
|
|
|
err_prefix = "json key \"" + prefix + "\"";
|
|
|
|
}
|
|
|
|
|
2020-04-04 21:31:07 +00:00
|
|
|
std::string pattern_key;
|
2022-04-02 21:14:10 +00:00
|
|
|
if (sch_dict) {
|
|
|
|
if (!this_dict) {
|
2018-12-17 16:55:11 +00:00
|
|
|
QTC::TC("libtests", "JSON wanted dictionary");
|
|
|
|
errors.push_back(err_prefix + " is supposed to be a dictionary");
|
|
|
|
return false;
|
|
|
|
}
|
2020-04-04 21:31:07 +00:00
|
|
|
auto members = sch_dict->members;
|
|
|
|
std::string key;
|
|
|
|
if ((members.size() == 1) &&
|
|
|
|
((key = members.begin()->first, key.length() > 2) &&
|
2022-04-02 21:14:10 +00:00
|
|
|
(key.at(0) == '<') && (key.at(key.length() - 1) == '>'))) {
|
2020-04-04 21:31:07 +00:00
|
|
|
pattern_key = key;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
if (sch_dict && (!pattern_key.empty())) {
|
2022-02-04 15:10:19 +00:00
|
|
|
auto pattern_schema = sch_dict->members[pattern_key].get();
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& iter: this_dict->members) {
|
2020-04-04 21:31:07 +00:00
|
|
|
std::string const& key = iter.first;
|
|
|
|
checkSchemaInternal(
|
2022-04-02 21:14:10 +00:00
|
|
|
this_dict->members[key].get(),
|
|
|
|
pattern_schema,
|
|
|
|
flags,
|
|
|
|
errors,
|
|
|
|
prefix + "." + key);
|
2020-04-04 21:31:07 +00:00
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (sch_dict) {
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto& iter: sch_dict->members) {
|
2022-01-22 19:33:26 +00:00
|
|
|
std::string const& key = iter.first;
|
2022-04-02 21:14:10 +00:00
|
|
|
if (this_dict->members.count(key)) {
|
2018-12-17 16:55:11 +00:00
|
|
|
checkSchemaInternal(
|
2022-02-04 15:10:19 +00:00
|
|
|
this_dict->members[key].get(),
|
|
|
|
iter.second.get(),
|
2022-04-02 21:14:10 +00:00
|
|
|
flags,
|
|
|
|
errors,
|
|
|
|
prefix + "." + key);
|
|
|
|
} else {
|
|
|
|
if (flags & f_optional) {
|
2022-01-22 19:33:26 +00:00
|
|
|
QTC::TC("libtests", "JSON optional key");
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-22 19:33:26 +00:00
|
|
|
QTC::TC("libtests", "JSON key missing in object");
|
|
|
|
errors.push_back(
|
|
|
|
err_prefix + ": key \"" + key +
|
|
|
|
"\" is present in schema but missing in object");
|
|
|
|
}
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
|
|
|
}
|
2022-04-30 13:43:07 +00:00
|
|
|
for (auto const& iter: this_dict->members) {
|
2022-02-04 15:52:37 +00:00
|
|
|
std::string const& key = iter.first;
|
2022-04-02 21:14:10 +00:00
|
|
|
if (sch_dict->members.count(key) == 0) {
|
2018-12-17 16:55:11 +00:00
|
|
|
QTC::TC("libtests", "JSON key extra in object");
|
|
|
|
errors.push_back(
|
|
|
|
err_prefix + ": key \"" + key +
|
|
|
|
"\" is not present in schema but appears in object");
|
|
|
|
}
|
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (sch_arr) {
|
2022-07-24 20:44:51 +00:00
|
|
|
auto n_elements = sch_arr->elements.size();
|
|
|
|
if (n_elements == 1) {
|
|
|
|
// A single-element array in the schema allows a single
|
|
|
|
// element in the object or a variable-length array, each
|
|
|
|
// of whose items must conform to the single element of
|
|
|
|
// the schema array. This doesn't apply to arrays of
|
|
|
|
// arrays -- we fall back to the behavior of allowing a
|
|
|
|
// single item only when the object is not an array.
|
|
|
|
if (this_arr) {
|
|
|
|
int i = 0;
|
|
|
|
for (auto const& element: this_arr->elements) {
|
|
|
|
checkSchemaInternal(
|
|
|
|
element.get(),
|
|
|
|
sch_arr->elements.at(0).get(),
|
|
|
|
flags,
|
|
|
|
errors,
|
2022-09-21 16:49:21 +00:00
|
|
|
prefix + "." + std::to_string(i));
|
2022-07-24 20:44:51 +00:00
|
|
|
++i;
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
QTC::TC("libtests", "JSON schema array for single item");
|
|
|
|
checkSchemaInternal(
|
|
|
|
this_v,
|
|
|
|
sch_arr->elements.at(0).get(),
|
|
|
|
flags,
|
|
|
|
errors,
|
|
|
|
prefix);
|
|
|
|
}
|
|
|
|
} else if (!this_arr || (this_arr->elements.size() != n_elements)) {
|
|
|
|
QTC::TC("libtests", "JSON schema array length mismatch");
|
2022-04-02 21:14:10 +00:00
|
|
|
errors.push_back(
|
2022-07-24 20:44:51 +00:00
|
|
|
err_prefix + " is supposed to be an array of length " +
|
2022-09-21 16:49:21 +00:00
|
|
|
std::to_string(n_elements));
|
2018-12-17 16:55:11 +00:00
|
|
|
return false;
|
2022-07-24 20:44:51 +00:00
|
|
|
} else {
|
|
|
|
// A multi-element array in the schema must correspond to
|
|
|
|
// an element of the same length in the object. Each
|
|
|
|
// element in the object is validated against the
|
|
|
|
// corresponding element in the schema.
|
|
|
|
size_t i = 0;
|
2022-07-24 20:17:03 +00:00
|
|
|
for (auto const& element: this_arr->elements) {
|
|
|
|
checkSchemaInternal(
|
|
|
|
element.get(),
|
2022-07-24 20:44:51 +00:00
|
|
|
sch_arr->elements.at(i).get(),
|
2022-07-24 20:17:03 +00:00
|
|
|
flags,
|
|
|
|
errors,
|
2022-09-21 16:49:21 +00:00
|
|
|
prefix + "." + std::to_string(i));
|
2022-07-24 20:17:03 +00:00
|
|
|
++i;
|
|
|
|
}
|
2018-12-17 16:55:11 +00:00
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (!sch_str) {
|
2022-01-22 19:33:26 +00:00
|
|
|
QTC::TC("libtests", "JSON schema other type");
|
2022-04-02 21:14:10 +00:00
|
|
|
errors.push_back(
|
|
|
|
err_prefix + " schema value is not dictionary, array, or string");
|
2022-01-22 19:33:26 +00:00
|
|
|
return false;
|
|
|
|
}
|
2018-12-17 16:55:11 +00:00
|
|
|
|
|
|
|
return errors.empty();
|
|
|
|
}
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
namespace
|
|
|
|
{
|
2022-01-17 23:40:38 +00:00
|
|
|
class JSONParser
|
|
|
|
{
|
|
|
|
public:
|
2022-05-04 13:12:57 +00:00
|
|
|
JSONParser(InputSource& is, JSON::Reactor* reactor) :
|
|
|
|
is(is),
|
2022-05-01 18:06:31 +00:00
|
|
|
reactor(reactor),
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state(ls_top),
|
2022-05-04 13:12:57 +00:00
|
|
|
bytes(0),
|
|
|
|
p(buf),
|
|
|
|
u_count(0),
|
|
|
|
offset(0),
|
|
|
|
done(false),
|
2022-05-20 13:57:30 +00:00
|
|
|
parser_state(ps_top),
|
|
|
|
dict_key_offset(0)
|
2022-01-17 23:40:38 +00:00
|
|
|
{
|
|
|
|
}
|
|
|
|
|
2022-05-04 13:12:57 +00:00
|
|
|
std::shared_ptr<JSON> parse();
|
2022-01-17 23:40:38 +00:00
|
|
|
|
|
|
|
private:
|
|
|
|
void getToken();
|
|
|
|
void handleToken();
|
2023-01-24 00:44:16 +00:00
|
|
|
void numberError();
|
2022-05-20 00:28:13 +00:00
|
|
|
static std::string
|
2022-05-21 13:05:54 +00:00
|
|
|
decode_string(std::string const& json, qpdf_offset_t offset);
|
2022-05-20 00:28:13 +00:00
|
|
|
static void handle_u_code(
|
|
|
|
char const* s,
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t offset,
|
|
|
|
qpdf_offset_t i,
|
2022-05-20 00:28:13 +00:00
|
|
|
unsigned long& high_surrogate,
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t& high_offset,
|
2022-05-20 00:28:13 +00:00
|
|
|
std::string& result);
|
2022-01-17 23:40:38 +00:00
|
|
|
|
|
|
|
enum parser_state_e {
|
|
|
|
ps_top,
|
|
|
|
ps_dict_begin,
|
|
|
|
ps_dict_after_key,
|
|
|
|
ps_dict_after_colon,
|
|
|
|
ps_dict_after_item,
|
|
|
|
ps_dict_after_comma,
|
|
|
|
ps_array_begin,
|
|
|
|
ps_array_after_item,
|
|
|
|
ps_array_after_comma,
|
|
|
|
ps_done,
|
|
|
|
};
|
|
|
|
|
|
|
|
enum lex_state_e {
|
|
|
|
ls_top,
|
|
|
|
ls_number,
|
2023-01-24 11:47:06 +00:00
|
|
|
ls_number_minus,
|
2023-01-24 11:22:06 +00:00
|
|
|
ls_number_leading_zero,
|
2023-01-23 19:51:27 +00:00
|
|
|
ls_number_before_point,
|
2023-01-23 20:08:11 +00:00
|
|
|
ls_number_point,
|
2023-01-24 00:34:22 +00:00
|
|
|
ls_number_after_point,
|
2023-01-24 01:36:38 +00:00
|
|
|
ls_number_e,
|
2023-01-24 13:04:31 +00:00
|
|
|
ls_number_e_sign,
|
2022-01-17 23:40:38 +00:00
|
|
|
ls_alpha,
|
|
|
|
ls_string,
|
|
|
|
ls_backslash,
|
2022-05-04 13:12:57 +00:00
|
|
|
ls_u4,
|
2023-01-26 13:21:45 +00:00
|
|
|
ls_begin_array,
|
|
|
|
ls_end_array,
|
|
|
|
ls_begin_dict,
|
|
|
|
ls_end_dict,
|
|
|
|
ls_colon,
|
|
|
|
ls_comma,
|
2022-01-17 23:40:38 +00:00
|
|
|
};
|
|
|
|
|
2023-01-27 14:37:25 +00:00
|
|
|
struct StackFrame
|
|
|
|
{
|
|
|
|
StackFrame(parser_state_e state, std::shared_ptr<JSON>& item) :
|
|
|
|
state(state),
|
|
|
|
item(item)
|
|
|
|
{
|
|
|
|
}
|
|
|
|
|
|
|
|
parser_state_e state;
|
|
|
|
std::shared_ptr<JSON> item;
|
|
|
|
};
|
|
|
|
|
2022-05-04 13:12:57 +00:00
|
|
|
InputSource& is;
|
2022-05-01 18:06:31 +00:00
|
|
|
JSON::Reactor* reactor;
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state_e lex_state;
|
2022-05-04 13:12:57 +00:00
|
|
|
char buf[16384];
|
|
|
|
size_t bytes;
|
2022-01-17 23:40:38 +00:00
|
|
|
char const* p;
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t u_count;
|
|
|
|
qpdf_offset_t offset;
|
2022-05-04 13:12:57 +00:00
|
|
|
bool done;
|
|
|
|
std::string token;
|
2023-01-26 11:36:44 +00:00
|
|
|
qpdf_offset_t token_start{0};
|
2022-01-17 23:40:38 +00:00
|
|
|
parser_state_e parser_state;
|
2023-01-27 14:37:25 +00:00
|
|
|
std::vector<StackFrame> stack;
|
2022-01-17 23:40:38 +00:00
|
|
|
std::string dict_key;
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t dict_key_offset;
|
2022-01-17 23:40:38 +00:00
|
|
|
};
|
2022-04-02 21:14:10 +00:00
|
|
|
} // namespace
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2022-05-20 00:28:13 +00:00
|
|
|
void
|
|
|
|
JSONParser::handle_u_code(
|
|
|
|
char const* s,
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t offset,
|
|
|
|
qpdf_offset_t i,
|
2022-05-20 00:28:13 +00:00
|
|
|
unsigned long& high_surrogate,
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t& high_offset,
|
2022-05-20 00:28:13 +00:00
|
|
|
std::string& result)
|
|
|
|
{
|
|
|
|
std::string hex = QUtil::hex_decode(std::string(s + i + 1, s + i + 5));
|
|
|
|
unsigned char high = static_cast<unsigned char>(hex.at(0));
|
|
|
|
unsigned char low = static_cast<unsigned char>(hex.at(1));
|
|
|
|
unsigned long codepoint = high;
|
|
|
|
codepoint <<= 8;
|
|
|
|
codepoint += low;
|
|
|
|
if ((codepoint & 0xFC00) == 0xD800) {
|
|
|
|
// high surrogate
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t new_high_offset = offset + i;
|
2022-05-20 00:28:13 +00:00
|
|
|
if (high_offset) {
|
|
|
|
QTC::TC("libtests", "JSON 16 high high");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(new_high_offset) +
|
2022-05-20 00:28:13 +00:00
|
|
|
": UTF-16 high surrogate found after previous high surrogate"
|
|
|
|
" at offset " +
|
2022-09-21 16:49:21 +00:00
|
|
|
std::to_string(high_offset));
|
2022-05-20 00:28:13 +00:00
|
|
|
}
|
|
|
|
high_offset = new_high_offset;
|
|
|
|
high_surrogate = codepoint;
|
|
|
|
} else if ((codepoint & 0xFC00) == 0xDC00) {
|
|
|
|
// low surrogate
|
|
|
|
if (offset + i != (high_offset + 6)) {
|
|
|
|
QTC::TC("libtests", "JSON 16 low not after high");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset + i) +
|
2022-05-20 00:28:13 +00:00
|
|
|
": UTF-16 low surrogate found not immediately after high"
|
|
|
|
" surrogate");
|
|
|
|
}
|
|
|
|
high_offset = 0;
|
|
|
|
codepoint =
|
|
|
|
0x10000U + ((high_surrogate & 0x3FFU) << 10U) + (codepoint & 0x3FF);
|
|
|
|
result += QUtil::toUTF8(codepoint);
|
|
|
|
} else {
|
|
|
|
result += QUtil::toUTF8(codepoint);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-01-17 23:40:38 +00:00
|
|
|
std::string
|
2022-05-21 13:05:54 +00:00
|
|
|
JSONParser::decode_string(std::string const& str, qpdf_offset_t offset)
|
2022-01-17 23:40:38 +00:00
|
|
|
{
|
|
|
|
// The string has already been validated when this private method
|
|
|
|
// is called, so errors are logic errors instead of runtime
|
|
|
|
// errors.
|
|
|
|
size_t len = str.length();
|
2022-04-02 21:14:10 +00:00
|
|
|
if ((len < 2) || (str.at(0) != '"') || (str.at(len - 1) != '"')) {
|
2022-01-17 23:40:38 +00:00
|
|
|
throw std::logic_error(
|
|
|
|
"JSON Parse: decode_string called with other than \"...\"");
|
|
|
|
}
|
|
|
|
char const* s = str.c_str();
|
|
|
|
// Move inside the quotation marks
|
|
|
|
++s;
|
|
|
|
len -= 2;
|
2022-05-20 00:28:13 +00:00
|
|
|
// Keep track of UTF-16 surrogate pairs.
|
|
|
|
unsigned long high_surrogate = 0;
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t high_offset = 0;
|
2022-01-17 23:40:38 +00:00
|
|
|
std::string result;
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t olen = toO(len);
|
|
|
|
for (qpdf_offset_t i = 0; i < olen; ++i) {
|
2022-04-02 21:14:10 +00:00
|
|
|
if (s[i] == '\\') {
|
2022-05-21 13:05:54 +00:00
|
|
|
if (i + 1 >= olen) {
|
2022-01-17 23:40:38 +00:00
|
|
|
throw std::logic_error("JSON parse: nothing after \\");
|
|
|
|
}
|
|
|
|
char ch = s[++i];
|
2022-04-02 21:14:10 +00:00
|
|
|
switch (ch) {
|
|
|
|
case '\\':
|
|
|
|
case '\"':
|
|
|
|
case '/':
|
2022-02-25 16:42:50 +00:00
|
|
|
// \/ is allowed in json input, but so is /, so we
|
|
|
|
// don't map / to \/ in output.
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, ch);
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 'b':
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, '\b');
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 'f':
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, '\f');
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 'n':
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, '\n');
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 'r':
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, '\r');
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 't':
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, '\t');
|
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
case 'u':
|
2022-05-21 13:05:54 +00:00
|
|
|
if (i + 4 >= olen) {
|
2022-01-17 23:40:38 +00:00
|
|
|
throw std::logic_error(
|
|
|
|
"JSON parse: not enough characters after \\u");
|
|
|
|
}
|
2022-05-20 00:28:13 +00:00
|
|
|
handle_u_code(
|
|
|
|
s, offset, i, high_surrogate, high_offset, result);
|
|
|
|
i += 4;
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
2022-04-02 21:14:10 +00:00
|
|
|
default:
|
|
|
|
throw std::logic_error("JSON parse: bad character after \\");
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-17 23:40:38 +00:00
|
|
|
result.append(1, s[i]);
|
|
|
|
}
|
|
|
|
}
|
2022-05-20 00:28:13 +00:00
|
|
|
if (high_offset) {
|
|
|
|
QTC::TC("libtests", "JSON 16 dangling high");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(high_offset) +
|
2022-05-20 00:28:13 +00:00
|
|
|
": UTF-16 high surrogate not followed by low surrogate");
|
|
|
|
}
|
2022-01-17 23:40:38 +00:00
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
2023-01-24 00:44:16 +00:00
|
|
|
void
|
|
|
|
JSONParser::numberError()
|
|
|
|
{
|
|
|
|
if (*p == '.') {
|
2023-01-25 19:46:53 +00:00
|
|
|
if (lex_state == ls_number || lex_state == ls_number_e ||
|
|
|
|
lex_state == ls_number_e_sign) {
|
2023-01-25 19:14:36 +00:00
|
|
|
QTC::TC("libtests", "JSON parse point after e");
|
2023-01-24 00:44:16 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: decimal point after e");
|
|
|
|
} else {
|
2023-01-24 00:34:22 +00:00
|
|
|
QTC::TC("libtests", "JSON parse duplicate point");
|
2023-01-24 00:44:16 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: decimal point already seen");
|
|
|
|
}
|
2023-01-24 01:36:38 +00:00
|
|
|
} else if (*p == 'e' || *p == 'E') {
|
2023-01-25 19:14:36 +00:00
|
|
|
QTC::TC("libtests", "JSON parse duplicate e");
|
2023-01-24 00:44:16 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: e already seen");
|
|
|
|
} else if ((*p == '+') || (*p == '-')) {
|
2023-01-23 19:51:27 +00:00
|
|
|
QTC::TC("libtests", "JSON parse unexpected sign");
|
2023-01-24 00:44:16 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: unexpected sign");
|
2023-01-23 20:08:11 +00:00
|
|
|
} else if (QUtil::is_space(*p) || strchr("{}[]:,", *p)) {
|
|
|
|
QTC::TC("libtests", "JSON parse incomplete number");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: incomplete number");
|
2023-01-24 00:44:16 +00:00
|
|
|
} else {
|
|
|
|
QTC::TC("libtests", "JSON parse numeric bad character");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: unexpected character " + std::string(p, 1));
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
void
|
|
|
|
JSONParser::getToken()
|
2022-01-17 23:40:38 +00:00
|
|
|
{
|
2022-05-04 13:12:57 +00:00
|
|
|
enum { append, ignore, reread } action = append;
|
|
|
|
bool ready = false;
|
|
|
|
token.clear();
|
|
|
|
while (!done) {
|
|
|
|
if (p == (buf + bytes)) {
|
|
|
|
p = buf;
|
|
|
|
bytes = is.read(buf, sizeof(buf));
|
|
|
|
if (bytes == 0) {
|
|
|
|
done = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
if (*p == 0) {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse null character");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: null character at offset " + std::to_string(offset));
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
2022-05-04 13:12:57 +00:00
|
|
|
action = append;
|
2022-04-02 21:14:10 +00:00
|
|
|
switch (lex_state) {
|
|
|
|
case ls_top:
|
2023-01-26 11:36:44 +00:00
|
|
|
token_start = offset;
|
2022-04-02 21:14:10 +00:00
|
|
|
if (*p == '"') {
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state = ls_string;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2022-05-04 13:12:57 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
} else if (*p == ',') {
|
|
|
|
lex_state = ls_comma;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
|
|
|
} else if (*p == ':') {
|
|
|
|
lex_state = ls_colon;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
|
|
|
} else if (*p == '{') {
|
|
|
|
lex_state = ls_begin_dict;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
|
|
|
} else if (*p == '}') {
|
|
|
|
lex_state = ls_end_dict;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
|
|
|
} else if (*p == '[') {
|
|
|
|
lex_state = ls_begin_array;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
|
|
|
} else if (*p == ']') {
|
|
|
|
lex_state = ls_end_array;
|
2023-01-26 13:31:53 +00:00
|
|
|
action = ignore;
|
2023-01-26 13:21:45 +00:00
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if ((*p >= 'a') && (*p <= 'z')) {
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state = ls_alpha;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (*p == '-') {
|
2023-01-24 11:47:06 +00:00
|
|
|
lex_state = ls_number_minus;
|
2023-01-24 11:22:06 +00:00
|
|
|
} else if ((*p >= '1') && (*p <= '9')) {
|
2023-01-23 19:51:27 +00:00
|
|
|
lex_state = ls_number_before_point;
|
2023-01-24 11:22:06 +00:00
|
|
|
} else if (*p == '0') {
|
|
|
|
lex_state = ls_number_leading_zero;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse bad character");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset) +
|
2022-01-17 23:40:38 +00:00
|
|
|
": unexpected character " + std::string(p, 1));
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2023-01-24 11:47:06 +00:00
|
|
|
case ls_number_minus:
|
|
|
|
if ((*p >= '1') && (*p <= '9')) {
|
2023-01-23 19:51:27 +00:00
|
|
|
lex_state = ls_number_before_point;
|
2023-01-24 11:47:06 +00:00
|
|
|
} else if (*p == '0') {
|
2023-01-24 11:22:06 +00:00
|
|
|
lex_state = ls_number_leading_zero;
|
2023-01-24 11:47:06 +00:00
|
|
|
} else {
|
|
|
|
QTC::TC("libtests", "JSON parse number minus no digits");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": numeric literal: no digit after minus sign");
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2023-01-24 11:22:06 +00:00
|
|
|
case ls_number_leading_zero:
|
|
|
|
if (*p == '.') {
|
2023-01-23 20:08:11 +00:00
|
|
|
lex_state = ls_number_point;
|
2023-01-24 11:22:06 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-24 11:22:06 +00:00
|
|
|
action = ignore;
|
|
|
|
ready = true;
|
|
|
|
} else if (strchr("{}[]:,", *p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-24 11:22:06 +00:00
|
|
|
action = reread;
|
|
|
|
ready = true;
|
2023-01-24 01:36:38 +00:00
|
|
|
} else if (*p == 'e' || *p == 'E') {
|
|
|
|
lex_state = ls_number_e;
|
2023-01-24 11:22:06 +00:00
|
|
|
} else {
|
|
|
|
QTC::TC("libtests", "JSON parse leading zero");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": number with leading zero");
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2023-01-23 19:51:27 +00:00
|
|
|
case ls_number_before_point:
|
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
2023-01-25 19:46:53 +00:00
|
|
|
// continue
|
2023-01-23 19:51:27 +00:00
|
|
|
} else if (*p == '.') {
|
2023-01-23 20:08:11 +00:00
|
|
|
lex_state = ls_number_point;
|
2023-01-23 19:51:27 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-23 19:51:27 +00:00
|
|
|
action = ignore;
|
|
|
|
ready = true;
|
|
|
|
} else if (strchr("{}[]:,", *p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-23 19:51:27 +00:00
|
|
|
action = reread;
|
|
|
|
ready = true;
|
2023-01-24 01:36:38 +00:00
|
|
|
} else if (*p == 'e' || *p == 'E') {
|
|
|
|
lex_state = ls_number_e;
|
2023-01-23 19:51:27 +00:00
|
|
|
} else {
|
|
|
|
numberError();
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2023-01-23 20:08:11 +00:00
|
|
|
case ls_number_point:
|
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
2023-01-24 00:34:22 +00:00
|
|
|
lex_state = ls_number_after_point;
|
|
|
|
} else {
|
|
|
|
numberError();
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ls_number_after_point:
|
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
2023-01-25 19:46:53 +00:00
|
|
|
// continue
|
2023-01-24 00:34:22 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-24 00:34:22 +00:00
|
|
|
action = ignore;
|
|
|
|
ready = true;
|
|
|
|
} else if (strchr("{}[]:,", *p)) {
|
2023-01-25 19:14:36 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-24 00:34:22 +00:00
|
|
|
action = reread;
|
|
|
|
ready = true;
|
2023-01-24 01:36:38 +00:00
|
|
|
} else if (*p == 'e' || *p == 'E') {
|
|
|
|
lex_state = ls_number_e;
|
|
|
|
} else {
|
|
|
|
numberError();
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ls_number_e:
|
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
|
|
|
lex_state = ls_number;
|
|
|
|
} else if ((*p == '+') || (*p == '-')) {
|
2023-01-24 13:04:31 +00:00
|
|
|
lex_state = ls_number_e_sign;
|
|
|
|
} else {
|
|
|
|
numberError();
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ls_number_e_sign:
|
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
2023-01-24 01:36:38 +00:00
|
|
|
lex_state = ls_number;
|
2023-01-23 20:08:11 +00:00
|
|
|
} else {
|
|
|
|
numberError();
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_number:
|
2023-01-25 19:14:36 +00:00
|
|
|
// We only get here after we have seen an exponent.
|
2022-04-02 21:14:10 +00:00
|
|
|
if ((*p >= '0') && (*p <= '9')) {
|
2023-01-25 19:46:53 +00:00
|
|
|
// continue
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2022-05-04 13:12:57 +00:00
|
|
|
action = ignore;
|
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (strchr("{}[]:,", *p)) {
|
2022-05-04 13:12:57 +00:00
|
|
|
action = reread;
|
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2023-01-24 00:44:16 +00:00
|
|
|
numberError();
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_alpha:
|
|
|
|
if ((*p >= 'a') && (*p <= 'z')) {
|
2022-01-17 23:40:38 +00:00
|
|
|
// okay
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (QUtil::is_space(*p)) {
|
2022-05-04 13:12:57 +00:00
|
|
|
action = ignore;
|
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (strchr("{}[]:,", *p)) {
|
2022-05-04 13:12:57 +00:00
|
|
|
action = reread;
|
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse keyword bad character");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset) +
|
2022-01-17 23:40:38 +00:00
|
|
|
": keyword: unexpected character " + std::string(p, 1));
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_string:
|
|
|
|
if (*p == '"') {
|
2023-01-27 17:03:24 +00:00
|
|
|
token += '"';
|
|
|
|
token = decode_string(token, token_start);
|
|
|
|
action = ignore;
|
2022-05-04 13:12:57 +00:00
|
|
|
ready = true;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (*p == '\\') {
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state = ls_backslash;
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_backslash:
|
2022-01-28 14:05:06 +00:00
|
|
|
/* cSpell: ignore bfnrt */
|
2022-04-02 21:14:10 +00:00
|
|
|
if (strchr("\\\"/bfnrt", *p)) {
|
2022-01-17 23:40:38 +00:00
|
|
|
lex_state = ls_string;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else if (*p == 'u') {
|
2022-05-04 13:12:57 +00:00
|
|
|
lex_state = ls_u4;
|
|
|
|
u_count = 0;
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse backslash bad character");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset) +
|
2022-01-17 23:40:38 +00:00
|
|
|
": invalid character after backslash: " +
|
|
|
|
std::string(p, 1));
|
|
|
|
}
|
|
|
|
break;
|
2022-05-04 13:12:57 +00:00
|
|
|
|
|
|
|
case ls_u4:
|
|
|
|
if (!QUtil::is_hex_digit(*p)) {
|
|
|
|
QTC::TC("libtests", "JSON parse bad hex after u");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset - u_count - 1) +
|
2022-05-04 13:12:57 +00:00
|
|
|
": \\u must be followed by four hex digits");
|
|
|
|
}
|
|
|
|
if (++u_count == 4) {
|
|
|
|
lex_state = ls_string;
|
|
|
|
}
|
|
|
|
break;
|
2023-01-26 13:21:45 +00:00
|
|
|
|
|
|
|
default:
|
|
|
|
throw std::logic_error(
|
|
|
|
"JSONParser::getToken : trying to handle delimiter state");
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
2022-05-04 13:12:57 +00:00
|
|
|
switch (action) {
|
|
|
|
case reread:
|
|
|
|
break;
|
|
|
|
case append:
|
|
|
|
token.append(1, *p);
|
|
|
|
// fall through
|
|
|
|
case ignore:
|
|
|
|
++p;
|
|
|
|
++offset;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (ready) {
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
2022-05-04 13:12:57 +00:00
|
|
|
if (done) {
|
2023-01-26 09:55:34 +00:00
|
|
|
if (!token.empty() && !ready) {
|
2022-04-02 21:14:10 +00:00
|
|
|
switch (lex_state) {
|
|
|
|
case ls_top:
|
2022-01-17 23:40:38 +00:00
|
|
|
// Can't happen
|
2022-05-04 13:12:57 +00:00
|
|
|
throw std::logic_error("tok_start set in ls_top while parsing");
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
|
2023-01-24 13:04:31 +00:00
|
|
|
case ls_number_leading_zero:
|
2023-01-23 19:51:27 +00:00
|
|
|
case ls_number_before_point:
|
2023-01-24 00:34:22 +00:00
|
|
|
case ls_number_after_point:
|
2023-01-26 09:55:34 +00:00
|
|
|
lex_state = ls_number;
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
|
2023-01-26 09:55:34 +00:00
|
|
|
case ls_number:
|
|
|
|
case ls_alpha:
|
|
|
|
// terminal state
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
2023-01-26 09:55:34 +00:00
|
|
|
|
|
|
|
default:
|
|
|
|
QTC::TC("libtests", "JSON parse ls premature end of input");
|
|
|
|
throw std::runtime_error("JSON: premature end of input");
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
|
|
|
JSONParser::handleToken()
|
|
|
|
{
|
2023-01-26 13:21:45 +00:00
|
|
|
if (lex_state == ls_top) {
|
2022-01-17 23:40:38 +00:00
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
if (parser_state == ps_done) {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse junk after object");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset) +
|
2022-05-04 13:12:57 +00:00
|
|
|
": material follows end of object: " + token);
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
|
2022-02-04 15:52:37 +00:00
|
|
|
std::shared_ptr<JSON> item;
|
2023-01-27 14:37:25 +00:00
|
|
|
auto tos = stack.empty() ? nullptr : stack.back().item;
|
2023-01-27 13:35:02 +00:00
|
|
|
auto ls = lex_state;
|
|
|
|
lex_state = ls_top;
|
2023-01-26 13:21:45 +00:00
|
|
|
|
2023-01-27 13:35:02 +00:00
|
|
|
switch (ls) {
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_begin_dict:
|
|
|
|
item = std::make_shared<JSON>(JSON::makeDictionary());
|
|
|
|
break;
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_begin_array:
|
|
|
|
item = std::make_shared<JSON>(JSON::makeArray());
|
|
|
|
break;
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_colon:
|
2023-01-26 14:10:48 +00:00
|
|
|
if (parser_state != ps_dict_after_key) {
|
|
|
|
QTC::TC("libtests", "JSON parse unexpected :");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": unexpected colon");
|
|
|
|
}
|
2023-01-27 12:26:27 +00:00
|
|
|
parser_state = ps_dict_after_colon;
|
|
|
|
return;
|
2023-01-26 14:10:48 +00:00
|
|
|
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_comma:
|
2023-01-26 14:10:48 +00:00
|
|
|
if (!((parser_state == ps_dict_after_item) ||
|
|
|
|
(parser_state == ps_array_after_item))) {
|
|
|
|
QTC::TC("libtests", "JSON parse unexpected ,");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": unexpected comma");
|
|
|
|
}
|
2023-01-27 12:26:27 +00:00
|
|
|
if (parser_state == ps_dict_after_item) {
|
|
|
|
parser_state = ps_dict_after_comma;
|
|
|
|
} else if (parser_state == ps_array_after_item) {
|
|
|
|
parser_state = ps_array_after_comma;
|
|
|
|
} else {
|
|
|
|
throw std::logic_error("JSONParser::handleToken: unexpected parser"
|
|
|
|
" state for comma");
|
|
|
|
}
|
|
|
|
return;
|
2023-01-26 14:10:48 +00:00
|
|
|
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_end_array:
|
2023-01-27 12:50:14 +00:00
|
|
|
if (!(parser_state == ps_array_begin ||
|
|
|
|
parser_state == ps_array_after_item)) {
|
2023-01-26 14:10:48 +00:00
|
|
|
QTC::TC("libtests", "JSON parse unexpected ]");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": unexpected array end delimiter");
|
|
|
|
}
|
2023-01-27 14:37:25 +00:00
|
|
|
parser_state = stack.back().state;
|
2023-01-27 12:50:14 +00:00
|
|
|
tos->setEnd(offset);
|
|
|
|
if (reactor) {
|
|
|
|
reactor->containerEnd(*tos);
|
|
|
|
}
|
|
|
|
if (parser_state != ps_done) {
|
|
|
|
stack.pop_back();
|
|
|
|
}
|
|
|
|
return;
|
2023-01-26 14:10:48 +00:00
|
|
|
|
2023-01-26 13:21:45 +00:00
|
|
|
case ls_end_dict:
|
2023-01-26 14:10:48 +00:00
|
|
|
if (!((parser_state == ps_dict_begin) ||
|
|
|
|
(parser_state == ps_dict_after_item))) {
|
|
|
|
QTC::TC("libtests", "JSON parse unexpected }");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": unexpected dictionary end delimiter");
|
|
|
|
}
|
2023-01-27 14:37:25 +00:00
|
|
|
parser_state = stack.back().state;
|
2023-01-27 12:50:14 +00:00
|
|
|
tos->setEnd(offset);
|
|
|
|
if (reactor) {
|
|
|
|
reactor->containerEnd(*tos);
|
|
|
|
}
|
|
|
|
if (parser_state != ps_done) {
|
|
|
|
stack.pop_back();
|
|
|
|
}
|
|
|
|
return;
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_number:
|
2022-05-04 13:12:57 +00:00
|
|
|
item = std::make_shared<JSON>(JSON::makeNumber(token));
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_alpha:
|
2022-05-04 13:12:57 +00:00
|
|
|
if (token == "true") {
|
2022-02-04 15:52:37 +00:00
|
|
|
item = std::make_shared<JSON>(JSON::makeBool(true));
|
2022-05-04 13:12:57 +00:00
|
|
|
} else if (token == "false") {
|
2022-02-04 15:52:37 +00:00
|
|
|
item = std::make_shared<JSON>(JSON::makeBool(false));
|
2022-05-04 13:12:57 +00:00
|
|
|
} else if (token == "null") {
|
2022-02-04 15:52:37 +00:00
|
|
|
item = std::make_shared<JSON>(JSON::makeNull());
|
2022-04-02 21:14:10 +00:00
|
|
|
} else {
|
2022-01-17 23:40:38 +00:00
|
|
|
QTC::TC("libtests", "JSON parse invalid keyword");
|
|
|
|
throw std::runtime_error(
|
2022-09-21 16:49:21 +00:00
|
|
|
"JSON: offset " + std::to_string(offset) +
|
2022-05-04 13:12:57 +00:00
|
|
|
": invalid keyword " + token);
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
|
2022-04-02 21:14:10 +00:00
|
|
|
case ls_string:
|
2023-01-27 13:35:02 +00:00
|
|
|
if (parser_state == ps_dict_begin ||
|
|
|
|
parser_state == ps_dict_after_comma) {
|
2023-01-27 17:03:24 +00:00
|
|
|
dict_key = token;
|
2023-01-27 13:35:02 +00:00
|
|
|
dict_key_offset = token_start;
|
|
|
|
parser_state = ps_dict_after_key;
|
|
|
|
return;
|
|
|
|
} else {
|
2023-01-27 17:03:24 +00:00
|
|
|
item = std::make_shared<JSON>(JSON::makeString(token));
|
2023-01-27 13:35:02 +00:00
|
|
|
}
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
|
2023-01-26 09:55:34 +00:00
|
|
|
default:
|
2022-05-04 13:12:57 +00:00
|
|
|
throw std::logic_error(
|
2023-01-26 09:55:34 +00:00
|
|
|
"JSONParser::handleToken : non-terminal lexer state encountered");
|
2022-01-17 23:40:38 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2023-01-27 11:12:30 +00:00
|
|
|
item->setStart(token_start);
|
|
|
|
item->setEnd(offset);
|
2023-01-27 12:50:14 +00:00
|
|
|
|
|
|
|
switch (parser_state) {
|
|
|
|
case ps_dict_begin:
|
|
|
|
case ps_dict_after_comma:
|
2023-01-27 13:35:02 +00:00
|
|
|
QTC::TC("libtests", "JSON parse string as dict key");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": expect string as dictionary key");
|
2023-01-27 12:50:14 +00:00
|
|
|
break;
|
|
|
|
|
|
|
|
case ps_dict_after_colon:
|
|
|
|
if (tos->checkDictionaryKeySeen(dict_key)) {
|
|
|
|
QTC::TC("libtests", "JSON parse duplicate key");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(dict_key_offset) +
|
|
|
|
": duplicated dictionary key");
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
2023-01-27 12:50:14 +00:00
|
|
|
if (!reactor || !reactor->dictionaryItem(dict_key, *item)) {
|
|
|
|
tos->addDictionaryMember(dict_key, *item);
|
2022-05-01 18:06:31 +00:00
|
|
|
}
|
2023-01-27 15:39:16 +00:00
|
|
|
parser_state = ps_dict_after_item;
|
2023-01-27 12:50:14 +00:00
|
|
|
break;
|
2022-05-01 18:06:31 +00:00
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_array_begin:
|
|
|
|
case ps_array_after_comma:
|
|
|
|
if (!reactor || !reactor->arrayItem(*item)) {
|
|
|
|
tos->addArrayElement(*item);
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
2023-01-27 15:39:16 +00:00
|
|
|
parser_state = ps_array_after_item;
|
2023-01-27 12:50:14 +00:00
|
|
|
break;
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_top:
|
2023-01-27 15:39:16 +00:00
|
|
|
if (!(item->isDictionary() || item->isArray())) {
|
2023-01-27 14:37:25 +00:00
|
|
|
stack.push_back({ps_done, item});
|
2023-01-27 15:39:16 +00:00
|
|
|
parser_state = ps_done;
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
parser_state = ps_done;
|
2023-01-27 12:50:14 +00:00
|
|
|
break;
|
2022-01-17 23:40:38 +00:00
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_dict_after_key:
|
2023-01-27 13:08:55 +00:00
|
|
|
QTC::TC("libtests", "JSON parse expected colon");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) + ": expected ':'");
|
|
|
|
break;
|
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_dict_after_item:
|
2023-01-27 13:08:55 +00:00
|
|
|
QTC::TC("libtests", "JSON parse expected , or }");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) + ": expected ',' or '}'");
|
|
|
|
break;
|
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_array_after_item:
|
2023-01-27 13:08:55 +00:00
|
|
|
QTC::TC("libtests", "JSON parse expected, or ]");
|
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) + ": expected ',' or ']'");
|
|
|
|
break;
|
|
|
|
|
2023-01-27 12:50:14 +00:00
|
|
|
case ps_done:
|
|
|
|
throw std::logic_error(
|
|
|
|
"JSONParser::handleToken: unexpected parser state");
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
|
2023-01-27 14:37:25 +00:00
|
|
|
if (item->isDictionary() || item->isArray()) {
|
2023-01-27 14:37:25 +00:00
|
|
|
stack.push_back({parser_state, item});
|
2022-05-14 14:11:52 +00:00
|
|
|
// Calling container start method is postponed until after
|
|
|
|
// adding the containers to their parent containers, if any.
|
|
|
|
// This makes it much easier to keep track of the current
|
|
|
|
// nesting level.
|
|
|
|
if (item->isDictionary()) {
|
2023-01-27 14:37:25 +00:00
|
|
|
if (reactor) {
|
|
|
|
reactor->dictionaryStart();
|
|
|
|
}
|
2023-01-27 15:39:16 +00:00
|
|
|
parser_state = ps_dict_begin;
|
2022-05-14 14:11:52 +00:00
|
|
|
} else if (item->isArray()) {
|
2023-01-27 14:37:25 +00:00
|
|
|
if (reactor) {
|
|
|
|
reactor->arrayStart();
|
|
|
|
}
|
2023-01-27 15:39:16 +00:00
|
|
|
parser_state = ps_array_begin;
|
2022-05-14 14:11:52 +00:00
|
|
|
}
|
2023-01-27 13:35:02 +00:00
|
|
|
|
2023-01-27 14:37:25 +00:00
|
|
|
if (stack.size() > 500) {
|
2023-01-27 14:37:25 +00:00
|
|
|
throw std::runtime_error(
|
|
|
|
"JSON: offset " + std::to_string(offset) +
|
|
|
|
": maximum object depth exceeded");
|
|
|
|
}
|
2022-05-01 13:34:17 +00:00
|
|
|
}
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
|
2022-02-04 15:52:37 +00:00
|
|
|
std::shared_ptr<JSON>
|
2022-05-04 13:12:57 +00:00
|
|
|
JSONParser::parse()
|
2022-01-17 23:40:38 +00:00
|
|
|
{
|
2022-05-04 13:12:57 +00:00
|
|
|
while (!done) {
|
2022-01-17 23:40:38 +00:00
|
|
|
getToken();
|
|
|
|
handleToken();
|
|
|
|
}
|
2022-04-02 21:14:10 +00:00
|
|
|
if (parser_state != ps_done) {
|
2022-01-28 14:05:06 +00:00
|
|
|
QTC::TC("libtests", "JSON parse premature EOF");
|
2022-01-17 23:40:38 +00:00
|
|
|
throw std::runtime_error("JSON: premature end of input");
|
|
|
|
}
|
2023-01-27 14:37:25 +00:00
|
|
|
auto const& tos = stack.back().item;
|
2022-05-01 18:06:31 +00:00
|
|
|
if (reactor && tos.get() && !(tos->isArray() || tos->isDictionary())) {
|
|
|
|
reactor->topLevelScalar();
|
|
|
|
}
|
|
|
|
return tos;
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
2022-05-04 13:12:57 +00:00
|
|
|
JSON::parse(InputSource& is, Reactor* reactor)
|
|
|
|
{
|
|
|
|
JSONParser jp(is, reactor);
|
|
|
|
return *jp.parse();
|
|
|
|
}
|
|
|
|
|
|
|
|
JSON
|
|
|
|
JSON::parse(std::string const& s)
|
2022-01-17 23:40:38 +00:00
|
|
|
{
|
2022-05-04 13:12:57 +00:00
|
|
|
BufferInputSource bis("json input", s);
|
|
|
|
JSONParser jp(bis, nullptr);
|
|
|
|
return *jp.parse();
|
2022-01-17 23:40:38 +00:00
|
|
|
}
|
2022-05-01 18:06:31 +00:00
|
|
|
|
|
|
|
void
|
2022-05-21 13:05:54 +00:00
|
|
|
JSON::setStart(qpdf_offset_t start)
|
2022-05-01 18:06:31 +00:00
|
|
|
{
|
|
|
|
this->m->start = start;
|
|
|
|
}
|
|
|
|
|
|
|
|
void
|
2022-05-21 13:05:54 +00:00
|
|
|
JSON::setEnd(qpdf_offset_t end)
|
2022-05-01 18:06:31 +00:00
|
|
|
{
|
|
|
|
this->m->end = end;
|
|
|
|
}
|
|
|
|
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t
|
2022-05-01 18:06:31 +00:00
|
|
|
JSON::getStart() const
|
|
|
|
{
|
|
|
|
return this->m->start;
|
|
|
|
}
|
|
|
|
|
2022-05-21 13:05:54 +00:00
|
|
|
qpdf_offset_t
|
2022-05-01 18:06:31 +00:00
|
|
|
JSON::getEnd() const
|
|
|
|
{
|
|
|
|
return this->m->end;
|
|
|
|
}
|