blob: 2ed5c418947013768092fff46699de59038ebf29 [file] [log] [blame]
/*
* Copyright 2014 Google Inc. All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#include <algorithm>
#include <cmath>
#include <list>
#include <string>
#include <utility>
#include "flatbuffers/base.h"
#include "flatbuffers/idl.h"
#include "flatbuffers/util.h"
namespace flatbuffers {
// Reflects the version at the compiling time of binary(lib/dll/so).
const char *FLATBUFFERS_VERSION() {
// clang-format off
return
FLATBUFFERS_STRING(FLATBUFFERS_VERSION_MAJOR) "."
FLATBUFFERS_STRING(FLATBUFFERS_VERSION_MINOR) "."
FLATBUFFERS_STRING(FLATBUFFERS_VERSION_REVISION);
// clang-format on
}
namespace {
static const double kPi = 3.14159265358979323846;
// The enums in the reflection schema should match the ones we use internally.
// Compare the last element to check if these go out of sync.
static_assert(BASE_TYPE_UNION == static_cast<BaseType>(reflection::Union),
"enums don't match");
// Any parsing calls have to be wrapped in this macro, which automates
// handling of recursive error checking a bit. It will check the received
// CheckedError object, and return straight away on error.
#define ECHECK(call) \
{ \
auto ce = (call); \
if (ce.Check()) return ce; \
}
// These two functions are called hundreds of times below, so define a short
// form:
#define NEXT() ECHECK(Next())
#define EXPECT(tok) ECHECK(Expect(tok))
static bool ValidateUTF8(const std::string &str) {
const char *s = &str[0];
const char *const sEnd = s + str.length();
while (s < sEnd) {
if (FromUTF8(&s) < 0) { return false; }
}
return true;
}
static bool IsLowerSnakeCase(const std::string &str) {
for (size_t i = 0; i < str.length(); i++) {
char c = str[i];
if (!check_ascii_range(c, 'a', 'z') && !is_digit(c) && c != '_') {
return false;
}
}
return true;
}
static void DeserializeDoc(std::vector<std::string> &doc,
const Vector<Offset<String>> *documentation) {
if (documentation == nullptr) return;
for (uoffset_t index = 0; index < documentation->size(); index++)
doc.push_back(documentation->Get(index)->str());
}
static CheckedError NoError() { return CheckedError(false); }
template<typename T> static std::string TypeToIntervalString() {
return "[" + NumToString((flatbuffers::numeric_limits<T>::lowest)()) + "; " +
NumToString((flatbuffers::numeric_limits<T>::max)()) + "]";
}
// atot: template version of atoi/atof: convert a string to an instance of T.
template<typename T>
static bool atot_scalar(const char *s, T *val, bool_constant<false>) {
return StringToNumber(s, val);
}
template<typename T>
static bool atot_scalar(const char *s, T *val, bool_constant<true>) {
// Normalize NaN parsed from fbs or json to unsigned NaN.
if (false == StringToNumber(s, val)) return false;
*val = (*val != *val) ? std::fabs(*val) : *val;
return true;
}
template<typename T>
static CheckedError atot(const char *s, Parser &parser, T *val) {
auto done = atot_scalar(s, val, bool_constant<is_floating_point<T>::value>());
if (done) return NoError();
if (0 == *val)
return parser.Error("invalid number: \"" + std::string(s) + "\"");
else
return parser.Error("invalid number: \"" + std::string(s) + "\"" +
", constant does not fit " + TypeToIntervalString<T>());
}
template<>
CheckedError atot<Offset<void>>(const char *s, Parser &parser,
Offset<void> *val) {
(void)parser;
*val = Offset<void>(atoi(s));
return NoError();
}
template<typename T>
static T *LookupTableByName(const SymbolTable<T> &table,
const std::string &name,
const Namespace &current_namespace,
size_t skip_top) {
const auto &components = current_namespace.components;
if (table.dict.empty()) return nullptr;
if (components.size() < skip_top) return nullptr;
const auto N = components.size() - skip_top;
std::string full_name;
for (size_t i = 0; i < N; i++) {
full_name += components[i];
full_name += '.';
}
for (size_t i = N; i > 0; i--) {
full_name += name;
auto obj = table.Lookup(full_name);
if (obj) return obj;
auto len = full_name.size() - components[i - 1].size() - 1 - name.size();
full_name.resize(len);
}
FLATBUFFERS_ASSERT(full_name.empty());
return table.Lookup(name); // lookup in global namespace
}
// Declare tokens we'll use. Single character tokens are represented by their
// ascii character code (e.g. '{'), others above 256.
// clang-format off
#define FLATBUFFERS_GEN_TOKENS(TD) \
TD(Eof, 256, "end of file") \
TD(StringConstant, 257, "string constant") \
TD(IntegerConstant, 258, "integer constant") \
TD(FloatConstant, 259, "float constant") \
TD(Identifier, 260, "identifier")
#ifdef __GNUC__
__extension__ // Stop GCC complaining about trailing comma with -Wpendantic.
#endif
enum {
#define FLATBUFFERS_TOKEN(NAME, VALUE, STRING) kToken ## NAME = VALUE,
FLATBUFFERS_GEN_TOKENS(FLATBUFFERS_TOKEN)
#undef FLATBUFFERS_TOKEN
};
static std::string TokenToString(int t) {
static const char * const tokens[] = {
#define FLATBUFFERS_TOKEN(NAME, VALUE, STRING) STRING,
FLATBUFFERS_GEN_TOKENS(FLATBUFFERS_TOKEN)
#undef FLATBUFFERS_TOKEN
#define FLATBUFFERS_TD(ENUM, IDLTYPE, ...) \
IDLTYPE,
FLATBUFFERS_GEN_TYPES(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
};
if (t < 256) { // A single ascii char token.
std::string s;
s.append(1, static_cast<char>(t));
return s;
} else { // Other tokens.
return tokens[t - 256];
}
}
// clang-format on
static bool IsIdentifierStart(char c) { return is_alpha(c) || (c == '_'); }
static bool CompareSerializedScalars(const uint8_t *a, const uint8_t *b,
const FieldDef &key) {
switch (key.value.type.base_type) {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_##ENUM: { \
CTYPE def = static_cast<CTYPE>(0); \
if (!a || !b) { StringToNumber(key.value.constant.c_str(), &def); } \
const auto av = a ? ReadScalar<CTYPE>(a) : def; \
const auto bv = b ? ReadScalar<CTYPE>(b) : def; \
return av < bv; \
}
FLATBUFFERS_GEN_TYPES_SCALAR(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
default: {
FLATBUFFERS_ASSERT(false && "scalar type expected");
return false;
}
}
}
static bool CompareTablesByScalarKey(const Offset<Table> *_a,
const Offset<Table> *_b,
const FieldDef &key) {
const voffset_t offset = key.value.offset;
// Indirect offset pointer to table pointer.
auto a = reinterpret_cast<const uint8_t *>(_a) + ReadScalar<uoffset_t>(_a);
auto b = reinterpret_cast<const uint8_t *>(_b) + ReadScalar<uoffset_t>(_b);
// Fetch field address from table.
a = reinterpret_cast<const Table *>(a)->GetAddressOf(offset);
b = reinterpret_cast<const Table *>(b)->GetAddressOf(offset);
return CompareSerializedScalars(a, b, key);
}
static bool CompareTablesByStringKey(const Offset<Table> *_a,
const Offset<Table> *_b,
const FieldDef &key) {
const voffset_t offset = key.value.offset;
// Indirect offset pointer to table pointer.
auto a = reinterpret_cast<const uint8_t *>(_a) + ReadScalar<uoffset_t>(_a);
auto b = reinterpret_cast<const uint8_t *>(_b) + ReadScalar<uoffset_t>(_b);
// Fetch field address from table.
a = reinterpret_cast<const Table *>(a)->GetAddressOf(offset);
b = reinterpret_cast<const Table *>(b)->GetAddressOf(offset);
if (a && b) {
// Indirect offset pointer to string pointer.
a += ReadScalar<uoffset_t>(a);
b += ReadScalar<uoffset_t>(b);
return *reinterpret_cast<const String *>(a) <
*reinterpret_cast<const String *>(b);
} else {
return a ? true : false;
}
}
static void SwapSerializedTables(Offset<Table> *a, Offset<Table> *b) {
// These are serialized offsets, so are relative where they are
// stored in memory, so compute the distance between these pointers:
ptrdiff_t diff = (b - a) * sizeof(Offset<Table>);
FLATBUFFERS_ASSERT(diff >= 0); // Guaranteed by SimpleQsort.
auto udiff = static_cast<uoffset_t>(diff);
a->o = EndianScalar(ReadScalar<uoffset_t>(a) - udiff);
b->o = EndianScalar(ReadScalar<uoffset_t>(b) + udiff);
std::swap(*a, *b);
}
// See below for why we need our own sort :(
template<typename T, typename F, typename S>
static void SimpleQsort(T *begin, T *end, size_t width, F comparator,
S swapper) {
if (end - begin <= static_cast<ptrdiff_t>(width)) return;
auto l = begin + width;
auto r = end;
while (l < r) {
if (comparator(begin, l)) {
r -= width;
swapper(l, r);
} else {
l += width;
}
}
l -= width;
swapper(begin, l);
SimpleQsort(begin, l, width, comparator, swapper);
SimpleQsort(r, end, width, comparator, swapper);
}
template<typename T> static inline void SingleValueRepack(Value &e, T val) {
// Remove leading zeros.
if (IsInteger(e.type.base_type)) { e.constant = NumToString(val); }
}
#if defined(FLATBUFFERS_HAS_NEW_STRTOD) && (FLATBUFFERS_HAS_NEW_STRTOD > 0)
// Normalize defaults NaN to unsigned quiet-NaN(0) if value was parsed from
// hex-float literal.
static void SingleValueRepack(Value &e, float val) {
if (val != val) e.constant = "nan";
}
static void SingleValueRepack(Value &e, double val) {
if (val != val) e.constant = "nan";
}
#endif
template<typename T> static uint64_t EnumDistanceImpl(T e1, T e2) {
if (e1 < e2) { std::swap(e1, e2); } // use std for scalars
// Signed overflow may occur, use unsigned calculation.
// The unsigned overflow is well-defined by C++ standard (modulo 2^n).
return static_cast<uint64_t>(e1) - static_cast<uint64_t>(e2);
}
static bool compareFieldDefs(const FieldDef *a, const FieldDef *b) {
auto a_id = atoi(a->attributes.Lookup("id")->constant.c_str());
auto b_id = atoi(b->attributes.Lookup("id")->constant.c_str());
return a_id < b_id;
}
static Namespace *GetNamespace(
const std::string &qualified_name, std::vector<Namespace *> &namespaces,
std::map<std::string, Namespace *> &namespaces_index) {
size_t dot = qualified_name.find_last_of('.');
std::string namespace_name = (dot != std::string::npos)
? std::string(qualified_name.c_str(), dot)
: "";
Namespace *&ns = namespaces_index[namespace_name];
if (!ns) {
ns = new Namespace();
namespaces.push_back(ns);
size_t pos = 0;
for (;;) {
dot = qualified_name.find('.', pos);
if (dot == std::string::npos) { break; }
ns->components.push_back(qualified_name.substr(pos, dot - pos));
pos = dot + 1;
}
}
return ns;
}
// Generate a unique hash for a file based on its name and contents (if any).
static uint64_t HashFile(const char *source_filename, const char *source) {
uint64_t hash = 0;
if (source_filename)
hash = HashFnv1a<uint64_t>(StripPath(source_filename).c_str());
if (source && *source) hash ^= HashFnv1a<uint64_t>(source);
return hash;
}
template<typename T> static bool compareName(const T *a, const T *b) {
return a->defined_namespace->GetFullyQualifiedName(a->name) <
b->defined_namespace->GetFullyQualifiedName(b->name);
}
template<typename T> static void AssignIndices(const std::vector<T *> &defvec) {
// Pre-sort these vectors, such that we can set the correct indices for them.
auto vec = defvec;
std::sort(vec.begin(), vec.end(), compareName<T>);
for (int i = 0; i < static_cast<int>(vec.size()); i++) vec[i]->index = i;
}
} // namespace
// clang-format off
const char *const kTypeNames[] = {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, ...) \
IDLTYPE,
FLATBUFFERS_GEN_TYPES(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
nullptr
};
const char kTypeSizes[] = {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
sizeof(CTYPE),
FLATBUFFERS_GEN_TYPES(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
};
// clang-format on
void Parser::Message(const std::string &msg) {
if (!error_.empty()) error_ += "\n"; // log all warnings and errors
error_ += file_being_parsed_.length() ? AbsolutePath(file_being_parsed_) : "";
// clang-format off
#ifdef _WIN32 // MSVC alike
error_ +=
"(" + NumToString(line_) + ", " + NumToString(CursorPosition()) + ")";
#else // gcc alike
if (file_being_parsed_.length()) error_ += ":";
error_ += NumToString(line_) + ": " + NumToString(CursorPosition());
#endif
// clang-format on
error_ += ": " + msg;
}
void Parser::Warning(const std::string &msg) {
if (!opts.no_warnings) {
Message("warning: " + msg);
has_warning_ = true; // for opts.warnings_as_errors
}
}
CheckedError Parser::Error(const std::string &msg) {
Message("error: " + msg);
return CheckedError(true);
}
CheckedError Parser::RecurseError() {
return Error("maximum parsing depth " + NumToString(parse_depth_counter_) +
" reached");
}
const std::string &Parser::GetPooledString(const std::string &s) const {
return *(string_cache_.insert(s).first);
}
class Parser::ParseDepthGuard {
public:
explicit ParseDepthGuard(Parser *parser_not_null)
: parser_(*parser_not_null), caller_depth_(parser_.parse_depth_counter_) {
FLATBUFFERS_ASSERT(caller_depth_ <= (FLATBUFFERS_MAX_PARSING_DEPTH) &&
"Check() must be called to prevent stack overflow");
parser_.parse_depth_counter_ += 1;
}
~ParseDepthGuard() { parser_.parse_depth_counter_ -= 1; }
CheckedError Check() {
return caller_depth_ >= (FLATBUFFERS_MAX_PARSING_DEPTH)
? parser_.RecurseError()
: CheckedError(false);
}
FLATBUFFERS_DELETE_FUNC(ParseDepthGuard(const ParseDepthGuard &));
FLATBUFFERS_DELETE_FUNC(ParseDepthGuard &operator=(const ParseDepthGuard &));
private:
Parser &parser_;
const int caller_depth_;
};
std::string Namespace::GetFullyQualifiedName(const std::string &name,
size_t max_components) const {
// Early exit if we don't have a defined namespace.
if (components.empty() || !max_components) { return name; }
std::string stream_str;
for (size_t i = 0; i < std::min(components.size(), max_components); i++) {
stream_str += components[i];
stream_str += '.';
}
if (!stream_str.empty()) stream_str.pop_back();
if (name.length()) {
stream_str += '.';
stream_str += name;
}
return stream_str;
}
std::string Parser::TokenToStringId(int t) const {
return t == kTokenIdentifier ? attribute_ : TokenToString(t);
}
// Parses exactly nibbles worth of hex digits into a number, or error.
CheckedError Parser::ParseHexNum(int nibbles, uint64_t *val) {
FLATBUFFERS_ASSERT(nibbles > 0);
for (int i = 0; i < nibbles; i++)
if (!is_xdigit(cursor_[i]))
return Error("escape code must be followed by " + NumToString(nibbles) +
" hex digits");
std::string target(cursor_, cursor_ + nibbles);
*val = StringToUInt(target.c_str(), 16);
cursor_ += nibbles;
return NoError();
}
CheckedError Parser::SkipByteOrderMark() {
if (static_cast<unsigned char>(*cursor_) != 0xef) return NoError();
cursor_++;
if (static_cast<unsigned char>(*cursor_) != 0xbb)
return Error("invalid utf-8 byte order mark");
cursor_++;
if (static_cast<unsigned char>(*cursor_) != 0xbf)
return Error("invalid utf-8 byte order mark");
cursor_++;
return NoError();
}
CheckedError Parser::Next() {
doc_comment_.clear();
bool seen_newline = cursor_ == source_;
attribute_.clear();
attr_is_trivial_ascii_string_ = true;
for (;;) {
char c = *cursor_++;
token_ = c;
switch (c) {
case '\0':
cursor_--;
token_ = kTokenEof;
return NoError();
case ' ':
case '\r':
case '\t': break;
case '\n':
MarkNewLine();
seen_newline = true;
break;
case '{':
case '}':
case '(':
case ')':
case '[':
case ']':
case ',':
case ':':
case ';':
case '=': return NoError();
case '\"':
case '\'': {
int unicode_high_surrogate = -1;
while (*cursor_ != c) {
if (*cursor_ < ' ' && static_cast<signed char>(*cursor_) >= 0)
return Error("illegal character in string constant");
if (*cursor_ == '\\') {
attr_is_trivial_ascii_string_ = false; // has escape sequence
cursor_++;
if (unicode_high_surrogate != -1 && *cursor_ != 'u') {
return Error(
"illegal Unicode sequence (unpaired high surrogate)");
}
switch (*cursor_) {
case 'n':
attribute_ += '\n';
cursor_++;
break;
case 't':
attribute_ += '\t';
cursor_++;
break;
case 'r':
attribute_ += '\r';
cursor_++;
break;
case 'b':
attribute_ += '\b';
cursor_++;
break;
case 'f':
attribute_ += '\f';
cursor_++;
break;
case '\"':
attribute_ += '\"';
cursor_++;
break;
case '\'':
attribute_ += '\'';
cursor_++;
break;
case '\\':
attribute_ += '\\';
cursor_++;
break;
case '/':
attribute_ += '/';
cursor_++;
break;
case 'x': { // Not in the JSON standard
cursor_++;
uint64_t val;
ECHECK(ParseHexNum(2, &val));
attribute_ += static_cast<char>(val);
break;
}
case 'u': {
cursor_++;
uint64_t val;
ECHECK(ParseHexNum(4, &val));
if (val >= 0xD800 && val <= 0xDBFF) {
if (unicode_high_surrogate != -1) {
return Error(
"illegal Unicode sequence (multiple high surrogates)");
} else {
unicode_high_surrogate = static_cast<int>(val);
}
} else if (val >= 0xDC00 && val <= 0xDFFF) {
if (unicode_high_surrogate == -1) {
return Error(
"illegal Unicode sequence (unpaired low surrogate)");
} else {
int code_point = 0x10000 +
((unicode_high_surrogate & 0x03FF) << 10) +
(val & 0x03FF);
ToUTF8(code_point, &attribute_);
unicode_high_surrogate = -1;
}
} else {
if (unicode_high_surrogate != -1) {
return Error(
"illegal Unicode sequence (unpaired high surrogate)");
}
ToUTF8(static_cast<int>(val), &attribute_);
}
break;
}
default: return Error("unknown escape code in string constant");
}
} else { // printable chars + UTF-8 bytes
if (unicode_high_surrogate != -1) {
return Error(
"illegal Unicode sequence (unpaired high surrogate)");
}
// reset if non-printable
attr_is_trivial_ascii_string_ &=
check_ascii_range(*cursor_, ' ', '~');
attribute_ += *cursor_++;
}
}
if (unicode_high_surrogate != -1) {
return Error("illegal Unicode sequence (unpaired high surrogate)");
}
cursor_++;
if (!attr_is_trivial_ascii_string_ && !opts.allow_non_utf8 &&
!ValidateUTF8(attribute_)) {
return Error("illegal UTF-8 sequence");
}
token_ = kTokenStringConstant;
return NoError();
}
case '/':
if (*cursor_ == '/') {
const char *start = ++cursor_;
while (*cursor_ && *cursor_ != '\n' && *cursor_ != '\r') cursor_++;
if (*start == '/') { // documentation comment
if (!seen_newline)
return Error(
"a documentation comment should be on a line on its own");
doc_comment_.push_back(std::string(start + 1, cursor_));
}
break;
} else if (*cursor_ == '*') {
cursor_++;
// TODO: make nested.
while (*cursor_ != '*' || cursor_[1] != '/') {
if (*cursor_ == '\n') MarkNewLine();
if (!*cursor_) return Error("end of file in comment");
cursor_++;
}
cursor_ += 2;
break;
}
FLATBUFFERS_FALLTHROUGH(); // else fall thru
default:
if (IsIdentifierStart(c)) {
// Collect all chars of an identifier:
const char *start = cursor_ - 1;
while (IsIdentifierStart(*cursor_) || is_digit(*cursor_)) cursor_++;
attribute_.append(start, cursor_);
token_ = kTokenIdentifier;
return NoError();
}
const auto has_sign = (c == '+') || (c == '-');
if (has_sign) {
// Check for +/-inf which is considered a float constant.
if (strncmp(cursor_, "inf", 3) == 0 &&
!(IsIdentifierStart(cursor_[3]) || is_digit(cursor_[3]))) {
attribute_.assign(cursor_ - 1, cursor_ + 3);
token_ = kTokenFloatConstant;
cursor_ += 3;
return NoError();
}
if (IsIdentifierStart(*cursor_)) {
// '-'/'+' and following identifier - it could be a predefined
// constant. Return the sign in token_, see ParseSingleValue.
return NoError();
}
}
auto dot_lvl =
(c == '.') ? 0 : 1; // dot_lvl==0 <=> exactly one '.' seen
if (!dot_lvl && !is_digit(*cursor_)) return NoError(); // enum?
// Parser accepts hexadecimal-floating-literal (see C++ 5.13.4).
if (is_digit(c) || has_sign || !dot_lvl) {
const auto start = cursor_ - 1;
auto start_digits = !is_digit(c) ? cursor_ : cursor_ - 1;
if (!is_digit(c) && is_digit(*cursor_)) {
start_digits = cursor_; // see digit in cursor_ position
c = *cursor_++;
}
// hex-float can't begind with '.'
auto use_hex = dot_lvl && (c == '0') && is_alpha_char(*cursor_, 'X');
if (use_hex) start_digits = ++cursor_; // '0x' is the prefix, skip it
// Read an integer number or mantisa of float-point number.
do {
if (use_hex) {
while (is_xdigit(*cursor_)) cursor_++;
} else {
while (is_digit(*cursor_)) cursor_++;
}
} while ((*cursor_ == '.') && (++cursor_) && (--dot_lvl >= 0));
// Exponent of float-point number.
if ((dot_lvl >= 0) && (cursor_ > start_digits)) {
// The exponent suffix of hexadecimal float number is mandatory.
if (use_hex && !dot_lvl) start_digits = cursor_;
if ((use_hex && is_alpha_char(*cursor_, 'P')) ||
is_alpha_char(*cursor_, 'E')) {
dot_lvl = 0; // Emulate dot to signal about float-point number.
cursor_++;
if (*cursor_ == '+' || *cursor_ == '-') cursor_++;
start_digits = cursor_; // the exponent-part has to have digits
// Exponent is decimal integer number
while (is_digit(*cursor_)) cursor_++;
if (*cursor_ == '.') {
cursor_++; // If see a dot treat it as part of invalid number.
dot_lvl = -1; // Fall thru to Error().
}
}
}
// Finalize.
if ((dot_lvl >= 0) && (cursor_ > start_digits)) {
attribute_.append(start, cursor_);
token_ = dot_lvl ? kTokenIntegerConstant : kTokenFloatConstant;
return NoError();
} else {
return Error("invalid number: " + std::string(start, cursor_));
}
}
std::string ch;
ch = c;
if (false == check_ascii_range(c, ' ', '~'))
ch = "code: " + NumToString(c);
return Error("illegal character: " + ch);
}
}
}
// Check if a given token is next.
bool Parser::Is(int t) const { return t == token_; }
bool Parser::IsIdent(const char *id) const {
return token_ == kTokenIdentifier && attribute_ == id;
}
// Expect a given token to be next, consume it, or error if not present.
CheckedError Parser::Expect(int t) {
if (t != token_) {
return Error("expecting: " + TokenToString(t) +
" instead got: " + TokenToStringId(token_));
}
NEXT();
return NoError();
}
CheckedError Parser::ParseNamespacing(std::string *id, std::string *last) {
while (Is('.')) {
NEXT();
*id += ".";
*id += attribute_;
if (last) *last = attribute_;
EXPECT(kTokenIdentifier);
}
return NoError();
}
EnumDef *Parser::LookupEnum(const std::string &id) {
// Search thru parent namespaces.
return LookupTableByName(enums_, id, *current_namespace_, 0);
}
StructDef *Parser::LookupStruct(const std::string &id) const {
auto sd = structs_.Lookup(id);
if (sd) sd->refcount++;
return sd;
}
StructDef *Parser::LookupStructThruParentNamespaces(
const std::string &id) const {
auto sd = LookupTableByName(structs_, id, *current_namespace_, 1);
if (sd) sd->refcount++;
return sd;
}
CheckedError Parser::ParseTypeIdent(Type &type) {
std::string id = attribute_;
EXPECT(kTokenIdentifier);
ECHECK(ParseNamespacing(&id, nullptr));
auto enum_def = LookupEnum(id);
if (enum_def) {
type = enum_def->underlying_type;
if (enum_def->is_union) type.base_type = BASE_TYPE_UNION;
} else {
type.base_type = BASE_TYPE_STRUCT;
type.struct_def = LookupCreateStruct(id);
}
return NoError();
}
// Parse any IDL type.
CheckedError Parser::ParseType(Type &type) {
if (token_ == kTokenIdentifier) {
if (IsIdent("bool")) {
type.base_type = BASE_TYPE_BOOL;
NEXT();
} else if (IsIdent("byte") || IsIdent("int8")) {
type.base_type = BASE_TYPE_CHAR;
NEXT();
} else if (IsIdent("ubyte") || IsIdent("uint8")) {
type.base_type = BASE_TYPE_UCHAR;
NEXT();
} else if (IsIdent("short") || IsIdent("int16")) {
type.base_type = BASE_TYPE_SHORT;
NEXT();
} else if (IsIdent("ushort") || IsIdent("uint16")) {
type.base_type = BASE_TYPE_USHORT;
NEXT();
} else if (IsIdent("int") || IsIdent("int32")) {
type.base_type = BASE_TYPE_INT;
NEXT();
} else if (IsIdent("uint") || IsIdent("uint32")) {
type.base_type = BASE_TYPE_UINT;
NEXT();
} else if (IsIdent("long") || IsIdent("int64")) {
type.base_type = BASE_TYPE_LONG;
NEXT();
} else if (IsIdent("ulong") || IsIdent("uint64")) {
type.base_type = BASE_TYPE_ULONG;
NEXT();
} else if (IsIdent("float") || IsIdent("float32")) {
type.base_type = BASE_TYPE_FLOAT;
NEXT();
} else if (IsIdent("double") || IsIdent("float64")) {
type.base_type = BASE_TYPE_DOUBLE;
NEXT();
} else if (IsIdent("string")) {
type.base_type = BASE_TYPE_STRING;
NEXT();
} else {
ECHECK(ParseTypeIdent(type));
}
} else if (token_ == '[') {
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
NEXT();
Type subtype;
ECHECK(ParseType(subtype));
if (IsSeries(subtype)) {
// We could support this, but it will complicate things, and it's
// easier to work around with a struct around the inner vector.
return Error("nested vector types not supported (wrap in table first)");
}
if (token_ == ':') {
NEXT();
if (token_ != kTokenIntegerConstant) {
return Error("length of fixed-length array must be an integer value");
}
uint16_t fixed_length = 0;
bool check = StringToNumber(attribute_.c_str(), &fixed_length);
if (!check || fixed_length < 1) {
return Error(
"length of fixed-length array must be positive and fit to "
"uint16_t type");
}
type = Type(BASE_TYPE_ARRAY, subtype.struct_def, subtype.enum_def,
fixed_length);
NEXT();
} else {
type = Type(BASE_TYPE_VECTOR, subtype.struct_def, subtype.enum_def);
}
type.element = subtype.base_type;
EXPECT(']');
} else {
return Error("illegal type syntax");
}
return NoError();
}
CheckedError Parser::AddField(StructDef &struct_def, const std::string &name,
const Type &type, FieldDef **dest) {
auto &field = *new FieldDef();
field.value.offset =
FieldIndexToOffset(static_cast<voffset_t>(struct_def.fields.vec.size()));
field.name = name;
field.file = struct_def.file;
field.value.type = type;
if (struct_def.fixed) { // statically compute the field offset
auto size = InlineSize(type);
auto alignment = InlineAlignment(type);
// structs_ need to have a predictable format, so we need to align to
// the largest scalar
struct_def.minalign = std::max(struct_def.minalign, alignment);
struct_def.PadLastField(alignment);
field.value.offset = static_cast<voffset_t>(struct_def.bytesize);
struct_def.bytesize += size;
}
if (struct_def.fields.Add(name, &field))
return Error("field already exists: " + name);
*dest = &field;
return NoError();
}
CheckedError Parser::ParseField(StructDef &struct_def) {
std::string name = attribute_;
if (LookupCreateStruct(name, false, false))
return Error("field name can not be the same as table/struct name");
if (!IsLowerSnakeCase(name)) {
Warning("field names should be lowercase snake_case, got: " + name);
}
std::vector<std::string> dc = doc_comment_;
EXPECT(kTokenIdentifier);
EXPECT(':');
Type type;
ECHECK(ParseType(type));
if (struct_def.fixed) {
auto valid = IsScalar(type.base_type) || IsStruct(type);
if (!valid && IsArray(type)) {
const auto &elem_type = type.VectorType();
valid |= IsScalar(elem_type.base_type) || IsStruct(elem_type);
}
if (!valid)
return Error("structs may contain only scalar or struct fields");
}
if (!struct_def.fixed && IsArray(type))
return Error("fixed-length array in table must be wrapped in struct");
if (IsArray(type)) {
advanced_features_ |= reflection::AdvancedArrayFeatures;
if (!SupportsAdvancedArrayFeatures()) {
return Error(
"Arrays are not yet supported in all "
"the specified programming languages.");
}
}
FieldDef *typefield = nullptr;
if (type.base_type == BASE_TYPE_UNION) {
// For union fields, add a second auto-generated field to hold the type,
// with a special suffix.
ECHECK(AddField(struct_def, name + UnionTypeFieldSuffix(),
type.enum_def->underlying_type, &typefield));
} else if (IsVector(type) && type.element == BASE_TYPE_UNION) {
advanced_features_ |= reflection::AdvancedUnionFeatures;
// Only cpp, js and ts supports the union vector feature so far.
if (!SupportsAdvancedUnionFeatures()) {
return Error(
"Vectors of unions are not yet supported in at least one of "
"the specified programming languages.");
}
// For vector of union fields, add a second auto-generated vector field to
// hold the types, with a special suffix.
Type union_vector(BASE_TYPE_VECTOR, nullptr, type.enum_def);
union_vector.element = BASE_TYPE_UTYPE;
ECHECK(AddField(struct_def, name + UnionTypeFieldSuffix(), union_vector,
&typefield));
}
FieldDef *field;
ECHECK(AddField(struct_def, name, type, &field));
if (token_ == '=') {
NEXT();
ECHECK(ParseSingleValue(&field->name, field->value, true));
if (IsStruct(type) || (struct_def.fixed && field->value.constant != "0"))
return Error(
"default values are not supported for struct fields, table fields, "
"or in structs.");
if (IsString(type) || IsVector(type)) {
advanced_features_ |= reflection::DefaultVectorsAndStrings;
if (field->value.constant != "0" && !SupportsDefaultVectorsAndStrings()) {
return Error(
"Default values for strings and vectors are not supported in one "
"of the specified programming languages");
}
}
if (IsVector(type) && field->value.constant != "0" &&
field->value.constant != "[]") {
return Error("The only supported default for vectors is `[]`.");
}
}
// Append .0 if the value has not it (skip hex and scientific floats).
// This suffix needed for generated C++ code.
if (IsFloat(type.base_type)) {
auto &text = field->value.constant;
FLATBUFFERS_ASSERT(false == text.empty());
auto s = text.c_str();
while (*s == ' ') s++;
if (*s == '-' || *s == '+') s++;
// 1) A float constants (nan, inf, pi, etc) is a kind of identifier.
// 2) A float number needn't ".0" at the end if it has exponent.
if ((false == IsIdentifierStart(*s)) &&
(std::string::npos == field->value.constant.find_first_of(".eEpP"))) {
field->value.constant += ".0";
}
}
field->doc_comment = dc;
ECHECK(ParseMetaData(&field->attributes));
field->deprecated = field->attributes.Lookup("deprecated") != nullptr;
auto hash_name = field->attributes.Lookup("hash");
if (hash_name) {
switch ((IsVector(type)) ? type.element : type.base_type) {
case BASE_TYPE_SHORT:
case BASE_TYPE_USHORT: {
if (FindHashFunction16(hash_name->constant.c_str()) == nullptr)
return Error("Unknown hashing algorithm for 16 bit types: " +
hash_name->constant);
break;
}
case BASE_TYPE_INT:
case BASE_TYPE_UINT: {
if (FindHashFunction32(hash_name->constant.c_str()) == nullptr)
return Error("Unknown hashing algorithm for 32 bit types: " +
hash_name->constant);
break;
}
case BASE_TYPE_LONG:
case BASE_TYPE_ULONG: {
if (FindHashFunction64(hash_name->constant.c_str()) == nullptr)
return Error("Unknown hashing algorithm for 64 bit types: " +
hash_name->constant);
break;
}
default:
return Error(
"only short, ushort, int, uint, long and ulong data types support "
"hashing.");
}
}
// For historical convenience reasons, string keys are assumed required.
// Scalars are kDefault unless otherwise specified.
// Nonscalars are kOptional unless required;
field->key = field->attributes.Lookup("key") != nullptr;
const bool required = field->attributes.Lookup("required") != nullptr ||
(IsString(type) && field->key);
const bool default_str_or_vec =
((IsString(type) || IsVector(type)) && field->value.constant != "0");
const bool optional = IsScalar(type.base_type)
? (field->value.constant == "null")
: !(required || default_str_or_vec);
if (required && optional) {
return Error("Fields cannot be both optional and required.");
}
field->presence = FieldDef::MakeFieldPresence(optional, required);
if (required && (struct_def.fixed || IsScalar(type.base_type))) {
return Error("only non-scalar fields in tables may be 'required'");
}
if (field->key) {
if (struct_def.has_key) return Error("only one field may be set as 'key'");
struct_def.has_key = true;
if (!IsScalar(type.base_type) && !IsString(type)) {
return Error("'key' field must be string or scalar type");
}
}
if (field->IsScalarOptional()) {
advanced_features_ |= reflection::OptionalScalars;
if (type.enum_def && type.enum_def->Lookup("null")) {
FLATBUFFERS_ASSERT(IsInteger(type.base_type));
return Error(
"the default 'null' is reserved for declaring optional scalar "
"fields, it conflicts with declaration of enum '" +
type.enum_def->name + "'.");
}
if (field->attributes.Lookup("key")) {
return Error(
"only a non-optional scalar field can be used as a 'key' field");
}
if (!SupportsOptionalScalars()) {
return Error(
"Optional scalars are not yet supported in at least one of "
"the specified programming languages.");
}
}
if (type.enum_def) {
// Verify the enum's type and default value.
const std::string &constant = field->value.constant;
if (type.base_type == BASE_TYPE_UNION) {
if (constant != "0") { return Error("Union defaults must be NONE"); }
} else if (IsVector(type)) {
if (constant != "0" && constant != "[]") {
return Error("Vector defaults may only be `[]`.");
}
} else if (IsArray(type)) {
if (constant != "0") {
return Error("Array defaults are not supported yet.");
}
} else {
if (!IsInteger(type.base_type)) {
return Error("Enums must have integer base types");
}
// Optional and bitflags enums may have default constants that are not
// their specified variants.
if (!field->IsOptional() &&
type.enum_def->attributes.Lookup("bit_flags") == nullptr) {
if (type.enum_def->FindByValue(constant) == nullptr) {
return Error("default value of `" + constant + "` for " + "field `" +
name + "` is not part of enum `" + type.enum_def->name +
"`.");
}
}
}
}
if (field->deprecated && struct_def.fixed)
return Error("can't deprecate fields in a struct");
auto cpp_type = field->attributes.Lookup("cpp_type");
if (cpp_type) {
if (!hash_name)
return Error("cpp_type can only be used with a hashed field");
/// forcing cpp_ptr_type to 'naked' if unset
auto cpp_ptr_type = field->attributes.Lookup("cpp_ptr_type");
if (!cpp_ptr_type) {
auto val = new Value();
val->type = cpp_type->type;
val->constant = "naked";
field->attributes.Add("cpp_ptr_type", val);
}
}
field->shared = field->attributes.Lookup("shared") != nullptr;
if (field->shared && field->value.type.base_type != BASE_TYPE_STRING)
return Error("shared can only be defined on strings");
auto field_native_custom_alloc =
field->attributes.Lookup("native_custom_alloc");
if (field_native_custom_alloc)
return Error(
"native_custom_alloc can only be used with a table or struct "
"definition");
field->native_inline = field->attributes.Lookup("native_inline") != nullptr;
if (field->native_inline && !IsStruct(field->value.type) &&
!IsVectorOfStruct(field->value.type) &&
!IsVectorOfTable(field->value.type))
return Error(
"'native_inline' can only be defined on structs, vector of structs or "
"vector of tables");
auto nested = field->attributes.Lookup("nested_flatbuffer");
if (nested) {
if (nested->type.base_type != BASE_TYPE_STRING)
return Error(
"nested_flatbuffer attribute must be a string (the root type)");
if (type.base_type != BASE_TYPE_VECTOR || type.element != BASE_TYPE_UCHAR)
return Error(
"nested_flatbuffer attribute may only apply to a vector of ubyte");
// This will cause an error if the root type of the nested flatbuffer
// wasn't defined elsewhere.
field->nested_flatbuffer = LookupCreateStruct(nested->constant);
}
if (field->attributes.Lookup("flexbuffer")) {
field->flexbuffer = true;
uses_flexbuffers_ = true;
if (type.base_type != BASE_TYPE_VECTOR || type.element != BASE_TYPE_UCHAR)
return Error("flexbuffer attribute may only apply to a vector of ubyte");
}
if (typefield) {
if (!IsScalar(typefield->value.type.base_type)) {
// this is a union vector field
typefield->presence = field->presence;
}
// If this field is a union, and it has a manually assigned id,
// the automatically added type field should have an id as well (of N - 1).
auto attr = field->attributes.Lookup("id");
if (attr) {
const auto &id_str = attr->constant;
voffset_t id = 0;
const auto done = !atot(id_str.c_str(), *this, &id).Check();
if (done && id > 0) {
auto val = new Value();
val->type = attr->type;
val->constant = NumToString(id - 1);
typefield->attributes.Add("id", val);
} else {
return Error(
"a union type effectively adds two fields with non-negative ids, "
"its id must be that of the second field (the first field is "
"the type field and not explicitly declared in the schema);\n"
"field: " +
field->name + ", id: " + id_str);
}
}
// if this field is a union that is deprecated,
// the automatically added type field should be deprecated as well
if (field->deprecated) { typefield->deprecated = true; }
}
EXPECT(';');
return NoError();
}
CheckedError Parser::ParseString(Value &val, bool use_string_pooling) {
auto s = attribute_;
EXPECT(kTokenStringConstant);
if (use_string_pooling) {
val.constant = NumToString(builder_.CreateSharedString(s).o);
} else {
val.constant = NumToString(builder_.CreateString(s).o);
}
return NoError();
}
CheckedError Parser::ParseComma() {
if (!opts.protobuf_ascii_alike) EXPECT(',');
return NoError();
}
CheckedError Parser::ParseAnyValue(Value &val, FieldDef *field,
size_t parent_fieldn,
const StructDef *parent_struct_def,
uoffset_t count, bool inside_vector) {
switch (val.type.base_type) {
case BASE_TYPE_UNION: {
FLATBUFFERS_ASSERT(field);
std::string constant;
Vector<uint8_t> *vector_of_union_types = nullptr;
// Find corresponding type field we may have already parsed.
for (auto elem = field_stack_.rbegin() + count;
elem != field_stack_.rbegin() + parent_fieldn + count; ++elem) {
auto &type = elem->second->value.type;
if (type.enum_def == val.type.enum_def) {
if (inside_vector) {
if (IsVector(type) && type.element == BASE_TYPE_UTYPE) {
// Vector of union type field.
uoffset_t offset;
ECHECK(atot(elem->first.constant.c_str(), *this, &offset));
vector_of_union_types = reinterpret_cast<Vector<uint8_t> *>(
builder_.GetCurrentBufferPointer() + builder_.GetSize() -
offset);
break;
}
} else {
if (type.base_type == BASE_TYPE_UTYPE) {
// Union type field.
constant = elem->first.constant;
break;
}
}
}
}
if (constant.empty() && !inside_vector) {
// We haven't seen the type field yet. Sadly a lot of JSON writers
// output these in alphabetical order, meaning it comes after this
// value. So we scan past the value to find it, then come back here.
// We currently don't do this for vectors of unions because the
// scanning/serialization logic would get very complicated.
auto type_name = field->name + UnionTypeFieldSuffix();
FLATBUFFERS_ASSERT(parent_struct_def);
auto type_field = parent_struct_def->fields.Lookup(type_name);
FLATBUFFERS_ASSERT(type_field); // Guaranteed by ParseField().
// Remember where we are in the source file, so we can come back here.
auto backup = *static_cast<ParserState *>(this);
ECHECK(SkipAnyJsonValue()); // The table.
ECHECK(ParseComma());
auto next_name = attribute_;
if (Is(kTokenStringConstant)) {
NEXT();
} else {
EXPECT(kTokenIdentifier);
}
if (next_name == type_name) {
EXPECT(':');
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
Value type_val = type_field->value;
ECHECK(ParseAnyValue(type_val, type_field, 0, nullptr, 0));
constant = type_val.constant;
// Got the information we needed, now rewind:
*static_cast<ParserState *>(this) = backup;
}
}
if (constant.empty() && !vector_of_union_types) {
return Error("missing type field for this union value: " + field->name);
}
uint8_t enum_idx;
if (vector_of_union_types) {
if (vector_of_union_types->size() <= count)
return Error(
"union types vector smaller than union values vector for: " +
field->name);
enum_idx = vector_of_union_types->Get(count);
} else {
ECHECK(atot(constant.c_str(), *this, &enum_idx));
}
auto enum_val = val.type.enum_def->ReverseLookup(enum_idx, true);
if (!enum_val) return Error("illegal type id for: " + field->name);
if (enum_val->union_type.base_type == BASE_TYPE_STRUCT) {
ECHECK(ParseTable(*enum_val->union_type.struct_def, &val.constant,
nullptr));
if (enum_val->union_type.struct_def->fixed) {
// All BASE_TYPE_UNION values are offsets, so turn this into one.
SerializeStruct(*enum_val->union_type.struct_def, val);
builder_.ClearOffsets();
val.constant = NumToString(builder_.GetSize());
}
} else if (IsString(enum_val->union_type)) {
ECHECK(ParseString(val, field->shared));
} else {
FLATBUFFERS_ASSERT(false);
}
break;
}
case BASE_TYPE_STRUCT:
ECHECK(ParseTable(*val.type.struct_def, &val.constant, nullptr));
break;
case BASE_TYPE_STRING: {
ECHECK(ParseString(val, field->shared));
break;
}
case BASE_TYPE_VECTOR: {
uoffset_t off;
ECHECK(ParseVector(val.type.VectorType(), &off, field, parent_fieldn));
val.constant = NumToString(off);
break;
}
case BASE_TYPE_ARRAY: {
ECHECK(ParseArray(val));
break;
}
case BASE_TYPE_INT:
case BASE_TYPE_UINT:
case BASE_TYPE_LONG:
case BASE_TYPE_ULONG: {
if (field && field->attributes.Lookup("hash") &&
(token_ == kTokenIdentifier || token_ == kTokenStringConstant)) {
ECHECK(ParseHash(val, field));
} else {
ECHECK(ParseSingleValue(field ? &field->name : nullptr, val, false));
}
break;
}
default:
ECHECK(ParseSingleValue(field ? &field->name : nullptr, val, false));
break;
}
return NoError();
}
void Parser::SerializeStruct(const StructDef &struct_def, const Value &val) {
SerializeStruct(builder_, struct_def, val);
}
void Parser::SerializeStruct(FlatBufferBuilder &builder,
const StructDef &struct_def, const Value &val) {
FLATBUFFERS_ASSERT(val.constant.length() == struct_def.bytesize);
builder.Align(struct_def.minalign);
builder.PushBytes(reinterpret_cast<const uint8_t *>(val.constant.c_str()),
struct_def.bytesize);
builder.AddStructOffset(val.offset, builder.GetSize());
}
template<typename F>
CheckedError Parser::ParseTableDelimiters(size_t &fieldn,
const StructDef *struct_def, F body) {
// We allow tables both as JSON object{ .. } with field names
// or vector[..] with all fields in order
char terminator = '}';
bool is_nested_vector = struct_def && Is('[');
if (is_nested_vector) {
NEXT();
terminator = ']';
} else {
EXPECT('{');
}
for (;;) {
if ((!opts.strict_json || !fieldn) && Is(terminator)) break;
std::string name;
if (is_nested_vector) {
if (fieldn >= struct_def->fields.vec.size()) {
return Error("too many unnamed fields in nested array");
}
name = struct_def->fields.vec[fieldn]->name;
} else {
name = attribute_;
if (Is(kTokenStringConstant)) {
NEXT();
} else {
EXPECT(opts.strict_json ? kTokenStringConstant : kTokenIdentifier);
}
if (!opts.protobuf_ascii_alike || !(Is('{') || Is('['))) EXPECT(':');
}
ECHECK(body(name, fieldn, struct_def));
if (Is(terminator)) break;
ECHECK(ParseComma());
}
NEXT();
if (is_nested_vector && fieldn != struct_def->fields.vec.size()) {
return Error("wrong number of unnamed fields in table vector");
}
return NoError();
}
CheckedError Parser::ParseTable(const StructDef &struct_def, std::string *value,
uoffset_t *ovalue) {
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
size_t fieldn_outer = 0;
auto err = ParseTableDelimiters(
fieldn_outer, &struct_def,
[&](const std::string &name, size_t &fieldn,
const StructDef *struct_def_inner) -> CheckedError {
if (name == "$schema") {
ECHECK(Expect(kTokenStringConstant));
return NoError();
}
auto field = struct_def_inner->fields.Lookup(name);
if (!field) {
if (!opts.skip_unexpected_fields_in_json) {
return Error("unknown field: " + name);
} else {
ECHECK(SkipAnyJsonValue());
}
} else {
if (IsIdent("null") && !IsScalar(field->value.type.base_type)) {
ECHECK(Next()); // Ignore this field.
} else {
Value val = field->value;
if (field->flexbuffer) {
flexbuffers::Builder builder(1024,
flexbuffers::BUILDER_FLAG_SHARE_ALL);
ECHECK(ParseFlexBufferValue(&builder));
builder.Finish();
// Force alignment for nested flexbuffer
builder_.ForceVectorAlignment(builder.GetSize(), sizeof(uint8_t),
sizeof(largest_scalar_t));
auto off = builder_.CreateVector(builder.GetBuffer());
val.constant = NumToString(off.o);
} else if (field->nested_flatbuffer) {
ECHECK(
ParseNestedFlatbuffer(val, field, fieldn, struct_def_inner));
} else {
ECHECK(ParseAnyValue(val, field, fieldn, struct_def_inner, 0));
}
// Hardcoded insertion-sort with error-check.
// If fields are specified in order, then this loop exits
// immediately.
auto elem = field_stack_.rbegin();
for (; elem != field_stack_.rbegin() + fieldn; ++elem) {
auto existing_field = elem->second;
if (existing_field == field)
return Error("field set more than once: " + field->name);
if (existing_field->value.offset < field->value.offset) break;
}
// Note: elem points to before the insertion point, thus .base()
// points to the correct spot.
field_stack_.insert(elem.base(), std::make_pair(val, field));
fieldn++;
}
}
return NoError();
});
ECHECK(err);
// Check if all required fields are parsed.
for (auto field_it = struct_def.fields.vec.begin();
field_it != struct_def.fields.vec.end(); ++field_it) {
auto required_field = *field_it;
if (!required_field->IsRequired()) { continue; }
bool found = false;
for (auto pf_it = field_stack_.end() - fieldn_outer;
pf_it != field_stack_.end(); ++pf_it) {
auto parsed_field = pf_it->second;
if (parsed_field == required_field) {
found = true;
break;
}
}
if (!found) {
return Error("required field is missing: " + required_field->name +
" in " + struct_def.name);
}
}
if (struct_def.fixed && fieldn_outer != struct_def.fields.vec.size())
return Error("struct: wrong number of initializers: " + struct_def.name);
auto start = struct_def.fixed ? builder_.StartStruct(struct_def.minalign)
: builder_.StartTable();
for (size_t size = struct_def.sortbysize ? sizeof(largest_scalar_t) : 1; size;
size /= 2) {
// Go through elements in reverse, since we're building the data backwards.
for (auto it = field_stack_.rbegin();
it != field_stack_.rbegin() + fieldn_outer; ++it) {
auto &field_value = it->first;
auto field = it->second;
if (!struct_def.sortbysize ||
size == SizeOf(field_value.type.base_type)) {
switch (field_value.type.base_type) {
// clang-format off
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_ ## ENUM: \
builder_.Pad(field->padding); \
if (struct_def.fixed) { \
CTYPE val; \
ECHECK(atot(field_value.constant.c_str(), *this, &val)); \
builder_.PushElement(val); \
} else { \
if (field->IsScalarOptional()) { \
if (field_value.constant != "null") { \
CTYPE val; \
ECHECK(atot(field_value.constant.c_str(), *this, &val)); \
builder_.AddElement(field_value.offset, val); \
} \
} else { \
CTYPE val, valdef; \
ECHECK(atot(field_value.constant.c_str(), *this, &val)); \
ECHECK(atot(field->value.constant.c_str(), *this, &valdef)); \
builder_.AddElement(field_value.offset, val, valdef); \
} \
} \
break;
FLATBUFFERS_GEN_TYPES_SCALAR(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_ ## ENUM: \
builder_.Pad(field->padding); \
if (IsStruct(field->value.type)) { \
SerializeStruct(*field->value.type.struct_def, field_value); \
} else { \
CTYPE val; \
ECHECK(atot(field_value.constant.c_str(), *this, &val)); \
builder_.AddOffset(field_value.offset, val); \
} \
break;
FLATBUFFERS_GEN_TYPES_POINTER(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
case BASE_TYPE_ARRAY:
builder_.Pad(field->padding);
builder_.PushBytes(
reinterpret_cast<const uint8_t*>(field_value.constant.c_str()),
InlineSize(field_value.type));
break;
// clang-format on
}
}
}
}
for (size_t i = 0; i < fieldn_outer; i++) field_stack_.pop_back();
if (struct_def.fixed) {
builder_.ClearOffsets();
builder_.EndStruct();
FLATBUFFERS_ASSERT(value);
// Temporarily store this struct in the value string, since it is to
// be serialized in-place elsewhere.
value->assign(
reinterpret_cast<const char *>(builder_.GetCurrentBufferPointer()),
struct_def.bytesize);
builder_.PopBytes(struct_def.bytesize);
FLATBUFFERS_ASSERT(!ovalue);
} else {
auto val = builder_.EndTable(start);
if (ovalue) *ovalue = val;
if (value) *value = NumToString(val);
}
return NoError();
}
template<typename F>
CheckedError Parser::ParseVectorDelimiters(uoffset_t &count, F body) {
EXPECT('[');
for (;;) {
if ((!opts.strict_json || !count) && Is(']')) break;
ECHECK(body(count));
count++;
if (Is(']')) break;
ECHECK(ParseComma());
}
NEXT();
return NoError();
}
CheckedError Parser::ParseAlignAttribute(const std::string &align_constant,
size_t min_align, size_t *align) {
// Use uint8_t to avoid problems with size_t==`unsigned long` on LP64.
uint8_t align_value;
if (StringToNumber(align_constant.c_str(), &align_value) &&
VerifyAlignmentRequirements(static_cast<size_t>(align_value),
min_align)) {
*align = align_value;
return NoError();
}
return Error("unexpected force_align value '" + align_constant +
"', alignment must be a power of two integer ranging from the "
"type\'s natural alignment " +
NumToString(min_align) + " to " +
NumToString(FLATBUFFERS_MAX_ALIGNMENT));
}
CheckedError Parser::ParseVector(const Type &type, uoffset_t *ovalue,
FieldDef *field, size_t fieldn) {
uoffset_t count = 0;
auto err = ParseVectorDelimiters(count, [&](uoffset_t &) -> CheckedError {
Value val;
val.type = type;
ECHECK(ParseAnyValue(val, field, fieldn, nullptr, count, true));
field_stack_.push_back(std::make_pair(val, nullptr));
return NoError();
});
ECHECK(err);
const size_t alignment = InlineAlignment(type);
const size_t len = count * InlineSize(type) / InlineAlignment(type);
const size_t elemsize = InlineAlignment(type);
const auto force_align = field->attributes.Lookup("force_align");
if (force_align) {
size_t align;
ECHECK(ParseAlignAttribute(force_align->constant, 1, &align));
if (align > 1) { builder_.ForceVectorAlignment(len, elemsize, align); }
}
// TODO Fix using element alignment as size (`elemsize`)!
builder_.StartVector(len, elemsize, alignment);
for (uoffset_t i = 0; i < count; i++) {
// start at the back, since we're building the data backwards.
auto &val = field_stack_.back().first;
switch (val.type.base_type) {
// clang-format off
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE,...) \
case BASE_TYPE_ ## ENUM: \
if (IsStruct(val.type)) SerializeStruct(*val.type.struct_def, val); \
else { \
CTYPE elem; \
ECHECK(atot(val.constant.c_str(), *this, &elem)); \
builder_.PushElement(elem); \
} \
break;
FLATBUFFERS_GEN_TYPES(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
// clang-format on
}
field_stack_.pop_back();
}
builder_.ClearOffsets();
*ovalue = builder_.EndVector(count);
if (type.base_type == BASE_TYPE_STRUCT && type.struct_def->has_key) {
// We should sort this vector. Find the key first.
const FieldDef *key = nullptr;
for (auto it = type.struct_def->fields.vec.begin();
it != type.struct_def->fields.vec.end(); ++it) {
if ((*it)->key) {
key = (*it);
break;
}
}
FLATBUFFERS_ASSERT(key);
// Now sort it.
// We can't use std::sort because for structs the size is not known at
// compile time, and for tables our iterators dereference offsets, so can't
// be used to swap elements.
// And we can't use C qsort either, since that would force use to use
// globals, making parsing thread-unsafe.
// So for now, we use SimpleQsort above.
// TODO: replace with something better, preferably not recursive.
if (type.struct_def->fixed) {
const voffset_t offset = key->value.offset;
const size_t struct_size = type.struct_def->bytesize;
auto v =
reinterpret_cast<VectorOfAny *>(builder_.GetCurrentBufferPointer());
SimpleQsort<uint8_t>(
v->Data(), v->Data() + v->size() * type.struct_def->bytesize,
type.struct_def->bytesize,
[offset, key](const uint8_t *a, const uint8_t *b) -> bool {
return CompareSerializedScalars(a + offset, b + offset, *key);
},
[struct_size](uint8_t *a, uint8_t *b) {
// FIXME: faster?
for (size_t i = 0; i < struct_size; i++) { std::swap(a[i], b[i]); }
});
} else {
auto v = reinterpret_cast<Vector<Offset<Table>> *>(
builder_.GetCurrentBufferPointer());
// Here also can't use std::sort. We do have an iterator type for it,
// but it is non-standard as it will dereference the offsets, and thus
// can't be used to swap elements.
if (key->value.type.base_type == BASE_TYPE_STRING) {
SimpleQsort<Offset<Table>>(
v->data(), v->data() + v->size(), 1,
[key](const Offset<Table> *_a, const Offset<Table> *_b) -> bool {
return CompareTablesByStringKey(_a, _b, *key);
},
SwapSerializedTables);
} else {
SimpleQsort<Offset<Table>>(
v->data(), v->data() + v->size(), 1,
[key](const Offset<Table> *_a, const Offset<Table> *_b) -> bool {
return CompareTablesByScalarKey(_a, _b, *key);
},
SwapSerializedTables);
}
}
}
return NoError();
}
CheckedError Parser::ParseArray(Value &array) {
std::vector<Value> stack;
FlatBufferBuilder builder;
const auto &type = array.type.VectorType();
auto length = array.type.fixed_length;
uoffset_t count = 0;
auto err = ParseVectorDelimiters(count, [&](uoffset_t &) -> CheckedError {
stack.emplace_back(Value());
auto &val = stack.back();
val.type = type;
if (IsStruct(type)) {
ECHECK(ParseTable(*val.type.struct_def, &val.constant, nullptr));
} else {
ECHECK(ParseSingleValue(nullptr, val, false));
}
return NoError();
});
ECHECK(err);
if (length != count) return Error("Fixed-length array size is incorrect.");
for (auto it = stack.rbegin(); it != stack.rend(); ++it) {
auto &val = *it;
// clang-format off
switch (val.type.base_type) {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_ ## ENUM: \
if (IsStruct(val.type)) { \
SerializeStruct(builder, *val.type.struct_def, val); \
} else { \
CTYPE elem; \
ECHECK(atot(val.constant.c_str(), *this, &elem)); \
builder.PushElement(elem); \
} \
break;
FLATBUFFERS_GEN_TYPES(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
default: FLATBUFFERS_ASSERT(0);
}
// clang-format on
}
array.constant.assign(
reinterpret_cast<const char *>(builder.GetCurrentBufferPointer()),
InlineSize(array.type));
return NoError();
}
CheckedError Parser::ParseNestedFlatbuffer(Value &val, FieldDef *field,
size_t fieldn,
const StructDef *parent_struct_def) {
if (token_ == '[') { // backwards compat for 'legacy' ubyte buffers
if (opts.json_nested_legacy_flatbuffers) {
ECHECK(ParseAnyValue(val, field, fieldn, parent_struct_def, 0));
} else {
return Error(
"cannot parse nested_flatbuffer as bytes unless"
" --json-nested-bytes is set");
}
} else {
auto cursor_at_value_begin = cursor_;
ECHECK(SkipAnyJsonValue());
std::string substring(cursor_at_value_begin - 1, cursor_ - 1);
// Create and initialize new parser
Parser nested_parser;
FLATBUFFERS_ASSERT(field->nested_flatbuffer);
nested_parser.root_struct_def_ = field->nested_flatbuffer;
nested_parser.enums_ = enums_;
nested_parser.opts = opts;
nested_parser.uses_flexbuffers_ = uses_flexbuffers_;
nested_parser.parse_depth_counter_ = parse_depth_counter_;
// Parse JSON substring into new flatbuffer builder using nested_parser
bool ok = nested_parser.Parse(substring.c_str(), nullptr, nullptr);
// Clean nested_parser to avoid deleting the elements in
// the SymbolTables on destruction
nested_parser.enums_.dict.clear();
nested_parser.enums_.vec.clear();
if (!ok) { ECHECK(Error(nested_parser.error_)); }
// Force alignment for nested flatbuffer
builder_.ForceVectorAlignment(
nested_parser.builder_.GetSize(), sizeof(uint8_t),
nested_parser.builder_.GetBufferMinAlignment());
auto off = builder_.CreateVector(nested_parser.builder_.GetBufferPointer(),
nested_parser.builder_.GetSize());
val.constant = NumToString(off.o);
}
return NoError();
}
CheckedError Parser::ParseMetaData(SymbolTable<Value> *attributes) {
if (Is('(')) {
NEXT();
for (;;) {
auto name = attribute_;
if (false == (Is(kTokenIdentifier) || Is(kTokenStringConstant)))
return Error("attribute name must be either identifier or string: " +
name);
if (known_attributes_.find(name) == known_attributes_.end())
return Error("user define attributes must be declared before use: " +
name);
NEXT();
auto e = new Value();
if (attributes->Add(name, e)) Warning("attribute already found: " + name);
if (Is(':')) {
NEXT();
ECHECK(ParseSingleValue(&name, *e, true));
}
if (Is(')')) {
NEXT();
break;
}
EXPECT(',');
}
}
return NoError();
}
CheckedError Parser::ParseEnumFromString(const Type &type,
std::string *result) {
const auto base_type =
type.enum_def ? type.enum_def->underlying_type.base_type : type.base_type;
if (!IsInteger(base_type)) return Error("not a valid value for this field");
uint64_t u64 = 0;
for (size_t pos = 0; pos != std::string::npos;) {
const auto delim = attribute_.find_first_of(' ', pos);
const auto last = (std::string::npos == delim);
auto word = attribute_.substr(pos, !last ? delim - pos : std::string::npos);
pos = !last ? delim + 1 : std::string::npos;
const EnumVal *ev = nullptr;
if (type.enum_def) {
ev = type.enum_def->Lookup(word);
} else {
auto dot = word.find_first_of('.');
if (std::string::npos == dot)
return Error("enum values need to be qualified by an enum type");
auto enum_def_str = word.substr(0, dot);
const auto enum_def = LookupEnum(enum_def_str);
if (!enum_def) return Error("unknown enum: " + enum_def_str);
auto enum_val_str = word.substr(dot + 1);
ev = enum_def->Lookup(enum_val_str);
}
if (!ev) return Error("unknown enum value: " + word);
u64 |= ev->GetAsUInt64();
}
*result = IsUnsigned(base_type) ? NumToString(u64)
: NumToString(static_cast<int64_t>(u64));
return NoError();
}
CheckedError Parser::ParseHash(Value &e, FieldDef *field) {
FLATBUFFERS_ASSERT(field);
Value *hash_name = field->attributes.Lookup("hash");
switch (e.type.base_type) {
case BASE_TYPE_SHORT: {
auto hash = FindHashFunction16(hash_name->constant.c_str());
int16_t hashed_value = static_cast<int16_t>(hash(attribute_.c_str()));
e.constant = NumToString(hashed_value);
break;
}
case BASE_TYPE_USHORT: {
auto hash = FindHashFunction16(hash_name->constant.c_str());
uint16_t hashed_value = hash(attribute_.c_str());
e.constant = NumToString(hashed_value);
break;
}
case BASE_TYPE_INT: {
auto hash = FindHashFunction32(hash_name->constant.c_str());
int32_t hashed_value = static_cast<int32_t>(hash(attribute_.c_str()));
e.constant = NumToString(hashed_value);
break;
}
case BASE_TYPE_UINT: {
auto hash = FindHashFunction32(hash_name->constant.c_str());
uint32_t hashed_value = hash(attribute_.c_str());
e.constant = NumToString(hashed_value);
break;
}
case BASE_TYPE_LONG: {
auto hash = FindHashFunction64(hash_name->constant.c_str());
int64_t hashed_value = static_cast<int64_t>(hash(attribute_.c_str()));
e.constant = NumToString(hashed_value);
break;
}
case BASE_TYPE_ULONG: {
auto hash = FindHashFunction64(hash_name->constant.c_str());
uint64_t hashed_value = hash(attribute_.c_str());
e.constant = NumToString(hashed_value);
break;
}
default: FLATBUFFERS_ASSERT(0);
}
NEXT();
return NoError();
}
CheckedError Parser::TokenError() {
return Error("cannot parse value starting with: " + TokenToStringId(token_));
}
CheckedError Parser::ParseFunction(const std::string *name, Value &e) {
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
// Copy name, attribute will be changed on NEXT().
const auto functionname = attribute_;
if (!IsFloat(e.type.base_type)) {
return Error(functionname + ": type of argument mismatch, expecting: " +
kTypeNames[BASE_TYPE_DOUBLE] +
", found: " + kTypeNames[e.type.base_type] +
", name: " + (name ? *name : "") + ", value: " + e.constant);
}
NEXT();
EXPECT('(');
ECHECK(ParseSingleValue(name, e, false));
EXPECT(')');
// calculate with double precision
double x, y = 0.0;
ECHECK(atot(e.constant.c_str(), *this, &x));
// clang-format off
auto func_match = false;
#define FLATBUFFERS_FN_DOUBLE(name, op) \
if (!func_match && functionname == name) { y = op; func_match = true; }
FLATBUFFERS_FN_DOUBLE("deg", x / kPi * 180);
FLATBUFFERS_FN_DOUBLE("rad", x * kPi / 180);
FLATBUFFERS_FN_DOUBLE("sin", sin(x));
FLATBUFFERS_FN_DOUBLE("cos", cos(x));
FLATBUFFERS_FN_DOUBLE("tan", tan(x));
FLATBUFFERS_FN_DOUBLE("asin", asin(x));
FLATBUFFERS_FN_DOUBLE("acos", acos(x));
FLATBUFFERS_FN_DOUBLE("atan", atan(x));
// TODO(wvo): add more useful conversion functions here.
#undef FLATBUFFERS_FN_DOUBLE
// clang-format on
if (true != func_match) {
return Error(std::string("Unknown conversion function: ") + functionname +
", field name: " + (name ? *name : "") +
", value: " + e.constant);
}
e.constant = NumToString(y);
return NoError();
}
CheckedError Parser::TryTypedValue(const std::string *name, int dtoken,
bool check, Value &e, BaseType req,
bool *destmatch) {
FLATBUFFERS_ASSERT(*destmatch == false && dtoken == token_);
*destmatch = true;
e.constant = attribute_;
// Check token match
if (!check) {
if (e.type.base_type == BASE_TYPE_NONE) {
e.type.base_type = req;
} else {
return Error(std::string("type mismatch: expecting: ") +
kTypeNames[e.type.base_type] +
", found: " + kTypeNames[req] +
", name: " + (name ? *name : "") + ", value: " + e.constant);
}
}
// The exponent suffix of hexadecimal float-point number is mandatory.
// A hex-integer constant is forbidden as an initializer of float number.
if ((kTokenFloatConstant != dtoken) && IsFloat(e.type.base_type)) {
const auto &s = e.constant;
const auto k = s.find_first_of("0123456789.");
if ((std::string::npos != k) && (s.length() > (k + 1)) &&
(s[k] == '0' && is_alpha_char(s[k + 1], 'X')) &&
(std::string::npos == s.find_first_of("pP", k + 2))) {
return Error(
"invalid number, the exponent suffix of hexadecimal "
"floating-point literals is mandatory: \"" +
s + "\"");
}
}
NEXT();
return NoError();
}
CheckedError Parser::ParseSingleValue(const std::string *name, Value &e,
bool check_now) {
if (token_ == '+' || token_ == '-') {
const char sign = static_cast<char>(token_);
// Get an indentifier: NAN, INF, or function name like cos/sin/deg.
NEXT();
if (token_ != kTokenIdentifier) return Error("constant name expected");
attribute_.insert(0, 1, sign);
}
const auto in_type = e.type.base_type;
const auto is_tok_ident = (token_ == kTokenIdentifier);
const auto is_tok_string = (token_ == kTokenStringConstant);
// First see if this could be a conversion function.
if (is_tok_ident && *cursor_ == '(') { return ParseFunction(name, e); }
// clang-format off
auto match = false;
#define IF_ECHECK_(force, dtoken, check, req) \
if (!match && ((dtoken) == token_) && ((check) || IsConstTrue(force))) \
ECHECK(TryTypedValue(name, dtoken, check, e, req, &match))
#define TRY_ECHECK(dtoken, check, req) IF_ECHECK_(false, dtoken, check, req)
#define FORCE_ECHECK(dtoken, check, req) IF_ECHECK_(true, dtoken, check, req)
// clang-format on
if (is_tok_ident || is_tok_string) {
const auto kTokenStringOrIdent = token_;
// The string type is a most probable type, check it first.
TRY_ECHECK(kTokenStringConstant, in_type == BASE_TYPE_STRING,
BASE_TYPE_STRING);
// avoid escaped and non-ascii in the string
if (!match && is_tok_string && IsScalar(in_type) &&
!attr_is_trivial_ascii_string_) {
return Error(
std::string("type mismatch or invalid value, an initializer of "
"non-string field must be trivial ASCII string: type: ") +
kTypeNames[in_type] + ", name: " + (name ? *name : "") +
", value: " + attribute_);
}
// A boolean as true/false. Boolean as Integer check below.
if (!match && IsBool(in_type)) {
auto is_true = attribute_ == "true";
if (is_true || attribute_ == "false") {
attribute_ = is_true ? "1" : "0";
// accepts both kTokenStringConstant and kTokenIdentifier
TRY_ECHECK(kTokenStringOrIdent, IsBool(in_type), BASE_TYPE_BOOL);
}
}
// Check for optional scalars.
if (!match && IsScalar(in_type) && attribute_ == "null") {
e.constant = "null";
NEXT();
match = true;
}
// Check if this could be a string/identifier enum value.
// Enum can have only true integer base type.
if (!match && IsInteger(in_type) && !IsBool(in_type) &&
IsIdentifierStart(*attribute_.c_str())) {
ECHECK(ParseEnumFromString(e.type, &e.constant));
NEXT();
match = true;
}
// Parse a float/integer number from the string.
// A "scalar-in-string" value needs extra checks.
if (!match && is_tok_string && IsScalar(in_type)) {
// Strip trailing whitespaces from attribute_.
auto last_non_ws = attribute_.find_last_not_of(' ');
if (std::string::npos != last_non_ws) attribute_.resize(last_non_ws + 1);
if (IsFloat(e.type.base_type)) {
// The functions strtod() and strtof() accept both 'nan' and
// 'nan(number)' literals. While 'nan(number)' is rejected by the parser
// as an unsupported function if is_tok_ident is true.
if (attribute_.find_last_of(')') != std::string::npos) {
return Error("invalid number: " + attribute_);
}
}
}
// Float numbers or nan, inf, pi, etc.
TRY_ECHECK(kTokenStringOrIdent, IsFloat(in_type), BASE_TYPE_FLOAT);
// An integer constant in string.
TRY_ECHECK(kTokenStringOrIdent, IsInteger(in_type), BASE_TYPE_INT);
// Unknown tokens will be interpreted as string type.
// An attribute value may be a scalar or string constant.
FORCE_ECHECK(kTokenStringConstant, in_type == BASE_TYPE_STRING,
BASE_TYPE_STRING);
} else {
// Try a float number.
TRY_ECHECK(kTokenFloatConstant, IsFloat(in_type), BASE_TYPE_FLOAT);
// Integer token can init any scalar (integer of float).
FORCE_ECHECK(kTokenIntegerConstant, IsScalar(in_type), BASE_TYPE_INT);
}
// Match empty vectors for default-empty-vectors.
if (!match && IsVector(e.type) && token_ == '[') {
NEXT();
if (token_ != ']') { return Error("Expected `]` in vector default"); }
NEXT();
match = true;
e.constant = "[]";
}
#undef FORCE_ECHECK
#undef TRY_ECHECK
#undef IF_ECHECK_
if (!match) {
std::string msg;
msg += "Cannot assign token starting with '" + TokenToStringId(token_) +
"' to value of <" + std::string(kTypeNames[in_type]) + "> type.";
return Error(msg);
}
const auto match_type = e.type.base_type; // may differ from in_type
// The check_now flag must be true when parse a fbs-schema.
// This flag forces to check default scalar values or metadata of field.
// For JSON parser the flag should be false.
// If it is set for JSON each value will be checked twice (see ParseTable).
// Special case 'null' since atot can't handle that.
if (check_now && IsScalar(match_type) && e.constant != "null") {
// clang-format off
switch (match_type) {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_ ## ENUM: {\
CTYPE val; \
ECHECK(atot(e.constant.c_str(), *this, &val)); \
SingleValueRepack(e, val); \
break; }
FLATBUFFERS_GEN_TYPES_SCALAR(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
default: break;
}
// clang-format on
}
return NoError();
}
StructDef *Parser::LookupCreateStruct(const std::string &name,
bool create_if_new, bool definition) {
std::string qualified_name = current_namespace_->GetFullyQualifiedName(name);
// See if it exists pre-declared by an unqualified use.
auto struct_def = LookupStruct(name);
if (struct_def && struct_def->predecl) {
if (definition) {
// Make sure it has the current namespace, and is registered under its
// qualified name.
struct_def->defined_namespace = current_namespace_;
structs_.Move(name, qualified_name);
}
return struct_def;
}
// See if it exists pre-declared by an qualified use.
struct_def = LookupStruct(qualified_name);
if (struct_def && struct_def->predecl) {
if (definition) {
// Make sure it has the current namespace.
struct_def->defined_namespace = current_namespace_;
}
return struct_def;
}
if (!definition && !struct_def) {
struct_def = LookupStructThruParentNamespaces(name);
}
if (!struct_def && create_if_new) {
struct_def = new StructDef();
if (definition) {
structs_.Add(qualified_name, struct_def);
struct_def->name = name;
struct_def->defined_namespace = current_namespace_;
} else {
// Not a definition.
// Rather than failing, we create a "pre declared" StructDef, due to
// circular references, and check for errors at the end of parsing.
// It is defined in the current namespace, as the best guess what the
// final namespace will be.
structs_.Add(name, struct_def);
struct_def->name = name;
struct_def->defined_namespace = current_namespace_;
struct_def->original_location.reset(
new std::string(file_being_parsed_ + ":" + NumToString(line_)));
}
}
return struct_def;
}
const EnumVal *EnumDef::MinValue() const {
return vals.vec.empty() ? nullptr : vals.vec.front();
}
const EnumVal *EnumDef::MaxValue() const {
return vals.vec.empty() ? nullptr : vals.vec.back();
}
uint64_t EnumDef::Distance(const EnumVal *v1, const EnumVal *v2) const {
return IsUInt64() ? EnumDistanceImpl(v1->GetAsUInt64(), v2->GetAsUInt64())
: EnumDistanceImpl(v1->GetAsInt64(), v2->GetAsInt64());
}
std::string EnumDef::AllFlags() const {
FLATBUFFERS_ASSERT(attributes.Lookup("bit_flags"));
uint64_t u64 = 0;
for (auto it = Vals().begin(); it != Vals().end(); ++it) {
u64 |= (*it)->GetAsUInt64();
}
return IsUInt64() ? NumToString(u64) : NumToString(static_cast<int64_t>(u64));
}
EnumVal *EnumDef::ReverseLookup(int64_t enum_idx,
bool skip_union_default) const {
auto skip_first = static_cast<int>(is_union && skip_union_default);
for (auto it = Vals().begin() + skip_first; it != Vals().end(); ++it) {
if ((*it)->GetAsInt64() == enum_idx) { return *it; }
}
return nullptr;
}
EnumVal *EnumDef::FindByValue(const std::string &constant) const {
int64_t i64;
auto done = false;
if (IsUInt64()) {
uint64_t u64; // avoid reinterpret_cast of pointers
done = StringToNumber(constant.c_str(), &u64);
i64 = static_cast<int64_t>(u64);
} else {
done = StringToNumber(constant.c_str(), &i64);
}
FLATBUFFERS_ASSERT(done);
if (!done) return nullptr;
return ReverseLookup(i64, false);
}
void EnumDef::SortByValue() {
auto &v = vals.vec;
if (IsUInt64())
std::sort(v.begin(), v.end(), [](const EnumVal *e1, const EnumVal *e2) {
if (e1->GetAsUInt64() == e2->GetAsUInt64()) {
return e1->name < e2->name;
}
return e1->GetAsUInt64() < e2->GetAsUInt64();
});
else
std::sort(v.begin(), v.end(), [](const EnumVal *e1, const EnumVal *e2) {
if (e1->GetAsInt64() == e2->GetAsInt64()) { return e1->name < e2->name; }
return e1->GetAsInt64() < e2->GetAsInt64();
});
}
void EnumDef::RemoveDuplicates() {
// This method depends form SymbolTable implementation!
// 1) vals.vec - owner (raw pointer)
// 2) vals.dict - access map
auto first = vals.vec.begin();
auto last = vals.vec.end();
if (first == last) return;
auto result = first;
while (++first != last) {
if ((*result)->value != (*first)->value) {
*(++result) = *first;
} else {
auto ev = *first;
for (auto it = vals.dict.begin(); it != vals.dict.end(); ++it) {
if (it->second == ev) it->second = *result; // reassign
}
delete ev; // delete enum value
*first = nullptr;
}
}
vals.vec.erase(++result, last);
}
template<typename T> void EnumDef::ChangeEnumValue(EnumVal *ev, T new_value) {
ev->value = static_cast<int64_t>(new_value);
}
namespace EnumHelper {
template<BaseType E> struct EnumValType { typedef int64_t type; };
template<> struct EnumValType<BASE_TYPE_ULONG> { typedef uint64_t type; };
} // namespace EnumHelper
struct EnumValBuilder {
EnumVal *CreateEnumerator(const std::string &ev_name) {
FLATBUFFERS_ASSERT(!temp);
auto first = enum_def.vals.vec.empty();
user_value = first;
temp = new EnumVal(ev_name, first ? 0 : enum_def.vals.vec.back()->value);
return temp;
}
EnumVal *CreateEnumerator(const std::string &ev_name, int64_t val) {
FLATBUFFERS_ASSERT(!temp);
user_value = true;
temp = new EnumVal(ev_name, val);
return temp;
}
FLATBUFFERS_CHECKED_ERROR AcceptEnumerator(const std::string &name) {
FLATBUFFERS_ASSERT(temp);
ECHECK(ValidateValue(&temp->value, false == user_value));
FLATBUFFERS_ASSERT((temp->union_type.enum_def == nullptr) ||
(temp->union_type.enum_def == &enum_def));
auto not_unique = enum_def.vals.Add(name, temp);
temp = nullptr;
if (not_unique) return parser.Error("enum value already exists: " + name);
return NoError();
}
FLATBUFFERS_CHECKED_ERROR AcceptEnumerator() {
return AcceptEnumerator(temp->name);
}
FLATBUFFERS_CHECKED_ERROR AssignEnumeratorValue(const std::string &value) {
user_value = true;
auto fit = false;
if (enum_def.IsUInt64()) {
uint64_t u64;
fit = StringToNumber(value.c_str(), &u64);
temp->value = static_cast<int64_t>(u64); // well-defined since C++20.
} else {
int64_t i64;
fit = StringToNumber(value.c_str(), &i64);
temp->value = i64;
}
if (!fit) return parser.Error("enum value does not fit, \"" + value + "\"");
return NoError();
}
template<BaseType E, typename CTYPE>
inline FLATBUFFERS_CHECKED_ERROR ValidateImpl(int64_t *ev, int m) {
typedef typename EnumHelper::EnumValType<E>::type T; // int64_t or uint64_t
static_assert(sizeof(T) == sizeof(int64_t), "invalid EnumValType");
const auto v = static_cast<T>(*ev);
auto up = static_cast<T>((flatbuffers::numeric_limits<CTYPE>::max)());
auto dn = static_cast<T>((flatbuffers::numeric_limits<CTYPE>::lowest)());
if (v < dn || v > (up - m)) {
return parser.Error("enum value does not fit, \"" + NumToString(v) +
(m ? " + 1\"" : "\"") + " out of " +
TypeToIntervalString<CTYPE>());
}
*ev = static_cast<int64_t>(v + m); // well-defined since C++20.
return NoError();
}
FLATBUFFERS_CHECKED_ERROR ValidateValue(int64_t *ev, bool next) {
// clang-format off
switch (enum_def.underlying_type.base_type) {
#define FLATBUFFERS_TD(ENUM, IDLTYPE, CTYPE, ...) \
case BASE_TYPE_##ENUM: { \
if (!IsInteger(BASE_TYPE_##ENUM)) break; \
return ValidateImpl<BASE_TYPE_##ENUM, CTYPE>(ev, next ? 1 : 0); \
}
FLATBUFFERS_GEN_TYPES_SCALAR(FLATBUFFERS_TD)
#undef FLATBUFFERS_TD
default: break;
}
// clang-format on
return parser.Error("fatal: invalid enum underlying type");
}
EnumValBuilder(Parser &_parser, EnumDef &_enum_def)
: parser(_parser),
enum_def(_enum_def),
temp(nullptr),
user_value(false) {}
~EnumValBuilder() { delete temp; }
Parser &parser;
EnumDef &enum_def;
EnumVal *temp;
bool user_value;
};
CheckedError Parser::ParseEnum(const bool is_union, EnumDef **dest,
const char *filename) {
std::vector<std::string> enum_comment = doc_comment_;
NEXT();
std::string enum_name = attribute_;
EXPECT(kTokenIdentifier);
EnumDef *enum_def;
ECHECK(StartEnum(enum_name, is_union, &enum_def));
if (filename != nullptr && !opts.project_root.empty()) {
enum_def->declaration_file =
&GetPooledString(RelativeToRootPath(opts.project_root, filename));
}
enum_def->doc_comment = enum_comment;
if (!is_union && !opts.proto_mode) {
// Give specialized error message, since this type spec used to
// be optional in the first FlatBuffers release.
if (!Is(':')) {
return Error(
"must specify the underlying integer type for this"
" enum (e.g. \': short\', which was the default).");
} else {
NEXT();
}
// Specify the integer type underlying this enum.
ECHECK(ParseType(enum_def->underlying_type));
if (!IsInteger(enum_def->underlying_type.base_type) ||
IsBool(enum_def->underlying_type.base_type))
return Error("underlying enum type must be integral");
// Make this type refer back to the enum it was derived from.
enum_def->underlying_type.enum_def = enum_def;
}
ECHECK(ParseMetaData(&enum_def->attributes));
const auto underlying_type = enum_def->underlying_type.base_type;
if (enum_def->attributes.Lookup("bit_flags") &&
!IsUnsigned(underlying_type)) {
// todo: Convert to the Error in the future?
Warning("underlying type of bit_flags enum must be unsigned");
}
if (enum_def->attributes.Lookup("force_align")) {
return Error("`force_align` is not a valid attribute for Enums. ");
}
EnumValBuilder evb(*this, *enum_def);
EXPECT('{');
// A lot of code generatos expect that an enum is not-empty.
if ((is_union || Is('}')) && !opts.proto_mode) {
evb.CreateEnumerator("NONE");
ECHECK(evb.AcceptEnumerator());
}
std::set<std::pair<BaseType, StructDef *>> union_types;
while (!Is('}')) {
if (opts.proto_mode && attribute_ == "option") {
ECHECK(ParseProtoOption());
} else {
auto &ev = *evb.CreateEnumerator(attribute_);
auto full_name = ev.name;
ev.doc_comment = doc_comment_;
EXPECT(kTokenIdentifier);
if (is_union) {
ECHECK(ParseNamespacing(&full_name, &ev.name));
if (opts.union_value_namespacing) {
// Since we can't namespace the actual enum identifiers, turn
// namespace parts into part of the identifier.
ev.name = full_name;
std::replace(ev.name.begin(), ev.name.end(), '.', '_');
}
if (Is(':')) {
NEXT();
ECHECK(ParseType(ev.union_type));
if (ev.union_type.base_type != BASE_TYPE_STRUCT &&
ev.union_type.base_type != BASE_TYPE_STRING)
return Error("union value type may only be table/struct/string");
} else {
ev.union_type = Type(BASE_TYPE_STRUCT, LookupCreateStruct(full_name));
}
if (!enum_def->uses_multiple_type_instances) {
auto ins = union_types.insert(std::make_pair(
ev.union_type.base_type, ev.union_type.struct_def));
enum_def->uses_multiple_type_instances = (false == ins.second);
}
}
if (Is('=')) {
NEXT();
ECHECK(evb.AssignEnumeratorValue(attribute_));
EXPECT(kTokenIntegerConstant);
}
if (opts.proto_mode && Is('[')) {
NEXT();
// ignore attributes on enums.
while (token_ != ']') NEXT();
NEXT();
} else {
// parse attributes in fbs schema
ECHECK(ParseMetaData(&ev.attributes));
}
ECHECK(evb.AcceptEnumerator());
}
if (!Is(opts.proto_mode ? ';' : ',')) break;
NEXT();
}
EXPECT('}');
// At this point, the enum can be empty if input is invalid proto-file.
if (!enum_def->size())
return Error("incomplete enum declaration, values not found");
if (enum_def->attributes.Lookup("bit_flags")) {
const auto base_width = static_cast<uint64_t>(8 * SizeOf(underlying_type));
for (auto it = enum_def->Vals().begin(); it != enum_def->Vals().end();
++it) {
auto ev = *it;
const auto u = ev->GetAsUInt64();
// Stop manipulations with the sign.
if (!IsUnsigned(underlying_type) && u == (base_width - 1))
return Error("underlying type of bit_flags enum must be unsigned");
if (u >= base_width)
return Error("bit flag out of range of underlying integral type");
enum_def->ChangeEnumValue(ev, 1ULL << u);
}
}
enum_def->SortByValue(); // Must be sorted to use MinValue/MaxValue.
// Ensure enum value uniqueness.
auto prev_it = enum_def->Vals().begin();
for (auto it = prev_it + 1; it != enum_def->Vals().end(); ++it) {
auto prev_ev = *prev_it;
auto ev = *it;
if (prev_ev->GetAsUInt64() == ev->GetAsUInt64())
return Error("all enum values must be unique: " + prev_ev->name +
" and " + ev->name + " are both " +
NumToString(ev->GetAsInt64()));
}
if (dest) *dest = enum_def;
const auto qualified_name =
current_namespace_->GetFullyQualifiedName(enum_def->name);
if (types_.Add(qualified_name, new Type(BASE_TYPE_UNION, nullptr, enum_def)))
return Error("datatype already exists: " + qualified_name);
return NoError();
}
CheckedError Parser::StartStruct(const std::string &name, StructDef **dest) {
auto &struct_def = *LookupCreateStruct(name, true, true);
if (!struct_def.predecl)
return Error("datatype already exists: " +
current_namespace_->GetFullyQualifiedName(name));
struct_def.predecl = false;
struct_def.name = name;
struct_def.file = file_being_parsed_;
// Move this struct to the back of the vector just in case it was predeclared,
// to preserve declaration order.
*std::remove(structs_.vec.begin(), structs_.vec.end(), &struct_def) =
&struct_def;
*dest = &struct_def;
return NoError();
}
CheckedError Parser::CheckClash(std::vector<FieldDef *> &fields,
StructDef *struct_def, const char *suffix,
BaseType basetype) {
auto len = strlen(suffix);
for (auto it = fields.begin(); it != fields.end(); ++it) {
auto &fname = (*it)->name;
if (fname.length() > len &&
fname.compare(fname.length() - len, len, suffix) == 0 &&
(*it)->value.type.base_type != BASE_TYPE_UTYPE) {
auto field =
struct_def->fields.Lookup(fname.substr(0, fname.length() - len));
if (field && field->value.type.base_type == basetype)
return Error("Field " + fname +
" would clash with generated functions for field " +
field->name);
}
}
return NoError();
}
std::vector<IncludedFile> Parser::GetIncludedFiles() const {
const auto it = files_included_per_file_.find(file_being_parsed_);
if (it == files_included_per_file_.end()) { return {}; }
return { it->second.cbegin(), it->second.cend() };
}
bool Parser::SupportsOptionalScalars(const flatbuffers::IDLOptions &opts) {
static FLATBUFFERS_CONSTEXPR unsigned long supported_langs =
IDLOptions::kRust | IDLOptions::kSwift | IDLOptions::kLobster |
IDLOptions::kKotlin | IDLOptions::kCpp | IDLOptions::kJava |
IDLOptions::kCSharp | IDLOptions::kTs | IDLOptions::kBinary |
IDLOptions::kGo | IDLOptions::kPython | IDLOptions::kJson |
IDLOptions::kNim;
unsigned long langs = opts.lang_to_generate;
return (langs > 0 && langs < IDLOptions::kMAX) && !(langs & ~supported_langs);
}
bool Parser::SupportsOptionalScalars() const {
// Check in general if a language isn't specified.
return opts.lang_to_generate == 0 || SupportsOptionalScalars(opts);
}
bool Parser::SupportsDefaultVectorsAndStrings() const {
static FLATBUFFERS_CONSTEXPR unsigned long supported_langs =
IDLOptions::kRust | IDLOptions::kSwift | IDLOptions::kNim;
return !(opts.lang_to_generate & ~supported_langs);
}
bool Parser::SupportsAdvancedUnionFeatures() const {
return (opts.lang_to_generate &
~(IDLOptions::kCpp | IDLOptions::kTs | IDLOptions::kPhp |
IDLOptions::kJava | IDLOptions::kCSharp | IDLOptions::kKotlin |
IDLOptions::kBinary | IDLOptions::kSwift | IDLOptions::kNim)) == 0;
}
bool Parser::SupportsAdvancedArrayFeatures() const {
return (opts.lang_to_generate &
~(IDLOptions::kCpp | IDLOptions::kPython | IDLOptions::kJava |
IDLOptions::kCSharp | IDLOptions::kJsonSchema | IDLOptions::kJson |
IDLOptions::kBinary | IDLOptions::kRust)) == 0;
}
Namespace *Parser::UniqueNamespace(Namespace *ns) {
for (auto it = namespaces_.begin(); it != namespaces_.end(); ++it) {
if (ns->components == (*it)->components) {
delete ns;
return *it;
}
}
namespaces_.push_back(ns);
return ns;
}
std::string Parser::UnqualifiedName(const std::string &full_qualified_name) {
Namespace *ns = new Namespace();
std::size_t current, previous = 0;
current = full_qualified_name.find('.');
while (current != std::string::npos) {
ns->components.push_back(
full_qualified_name.substr(previous, current - previous));
previous = current + 1;
current = full_qualified_name.find('.', previous);
}
current_namespace_ = UniqueNamespace(ns);
return full_qualified_name.substr(previous, current - previous);
}
CheckedError Parser::ParseDecl(const char *filename) {
std::vector<std::string> dc = doc_comment_;
bool fixed = IsIdent("struct");
if (!fixed && !IsIdent("table")) return Error("declaration expected");
NEXT();
std::string name = attribute_;
EXPECT(kTokenIdentifier);
StructDef *struct_def;
ECHECK(StartStruct(name, &struct_def));
struct_def->doc_comment = dc;
struct_def->fixed = fixed;
if (filename && !opts.project_root.empty()) {
struct_def->declaration_file =
&GetPooledString(RelativeToRootPath(opts.project_root, filename));
}
ECHECK(ParseMetaData(&struct_def->attributes));
struct_def->sortbysize =
struct_def->attributes.Lookup("original_order") == nullptr && !fixed;
EXPECT('{');
while (token_ != '}') ECHECK(ParseField(*struct_def));
if (fixed) {
const auto force_align = struct_def->attributes.Lookup("force_align");
if (force_align) {
size_t align;
ECHECK(ParseAlignAttribute(force_align->constant, struct_def->minalign,
&align));
struct_def->minalign = align;
}
if (!struct_def->bytesize) return Error("size 0 structs not allowed");
}
struct_def->PadLastField(struct_def->minalign);
// Check if this is a table that has manual id assignments
auto &fields = struct_def->fields.vec;
if (!fixed && fields.size()) {
size_t num_id_fields = 0;
for (auto it = fields.begin(); it != fields.end(); ++it) {
if ((*it)->attributes.Lookup("id")) num_id_fields++;
}
// If any fields have ids..
if (num_id_fields || opts.require_explicit_ids) {
// Then all fields must have them.
if (num_id_fields != fields.size()) {
if (opts.require_explicit_ids) {
return Error(
"all fields must have an 'id' attribute when "
"--require-explicit-ids is used");
} else {
return Error(
"either all fields or no fields must have an 'id' attribute");
}
}
// Simply sort by id, then the fields are the same as if no ids had
// been specified.
std::sort(fields.begin(), fields.end(), compareFieldDefs);
// Verify we have a contiguous set, and reassign vtable offsets.
FLATBUFFERS_ASSERT(fields.size() <=
flatbuffers::numeric_limits<voffset_t>::max());
for (voffset_t i = 0; i < static_cast<voffset_t>(fields.size()); i++) {
auto &field = *fields[i];
const auto &id_str = field.attributes.Lookup("id")->constant;
// Metadata values have a dynamic type, they can be `float`, 'int', or
// 'string`.
// The FieldIndexToOffset(i) expects the voffset_t so `id` is limited by
// this type.
voffset_t id = 0;
const auto done = !atot(id_str.c_str(), *this, &id).Check();
if (!done)
return Error("field id\'s must be non-negative number, field: " +
field.name + ", id: " + id_str);
if (i != id)
return Error("field id\'s must be consecutive from 0, id " +
NumToString(i) + " missing or set twice, field: " +
field.name + ", id: " + id_str);
field.value.offset = FieldIndexToOffset(i);
}
}
}
ECHECK(
CheckClash(fields, struct_def, UnionTypeFieldSuffix(), BASE_TYPE_UNION));
ECHECK(CheckClash(fields, struct_def, "Type", BASE_TYPE_UNION));
ECHECK(CheckClash(fields, struct_def, "_length", BASE_TYPE_VECTOR));
ECHECK(CheckClash(fields, struct_def, "Length", BASE_TYPE_VECTOR));
ECHECK(CheckClash(fields, struct_def, "_byte_vector", BASE_TYPE_STRING));
ECHECK(CheckClash(fields, struct_def, "ByteVector", BASE_TYPE_STRING));
EXPECT('}');
const auto qualified_name =
current_namespace_->GetFullyQualifiedName(struct_def->name);
if (types_.Add(qualified_name,
new Type(BASE_TYPE_STRUCT, struct_def, nullptr)))
return Error("datatype already exists: " + qualified_name);
return NoError();
}
CheckedError Parser::ParseService(const char *filename) {
std::vector<std::string> service_comment = doc_comment_;
NEXT();
auto service_name = attribute_;
EXPECT(kTokenIdentifier);
auto &service_def = *new ServiceDef();
service_def.name = service_name;
service_def.file = file_being_parsed_;
service_def.doc_comment = service_comment;
service_def.defined_namespace = current_namespace_;
if (filename != nullptr && !opts.project_root.empty()) {
service_def.declaration_file =
&GetPooledString(RelativeToRootPath(opts.project_root, filename));
}
if (services_.Add(current_namespace_->GetFullyQualifiedName(service_name),
&service_def))
return Error("service already exists: " + service_name);
ECHECK(ParseMetaData(&service_def.attributes));
EXPECT('{');
do {
std::vector<std::string> doc_comment = doc_comment_;
auto rpc_name = attribute_;
EXPECT(kTokenIdentifier);
EXPECT('(');
Type reqtype, resptype;
ECHECK(ParseTypeIdent(reqtype));
EXPECT(')');
EXPECT(':');
ECHECK(ParseTypeIdent(resptype));
if (reqtype.base_type != BASE_TYPE_STRUCT || reqtype.struct_def->fixed ||
resptype.base_type != BASE_TYPE_STRUCT || resptype.struct_def->fixed)
return Error("rpc request and response types must be tables");
auto &rpc = *new RPCCall();
rpc.name = rpc_name;
rpc.request = reqtype.struct_def;
rpc.response = resptype.struct_def;
rpc.doc_comment = doc_comment;
if (service_def.calls.Add(rpc_name, &rpc))
return Error("rpc already exists: " + rpc_name);
ECHECK(ParseMetaData(&rpc.attributes));
EXPECT(';');
} while (token_ != '}');
NEXT();
return NoError();
}
bool Parser::SetRootType(const char *name) {
root_struct_def_ = LookupStruct(name);
if (!root_struct_def_)
root_struct_def_ =
LookupStruct(current_namespace_->GetFullyQualifiedName(name));
return root_struct_def_ != nullptr;
}
void Parser::MarkGenerated() {
// This function marks all existing definitions as having already
// been generated, which signals no code for included files should be
// generated.
for (auto it = enums_.vec.begin(); it != enums_.vec.end(); ++it) {
(*it)->generated = true;
}
for (auto it = structs_.vec.begin(); it != structs_.vec.end(); ++it) {
if (!(*it)->predecl) { (*it)->generated = true; }
}
for (auto it = services_.vec.begin(); it != services_.vec.end(); ++it) {
(*it)->generated = true;
}
}
CheckedError Parser::ParseNamespace() {
NEXT();
auto ns = new Namespace();
namespaces_.push_back(ns); // Store it here to not leak upon error.
if (token_ != ';') {
for (;;) {
ns->components.push_back(attribute_);
EXPECT(kTokenIdentifier);
if (Is('.')) NEXT() else break;
}
}
namespaces_.pop_back();
current_namespace_ = UniqueNamespace(ns);
EXPECT(';');
return NoError();
}
// Best effort parsing of .proto declarations, with the aim to turn them
// in the closest corresponding FlatBuffer equivalent.
// We parse everything as identifiers instead of keywords, since we don't
// want protobuf keywords to become invalid identifiers in FlatBuffers.
CheckedError Parser::ParseProtoDecl() {
bool isextend = IsIdent("extend");
if (IsIdent("package")) {
// These are identical in syntax to FlatBuffer's namespace decl.
ECHECK(ParseNamespace());
} else if (IsIdent("message") || isextend) {
std::vector<std::string> struct_comment = doc_comment_;
NEXT();
StructDef *struct_def = nullptr;
Namespace *parent_namespace = nullptr;
if (isextend) {
if (Is('.')) NEXT(); // qualified names may start with a . ?
auto id = attribute_;
EXPECT(kTokenIdentifier);
ECHECK(ParseNamespacing(&id, nullptr));
struct_def = LookupCreateStruct(id, false);
if (!struct_def)
return Error("cannot extend unknown message type: " + id);
} else {
std::string name = attribute_;
EXPECT(kTokenIdentifier);
ECHECK(StartStruct(name, &struct_def));
// Since message definitions can be nested, we create a new namespace.
auto ns = new Namespace();
// Copy of current namespace.
*ns = *current_namespace_;
// But with current message name.
ns->components.push_back(name);
ns->from_table++;
parent_namespace = current_namespace_;
current_namespace_ = UniqueNamespace(ns);
}
struct_def->doc_comment = struct_comment;
ECHECK(ParseProtoFields(struct_def, isextend, false));
if (!isextend) { current_namespace_ = parent_namespace; }
if (Is(';')) NEXT();
} else if (IsIdent("enum")) {
// These are almost the same, just with different terminator:
EnumDef *enum_def;
ECHECK(ParseEnum(false, &enum_def, nullptr));
if (Is(';')) NEXT();
// Temp: remove any duplicates, as .fbs files can't handle them.
enum_def->RemoveDuplicates();
} else if (IsIdent("syntax")) { // Skip these.
NEXT();
EXPECT('=');
EXPECT(kTokenStringConstant);
EXPECT(';');
} else if (IsIdent("option")) { // Skip these.
ECHECK(ParseProtoOption());
EXPECT(';');
} else if (IsIdent("service")) { // Skip these.
NEXT();
EXPECT(kTokenIdentifier);
ECHECK(ParseProtoCurliesOrIdent());
} else {
return Error("don\'t know how to parse .proto declaration starting with " +
TokenToStringId(token_));
}
return NoError();
}
CheckedError Parser::StartEnum(const std::string &name, bool is_union,
EnumDef **dest) {
auto &enum_def = *new EnumDef();
enum_def.name = name;
enum_def.file = file_being_parsed_;
enum_def.doc_comment = doc_comment_;
enum_def.is_union = is_union;
enum_def.defined_namespace = current_namespace_;
const auto qualified_name = current_namespace_->GetFullyQualifiedName(name);
if (enums_.Add(qualified_name, &enum_def))
return Error("enum already exists: " + qualified_name);
enum_def.underlying_type.base_type =
is_union ? BASE_TYPE_UTYPE : BASE_TYPE_INT;
enum_def.underlying_type.enum_def = &enum_def;
if (dest) *dest = &enum_def;
return NoError();
}
CheckedError Parser::ParseProtoFields(StructDef *struct_def, bool isextend,
bool inside_oneof) {
EXPECT('{');
while (token_ != '}') {
if (IsIdent("message") || IsIdent("extend") || IsIdent("enum")) {
// Nested declarations.
ECHECK(ParseProtoDecl());
} else if (IsIdent("extensions")) { // Skip these.
NEXT();
EXPECT(kTokenIntegerConstant);
if (Is(kTokenIdentifier)) {
NEXT(); // to
NEXT(); // num
}
EXPECT(';');
} else if (IsIdent("option")) { // Skip these.
ECHECK(ParseProtoOption());
EXPECT(';');
} else if (IsIdent("reserved")) { // Skip these.
NEXT();
while (!Is(';')) { NEXT(); } // A variety of formats, just skip.
NEXT();
} else {
std::vector<std::string> field_comment = doc_comment_;
// Parse the qualifier.
bool required = false;
bool repeated = false;
bool oneof = false;
if (!inside_oneof) {
if (IsIdent("optional")) {
// This is the default.
NEXT();
} else if (IsIdent("required")) {
required = true;
NEXT();
} else if (IsIdent("repeated")) {
repeated = true;
NEXT();
} else if (IsIdent("oneof")) {
oneof = true;
NEXT();
} else {
// can't error, proto3 allows decls without any of the above.
}
}
StructDef *anonymous_struct = nullptr;
EnumDef *oneof_union = nullptr;
Type type;
if (IsIdent("group") || oneof) {
if (!oneof) NEXT();
if (oneof && opts.proto_oneof_union) {
auto name = ConvertCase(attribute_, Case::kUpperCamel) + "Union";
ECHECK(StartEnum(name, true, &oneof_union));
type = Type(BASE_TYPE_UNION, nullptr, oneof_union);
} else {
auto name = "Anonymous" + NumToString(anonymous_counter_++);
ECHECK(StartStruct(name, &anonymous_struct));
type = Type(BASE_TYPE_STRUCT, anonymous_struct);
}
} else {
ECHECK(ParseTypeFromProtoType(&type));
}
// Repeated elements get mapped to a vector.
if (repeated) {
type.element = type.base_type;
type.base_type = BASE_TYPE_VECTOR;
if (type.element == BASE_TYPE_VECTOR) {
// We have a vector or vectors, which FlatBuffers doesn't support.
// For now make it a vector of string (since the source is likely
// "repeated bytes").
// TODO(wvo): A better solution would be to wrap this in a table.
type.element = BASE_TYPE_STRING;
}
}
std::string name = attribute_;
EXPECT(kTokenIdentifier);
if (!oneof) {
// Parse the field id. Since we're just translating schemas, not
// any kind of binary compatibility, we can safely ignore these, and
// assign our own.
EXPECT('=');
EXPECT(kTokenIntegerConstant);
}
FieldDef *field = nullptr;
if (isextend) {
// We allow a field to be re-defined when extending.
// TODO: are there situations where that is problematic?
field = struct_def->fields.Lookup(name);
}
if (!field) ECHECK(AddField(*struct_def, name, type, &field));
field->doc_comment = field_comment;
if (!IsScalar(type.base_type) && required) {
field->presence = FieldDef::kRequired;
}
// See if there's a default specified.
if (Is('[')) {
NEXT();
for (;;) {
auto key = attribute_;
ECHECK(ParseProtoKey());
EXPECT('=');
auto val = attribute_;
ECHECK(ParseProtoCurliesOrIdent());
if (key == "default") {
// Temp: skip non-numeric and non-boolean defaults (enums).
auto numeric = strpbrk(val.c_str(), "0123456789-+.");
if (IsFloat(type.base_type) &&
(val == "inf" || val == "+inf" || val == "-inf")) {
// Prefer to be explicit with +inf.
field->value.constant = val == "inf" ? "+inf" : val;
} else if (IsScalar(type.base_type) && numeric == val.c_str()) {
field->value.constant = val;
} else if (val == "true") {
field->value.constant = val;
} // "false" is default, no need to handle explicitly.
} else if (key == "deprecated") {
field->deprecated = val == "true";
}
if (!Is(',')) break;
NEXT();
}
EXPECT(']');
}
if (anonymous_struct) {
ECHECK(ParseProtoFields(anonymous_struct, false, oneof));
if (Is(';')) NEXT();
} else if (oneof_union) {
// Parse into a temporary StructDef, then transfer fields into an
// EnumDef describing the oneof as a union.
StructDef oneof_struct;
ECHECK(ParseProtoFields(&oneof_struct, false, oneof));
if (Is(';')) NEXT();
for (auto field_it = oneof_struct.fields.vec.begin();
field_it != oneof_struct.fields.vec.end(); ++field_it) {
const auto &oneof_field = **field_it;
const auto &oneof_type = oneof_field.value.type;
if (oneof_type.base_type != BASE_TYPE_STRUCT ||
!oneof_type.struct_def || oneof_type.struct_def->fixed)
return Error("oneof '" + name +
"' cannot be mapped to a union because member '" +
oneof_field.name + "' is not a table type.");
EnumValBuilder evb(*this, *oneof_union);
auto ev = evb.CreateEnumerator(oneof_type.struct_def->name);
ev->union_type = oneof_type;
ev->doc_comment = oneof_field.doc_comment;
ECHECK(evb.AcceptEnumerator(oneof_field.name));
}
} else {
EXPECT(';');
}
}
}
NEXT();
return NoError();
}
CheckedError Parser::ParseProtoKey() {
if (token_ == '(') {
NEXT();
// Skip "(a.b)" style custom attributes.
while (token_ == '.' || token_ == kTokenIdentifier) NEXT();
EXPECT(')');
while (Is('.')) {
NEXT();
EXPECT(kTokenIdentifier);
}
} else {
EXPECT(kTokenIdentifier);
}
return NoError();
}
CheckedError Parser::ParseProtoCurliesOrIdent() {
if (Is('{')) {
NEXT();
for (int nesting = 1; nesting;) {
if (token_ == '{')
nesting++;
else if (token_ == '}')
nesting--;
NEXT();
}
} else {
NEXT(); // Any single token.
}
return NoError();
}
CheckedError Parser::ParseProtoOption() {
NEXT();
ECHECK(ParseProtoKey());
EXPECT('=');
ECHECK(ParseProtoCurliesOrIdent());
return NoError();
}
// Parse a protobuf type, and map it to the corresponding FlatBuffer one.
CheckedError Parser::ParseTypeFromProtoType(Type *type) {
struct type_lookup {
const char *proto_type;
BaseType fb_type, element;
};
static type_lookup lookup[] = {
{ "float", BASE_TYPE_FLOAT, BASE_TYPE_NONE },
{ "double", BASE_TYPE_DOUBLE, BASE_TYPE_NONE },
{ "int32", BASE_TYPE_INT, BASE_TYPE_NONE },
{ "int64", BASE_TYPE_LONG, BASE_TYPE_NONE },
{ "uint32", BASE_TYPE_UINT, BASE_TYPE_NONE },
{ "uint64", BASE_TYPE_ULONG, BASE_TYPE_NONE },
{ "sint32", BASE_TYPE_INT, BASE_TYPE_NONE },
{ "sint64", BASE_TYPE_LONG, BASE_TYPE_NONE },
{ "fixed32", BASE_TYPE_UINT, BASE_TYPE_NONE },
{ "fixed64", BASE_TYPE_ULONG, BASE_TYPE_NONE },
{ "sfixed32", BASE_TYPE_INT, BASE_TYPE_NONE },
{ "sfixed64", BASE_TYPE_LONG, BASE_TYPE_NONE },
{ "bool", BASE_TYPE_BOOL, BASE_TYPE_NONE },
{ "string", BASE_TYPE_STRING, BASE_TYPE_NONE },
{ "bytes", BASE_TYPE_VECTOR, BASE_TYPE_UCHAR },
{ nullptr, BASE_TYPE_NONE, BASE_TYPE_NONE }
};
for (auto tl = lookup; tl->proto_type; tl++) {
if (attribute_ == tl->proto_type) {
type->base_type = tl->fb_type;
type->element = tl->element;
NEXT();
return NoError();
}
}
if (Is('.')) NEXT(); // qualified names may start with a . ?
ECHECK(ParseTypeIdent(*type));
return NoError();
}
CheckedError Parser::SkipAnyJsonValue() {
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
switch (token_) {
case '{': {
size_t fieldn_outer = 0;
return ParseTableDelimiters(fieldn_outer, nullptr,
[&](const std::string &, size_t &fieldn,
const StructDef *) -> CheckedError {
ECHECK(SkipAnyJsonValue());
fieldn++;
return NoError();
});
}
case '[': {
uoffset_t count = 0;
return ParseVectorDelimiters(count, [&](uoffset_t &) -> CheckedError {
return SkipAnyJsonValue();
});
}
case kTokenStringConstant:
case kTokenIntegerConstant:
case kTokenFloatConstant: NEXT(); break;
default:
if (IsIdent("true") || IsIdent("false") || IsIdent("null") ||
IsIdent("inf")) {
NEXT();
} else
return TokenError();
}
return NoError();
}
CheckedError Parser::ParseFlexBufferNumericConstant(
flexbuffers::Builder *builder) {
double d;
if (!StringToNumber(attribute_.c_str(), &d))
return Error("unexpected floating-point constant: " + attribute_);
builder->Double(d);
return NoError();
}
CheckedError Parser::ParseFlexBufferValue(flexbuffers::Builder *builder) {
ParseDepthGuard depth_guard(this);
ECHECK(depth_guard.Check());
switch (token_) {
case '{': {
auto start = builder->StartMap();
size_t fieldn_outer = 0;
auto err =
ParseTableDelimiters(fieldn_outer, nullptr,
[&](const std::string &name, size_t &fieldn,
const StructDef *) -> CheckedError {
builder->Key(name);
ECHECK(ParseFlexBufferValue(builder));
fieldn++;
return NoError();
});
ECHECK(err);
builder->EndMap(start);
if (builder->HasDuplicateKeys())
return Error("FlexBuffers map has duplicate keys");
break;
}
case '[': {
auto start = builder->StartVector();
uoffset_t count = 0;
ECHECK(ParseVectorDelimiters(count, [&](uoffset_t &) -> CheckedError {
return ParseFlexBufferValue(builder);
}));
builder->EndVector(start, false, false);
break;
}
case kTokenStringConstant:
builder->String(attribute_);
EXPECT(kTokenStringConstant);
break;
case kTokenIntegerConstant:
builder->Int(StringToInt(attribute_.c_str()));
EXPECT(kTokenIntegerConstant);
break;
case kTokenFloatConstant: {
double d;
StringToNumber(attribute_.c_str(), &d);
builder->Double(d);
EXPECT(kTokenFloatConstant);
break;
}
case '-':
case '+': {
// `[-+]?(nan|inf|infinity)`, see ParseSingleValue().
const auto sign = static_cast<char>(token_);
NEXT();
if (token_ != kTokenIdentifier)
return Error("floating-point constant expected");
attribute_.insert(0, 1, sign);
ECHECK(ParseFlexBufferNumericConstant(builder));
NEXT();
break;
}
default:
if (IsIdent("true")) {
builder->Bool(true);
NEXT();
} else if (IsIdent("false")) {
builder->Bool(false);
NEXT();
} else if (IsIdent("null")) {
builder->Null();
NEXT();
} else if (IsIdent("inf") || IsIdent("infinity") || IsIdent("nan")) {
ECHECK(ParseFlexBufferNumericConstant(builder));
NEXT();
} else
return TokenError();
}
return NoError();
}
bool Parser::ParseFlexBuffer(const char *source, const char *source_filename,
flexbuffers::Builder *builder) {
const auto initial_depth = parse_depth_counter_;
(void)initial_depth;
auto ok = !StartParseFile(source, source_filename).Check() &&
!ParseFlexBufferValue(builder).Check();
if (ok) builder->Finish();
FLATBUFFERS_ASSERT(initial_depth == parse_depth_counter_);
return ok;
}
bool Parser::Parse(const char *source, const char **include_paths,
const char *source_filename) {
const auto initial_depth = parse_depth_counter_;
(void)initial_depth;
bool r;
if (opts.use_flexbuffers) {
r = ParseFlexBuffer(source, source_filename, &flex_builder_);
} else {
r = !ParseRoot(source, include_paths, source_filename).Check();
}
FLATBUFFERS_ASSERT(initial_depth == parse_depth_counter_);
return r;
}
bool Parser::ParseJson(const char *json, const char *json_filename) {
const auto initial_depth = parse_depth_counter_;
(void)initial_depth;
builder_.Clear();
const auto done =
!StartParseFile(json, json_filename).Check() && !DoParseJson().Check();
FLATBUFFERS_ASSERT(initial_depth == parse_depth_counter_);
return done;
}
CheckedError Parser::StartParseFile(const char *source,
const char *source_filename) {
file_being_parsed_ = source_filename ? source_filename : "";
source_ = source;
ResetState(source_);
error_.clear();
ECHECK(SkipByteOrderMark());
NEXT();
if (Is(kTokenEof)) return Error("input file is empty");
return NoError();
}
CheckedError Parser::ParseRoot(const char *source, const char **include_paths,
const char *source_filename) {
ECHECK(DoParse(source, include_paths, source_filename, nullptr));
// Check that all types were defined.
for (auto it = structs_.vec.begin(); it != structs_.vec.end();) {
auto &struct_def = **it;
if (struct_def.predecl) {
if (opts.proto_mode) {
// Protos allow enums to be used before declaration, so check if that
// is the case here.
EnumDef *enum_def = nullptr;
for (size_t components =
struct_def.defined_namespace->components.size() + 1;
components && !enum_def; components--) {
auto qualified_name =
struct_def.defined_namespace->GetFullyQualifiedName(
struct_def.name, components - 1);
enum_def = LookupEnum(qualified_name);
}
if (enum_def) {
// This is pretty slow, but a simple solution for now.
auto initial_count = struct_def.refcount;
for (auto struct_it = structs_.vec.begin();
struct_it != structs_.vec.end(); ++struct_it) {
auto &sd = **struct_it;
for (auto field_it = sd.fields.vec.begin();
field_it != sd.fields.vec.end(); ++field_it) {
auto &field = **field_it;
if (field.value.type.struct_def == &struct_def) {
field.value.type.struct_def = nullptr;
field.value.type.enum_def = enum_def;
auto &bt = IsVector(field.value.type)
? field.value.type.element
: field.value.type.base_type;
FLATBUFFERS_ASSERT(bt == BASE_TYPE_STRUCT);
bt = enum_def->underlying_type.base_type;
struct_def.refcount--;
enum_def->refcount++;
}
}
}
if (struct_def.refcount)
return Error("internal: " + NumToString(struct_def.refcount) + "/" +
NumToString(initial_count) +
" use(s) of pre-declaration enum not accounted for: " +
enum_def->name);
structs_.dict.erase(structs_.dict.find(struct_def.name));
it = structs_.vec.erase(it);
delete &struct_def;
continue; // Skip error.
}
}
auto err = "type referenced but not defined (check namespace): " +
struct_def.name;
if (struct_def.original_location)
err += ", originally at: " + *struct_def.original_location;
return Error(err);
}
++it;
}
// This check has to happen here and not earlier, because only now do we
// know for sure what the type of these are.
for (auto it = enums_.vec.begin(); it != enums_.vec.end(); ++it) {
auto &enum_def = **it;
if (enum_def.is_union) {
for (auto val_it = enum_def.Vals().begin();
val_it != enum_def.Vals().end(); ++val_it) {
auto &val = **val_it;
if (!(opts.lang_to_generate != 0 && SupportsAdvancedUnionFeatures()) &&
(IsStruct(val.union_type) || IsString(val.union_type)))
return Error(
"only tables can be union elements in the generated language: " +
val.name);
}
}
}
auto err = CheckPrivateLeak();
if (err.Check()) return err;
// Parse JSON object only if the scheme has been parsed.
if (token_ == '{') { ECHECK(DoParseJson()); }
return NoError();
}
CheckedError Parser::CheckPrivateLeak() {
if (!opts.no_leak_private_annotations) return NoError();
// Iterate over all structs/tables to validate we arent leaking
// any private (structs/tables/enums)
for (auto it = structs_.vec.begin(); it != structs_.vec.end(); it++) {
auto &struct_def = **it;
for (auto fld_it = struct_def.fields.vec.begin();
fld_it != struct_def.fields.vec.end(); ++fld_it) {
auto &field = **fld_it;
if (field.value.type.enum_def) {
auto err =
CheckPrivatelyLeakedFields(struct_def, *field.value.type.enum_def);
if (err.Check()) { return err; }
} else if (field.value.type.struct_def) {
auto err = CheckPrivatelyLeakedFields(struct_def,
*field.value.type.struct_def);
if (err.Check()) { return err; }
}
}
}
// Iterate over all enums to validate we arent leaking
// any private (structs/tables)
for (auto it = enums_.vec.begin(); it != enums_.vec.end(); ++it) {
auto &enum_def = **it;
if (enum_def.is_union) {
for (auto val_it = enum_def.Vals().begin();
val_it != enum_def.Vals().end(); ++val_it) {
auto &val = **val_it;
if (val.union_type.struct_def) {
auto err =
CheckPrivatelyLeakedFields(enum_def, *val.union_type.struct_def);
if (err.Check()) { return err; }
}
}
}
}
return NoError();
}
CheckedError Parser::CheckPrivatelyLeakedFields(const Definition &def,
const Definition &value_type) {
if (!opts.no_leak_private_annotations) return NoError();
const auto is_private = def.attributes.Lookup("private");
const auto is_field_private = value_type.attributes.Lookup("private");
if (!is_private && is_field_private) {
return Error(
"Leaking private implementation, verify all objects have similar "
"annotations");
}
return NoError();
}
CheckedError Parser::DoParse(const char *source, const char **include_paths,
const char *source_filename,
const char *include_filename) {
uint64_t source_hash = 0;
if (source_filename) {
// If the file is in-memory, don't include its contents in the hash as we
// won't be able to load them later.
if (FileExists(source_filename))
source_hash = HashFile(source_filename, source);
else
source_hash = HashFile(source_filename, nullptr);
if (included_files_.find(source_hash) == included_files_.end()) {
included_files_[source_hash] = include_filename ? include_filename : "";
files_included_per_file_[source_filename] = std::set<IncludedFile>();
} else {
return NoError();
}
}
if (!include_paths) {
static const char *current_directory[] = { "", nullptr };
include_paths = current_directory;
}
field_stack_.clear();
builder_.Clear();
// Start with a blank namespace just in case this file doesn't have one.
current_namespace_ = empty_namespace_;
ECHECK(StartParseFile(source, source_filename));
// Includes must come before type declarations:
for (;;) {
// Parse pre-include proto statements if any:
if (opts.proto_mode && (attribute_ == "option" || attribute_ == "syntax" ||
attribute_ == "package")) {
ECHECK(ParseProtoDecl());
} else if (IsIdent("native_include")) {
NEXT();
native_included_files_.emplace_back(attribute_);
EXPECT(kTokenStringConstant);
EXPECT(';');
} else if (IsIdent("include") || (opts.proto_mode && IsIdent("import"))) {
NEXT();
if (opts.proto_mode && attribute_ == "public") NEXT();
auto name = flatbuffers::PosixPath(attribute_.c_str());
EXPECT(kTokenStringConstant);
// Look for the file relative to the directory of the current file.
std::string filepath;
if (source_filename) {
auto source_file_directory =
flatbuffers::StripFileName(source_filename);
filepath = flatbuffers::ConCatPathFileName(source_file_directory, name);
}
if (filepath.empty() || !FileExists(filepath.c_str())) {
// Look for the file in include_paths.
for (auto paths = include_paths; paths && *paths; paths++) {
filepath = flatbuffers::ConCatPathFileName(*paths, name);
if (FileExists(filepath.c_str())) break;
}
}
if (filepath.empty())
return Error("unable to locate include file: " + name);
if (source_filename) {
IncludedFile included_file;
included_file.filename = filepath;
included_file.schema_name = name;
files_included_per_file_[source_filename].insert(included_file);
}
std::string contents;
bool file_loaded = LoadFile(filepath.c_str(), true, &contents);
if (included_files_.find(HashFile(filepath.c_str(), contents.c_str())) ==
included_files_.end()) {
// We found an include file that we have not parsed yet.
// Parse it.
if (!file_loaded) return Error("unable to load include file: " + name);
ECHECK(DoParse(contents.c_str(), include_paths, filepath.c_str(),
name.c_str()));
// We generally do not want to output code for any included files:
if (!opts.generate_all) MarkGenerated();
// Reset these just in case the included file had them, and the
// parent doesn't.
root_struct_def_ = nullptr;
file_identifier_.clear();
file_extension_.clear();
// This is the easiest way to continue this file after an include:
// instead of saving and restoring all the state, we simply start the
// file anew. This will cause it to encounter the same include
// statement again, but this time it will skip it, because it was
// entered into included_files_.
// This is recursive, but only go as deep as the number of include
// statements.
included_files_.erase(source_hash);
return DoParse(source, include_paths, source_filename,
include_filename);
}
EXPECT(';');
} else {
break;
}
}
// Now parse all other kinds of declarations:
while (token_ != kTokenEof) {
if (opts.proto_mode) {
ECHECK(ParseProtoDecl());
} else if (IsIdent("namespace")) {
ECHECK(ParseNamespace());
} else if (token_ == '{') {
return NoError();
} else if (IsIdent("enum")) {
ECHECK(ParseEnum(false, nullptr, source_filename));
} else if (IsIdent("union")) {
ECHECK(ParseEnum(true, nullptr, source_filename));
} else if (IsIdent("root_type")) {
NEXT();
auto root_type = attribute_;
EXPECT(kTokenIdentifier);
ECHECK(ParseNamespacing(&root_type, nullptr));
if (opts.root_type.empty()) {
if (!SetRootType(root_type.c_str()))
return Error("unknown root type: " + root_type);
if (root_struct_def_->fixed) return Error("root type must be a table");
}
EXPECT(';');
} else if (IsIdent("file_identifier")) {
NEXT();
file_identifier_ = attribute_;
EXPECT(kTokenStringConstant);
if (file_identifier_.length() != flatbuffers::kFileIdentifierLength)
return Error("file_identifier must be exactly " +
NumToString(flatbuffers::kFileIdentifierLength) +
" characters");
EXPECT(';');
} else if (IsIdent("file_extension")) {
NEXT();
file_extension_ = attribute_;
EXPECT(kTokenStringConstant);
EXPECT(';');
} else if (IsIdent("include")) {
return Error("includes must come before declarations");
} else if (IsIdent("attribute")) {
NEXT();
auto name = attribute_;
if (Is(kTokenIdentifier)) {
NEXT();
} else {
EXPECT(kTokenStringConstant);
}
EXPECT(';');
known_attributes_[name] = false;
} else if (IsIdent("rpc_service")) {
ECHECK(ParseService(source_filename));
} else {
ECHECK(ParseDecl(source_filename));
}
}
EXPECT(kTokenEof);
if (opts.warnings_as_errors && has_warning_) {
return Error("treating warnings as errors, failed due to above warnings");
}
return NoError();
}
CheckedError Parser::DoParseJson() {
if (token_ != '{') {
EXPECT('{');
} else {
if (!root_struct_def_) return Error("no root type set to parse json with");
if (builder_.GetSize()) {
return Error("cannot have more than one json object in a file");
}
uoffset_t toff;
ECHECK(ParseTable(*root_struct_def_, nullptr, &toff));
if (opts.size_prefixed) {
builder_.FinishSizePrefixed(
Offset<Table>(toff),
file_identifier_.length() ? file_identifier_.c_str() : nullptr);
} else {
builder_.Finish(Offset<Table>(toff), file_identifier_.length()
? file_identifier_.c_str()
: nullptr);
}
}
// Check that JSON file doesn't contain more objects or IDL directives.
// Comments after JSON are allowed.
EXPECT(kTokenEof);
return NoError();
}
std::set<std::string> Parser::GetIncludedFilesRecursive(
const std::string &file_name) const {
std::set<std::string> included_files;
std::list<std::string> to_process;
if (file_name.empty()) return included_files;
to_process.push_back(file_name);
while (!to_process.empty()) {
std::string current = to_process.front();
to_process.pop_front();
included_files.insert(current);
// Workaround the lack of const accessor in C++98 maps.
auto &new_files =
(*const_cast<std::map<std::string, std::set<IncludedFile>> *>(
&files_included_per_file_))[current];
for (auto it = new_files.begin(); it != new_files.end(); ++it) {
if (included_files.find(it->filename) == included_files.end())
to_process.push_back(it->filename);
}
}
return included_files;
}
// Schema serialization functionality:
static flatbuffers::Offset<
flatbuffers::Vector<flatbuffers::Offset<reflection::KeyValue>>>
SerializeAttributesCommon(const SymbolTable<Value> &attributes,
FlatBufferBuilder *builder, const Parser &parser) {
std::vector<flatbuffers::Offset<reflection::KeyValue>> attrs;
for (auto kv = attributes.dict.begin(); kv != attributes.dict.end(); ++kv) {
auto it = parser.known_attributes_.find(kv->first);
FLATBUFFERS_ASSERT(it != parser.known_attributes_.end());
if (parser.opts.binary_schema_builtins || !it->second) {
auto key = builder->CreateString(kv->first);
auto val = builder->CreateString(kv->second->constant);
attrs.push_back(reflection::CreateKeyValue(*builder, key, val));
}
}
if (attrs.size()) {
return builder->CreateVectorOfSortedTables(&attrs);
} else {
return 0;
}
}
static bool DeserializeAttributesCommon(
SymbolTable<Value> &attributes, Parser &parser,
const Vector<Offset<reflection::KeyValue>> *attrs) {
if (attrs == nullptr) return true;
for (uoffset_t i = 0; i < attrs->size(); ++i) {
auto kv = attrs->Get(i);
auto value = new Value();
if (kv->value()) { value->constant = kv->value()->str(); }
if (attributes.Add(kv->key()->str(), value)) {
delete value;
return false;
}
parser.known_attributes_[kv->key()->str()];
}
return true;
}
void Parser::Serialize() {
builder_.Clear();
AssignIndices(structs_.vec);
AssignIndices(enums_.vec);
std::vector<Offset<reflection::Object>> object_offsets;
std::set<std::string> files;
for (auto it = structs_.vec.begin(); it != structs_.vec.end(); ++it) {
auto offset = (*it)->Serialize(&builder_, *this);
object_offsets.push_back(offset);
(*it)->serialized_location = offset.o;
const std::string *file = (*it)->declaration_file;
if (file) files.insert(*file);
}
std::vector<Offset<reflection::Enum>> enum_offsets;
for (auto it = enums_.vec.begin(); it != enums_.vec.end(); ++it) {
auto offset = (*it)->Serialize(&builder_, *this);
enum_offsets.push_back(offset);
const std::string *file = (*it)->declaration_file;
if (file) files.insert(*file);
}
std::vector<Offset<reflection::Service>> service_offsets;
for (auto it = services_.vec.begin(); it != services_.vec.end(); ++it) {
auto offset = (*it)->Serialize(&builder_, *this);
service_offsets.push_back(offset);
const std::string *file = (*it)->declaration_file;
if (file) files.insert(*file);
}
// Create Schemafiles vector of tables.
flatbuffers::Offset<
flatbuffers::Vector<flatbuffers::Offset<reflection::SchemaFile>>>
schema_files__;
if (!opts.project_root.empty()) {
std::vector<Offset<reflection::SchemaFile>> schema_files;
std::vector<Offset<flatbuffers::String>> included_files;
for (auto f = files_included_per_file_.begin();
f != files_included_per_file_.end(); f++) {
const auto filename__ = builder_.CreateSharedString(
RelativeToRootPath(opts.project_root, f->first));
for (auto i = f->second.begin(); i != f->second.end(); i++) {
included_files.push_back(builder_.CreateSharedString(
RelativeToRootPath(opts.project_root, i->filename)));
}
const auto included_files__ = builder_.CreateVector(included_files);
included_files.clear();
schema_files.push_back(
reflection::CreateSchemaFile(builder_, filename__, included_files__));
}
schema_files__ = builder_.CreateVectorOfSortedTables(&schema_files);
}
const auto objs__ = builder_.CreateVectorOfSortedTables(&object_offsets);
const auto enum__ = builder_.CreateVectorOfSortedTables(&enum_offsets);
const auto fiid__ = builder_.CreateString(file_identifier_);
const auto fext__ = builder_.CreateString(file_extension_);
const auto serv__ = builder_.CreateVectorOfSortedTables(&service_offsets);
const auto schema_offset = reflection::CreateSchema(
builder_, objs__, enum__, fiid__, fext__,
(root_struct_def_ ? root_struct_def_->serialized_location : 0), serv__,
static_cast<reflection::AdvancedFeatures>(advanced_features_),
schema_files__);
if (opts.size_prefixed) {
builder_.FinishSizePrefixed(schema_offset, reflection::SchemaIdentifier());
} else {
builder_.Finish(schema_offset, reflection::SchemaIdentifier());
}
}
Offset<reflection::Object> StructDef::Serialize(FlatBufferBuilder *builder,
const Parser &parser) const {
std::vector<Offset<reflection::Field>> field_offsets;
for (auto it = fields.vec.begin(); it != fields.vec.end(); ++it) {
field_offsets.push_back((*it)->Serialize(
builder, static_cast<uint16_t>(it - fields.vec.begin()), parser));
}
const auto qualified_name = defined_namespace->GetFullyQualifiedName(name);
const auto name__ = builder->CreateString(qualified_name);
const auto flds__ = builder->CreateVectorOfSortedTables(&field_offsets);
const auto attr__ = SerializeAttributes(builder, parser);
const auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
std::string decl_file_in_project = declaration_file ? *declaration_file : "";
const auto file__ = builder->CreateSharedString(decl_file_in_project);
return reflection::CreateObject(
*builder, name__, flds__, fixed, static_cast<int>(minalign),
static_cast<int>(bytesize), attr__, docs__, file__);
}
bool StructDef::Deserialize(Parser &parser, const reflection::Object *object) {
if (!DeserializeAttributes(parser, object->attributes())) return false;
DeserializeDoc(doc_comment, object->documentation());
name = parser.UnqualifiedName(object->name()->str());
predecl = false;
sortbysize = attributes.Lookup("original_order") == nullptr && !fixed;
const auto &of = *(object->fields());
auto indexes = std::vector<uoffset_t>(of.size());
for (uoffset_t i = 0; i < of.size(); i++) indexes[of.Get(i)->id()] = i;
size_t tmp_struct_size = 0;
for (size_t i = 0; i < indexes.size(); i++) {
auto field = of.Get(indexes[i]);
auto field_def = new FieldDef();
if (!field_def->Deserialize(parser, field) ||
fields.Add(field_def->name, field_def)) {
delete field_def;
return false;
}
if (field_def->key) {
if (has_key) {
// only one field may be set as key
delete field_def;
return false;
}
has_key = true;
}
if (fixed) {
// Recompute padding since that's currently not serialized.
auto size = InlineSize(field_def->value.type);
auto next_field =
i + 1 < indexes.size() ? of.Get(indexes[i + 1]) : nullptr;
tmp_struct_size += size;
field_def->padding =
next_field ? (next_field->offset() - field_def->value.offset) - size
: PaddingBytes(tmp_struct_size, minalign);
tmp_struct_size += field_def->padding;
}
}
FLATBUFFERS_ASSERT(static_cast<int>(tmp_struct_size) == object->bytesize());
return true;
}
Offset<reflection::Field> FieldDef::Serialize(FlatBufferBuilder *builder,
uint16_t id,
const Parser &parser) const {
auto name__ = builder->CreateString(name);
auto type__ = value.type.Serialize(builder);
auto attr__ = SerializeAttributes(builder, parser);
auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
double d;
StringToNumber(value.constant.c_str(), &d);
return reflection::CreateField(
*builder, name__, type__, id, value.offset,
// Is uint64>max(int64) tested?
IsInteger(value.type.base_type) ? StringToInt(value.constant.c_str()) : 0,
// result may be platform-dependent if underlying is float (not double)
IsFloat(value.type.base_type) ? d : 0.0, deprecated, IsRequired(), key,
attr__, docs__, IsOptional(), static_cast<uint16_t>(padding));
// TODO: value.constant is almost always "0", we could save quite a bit of
// space by sharing it. Same for common values of value.type.
}
bool FieldDef::Deserialize(Parser &parser, const reflection::Field *field) {
name = field->name()->str();
defined_namespace = parser.current_namespace_;
if (!value.type.Deserialize(parser, field->type())) return false;
value.offset = field->offset();
if (IsInteger(value.type.base_type)) {
value.constant = NumToString(field->default_integer());
} else if (IsFloat(value.type.base_type)) {
value.constant = FloatToString(field->default_real(), 16);
}
presence = FieldDef::MakeFieldPresence(field->optional(), field->required());
padding = field->padding();
key = field->key();
if (!DeserializeAttributes(parser, field->attributes())) return false;
// TODO: this should probably be handled by a separate attribute
if (attributes.Lookup("flexbuffer")) {
flexbuffer = true;
parser.uses_flexbuffers_ = true;
if (value.type.base_type != BASE_TYPE_VECTOR ||
value.type.element != BASE_TYPE_UCHAR)
return false;
}
if (auto nested = attributes.Lookup("nested_flatbuffer")) {
auto nested_qualified_name =
parser.current_namespace_->GetFullyQualifiedName(nested->constant);
nested_flatbuffer = parser.LookupStruct(nested_qualified_name);
if (!nested_flatbuffer) return false;
}
shared = attributes.Lookup("shared") != nullptr;
DeserializeDoc(doc_comment, field->documentation());
return true;
}
Offset<reflection::RPCCall> RPCCall::Serialize(FlatBufferBuilder *builder,
const Parser &parser) const {
auto name__ = builder->CreateString(name);
auto attr__ = SerializeAttributes(builder, parser);
auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
return reflection::CreateRPCCall(
*builder, name__, request->serialized_location,
response->serialized_location, attr__, docs__);
}
bool RPCCall::Deserialize(Parser &parser, const reflection::RPCCall *call) {
name = call->name()->str();
if (!DeserializeAttributes(parser, call->attributes())) return false;
DeserializeDoc(doc_comment, call->documentation());
request = parser.structs_.Lookup(call->request()->name()->str());
response = parser.structs_.Lookup(call->response()->name()->str());
if (!request || !response) { return false; }
return true;
}
Offset<reflection::Service> ServiceDef::Serialize(FlatBufferBuilder *builder,
const Parser &parser) const {
std::vector<Offset<reflection::RPCCall>> servicecall_offsets;
for (auto it = calls.vec.begin(); it != calls.vec.end(); ++it) {
servicecall_offsets.push_back((*it)->Serialize(builder, parser));
}
const auto qualified_name = defined_namespace->GetFullyQualifiedName(name);
const auto name__ = builder->CreateString(qualified_name);
const auto call__ = builder->CreateVector(servicecall_offsets);
const auto attr__ = SerializeAttributes(builder, parser);
const auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
std::string decl_file_in_project = declaration_file ? *declaration_file : "";
const auto file__ = builder->CreateSharedString(decl_file_in_project);
return reflection::CreateService(*builder, name__, call__, attr__, docs__,
file__);
}
bool ServiceDef::Deserialize(Parser &parser,
const reflection::Service *service) {
name = parser.UnqualifiedName(service->name()->str());
if (service->calls()) {
for (uoffset_t i = 0; i < service->calls()->size(); ++i) {
auto call = new RPCCall();
if (!call->Deserialize(parser, service->calls()->Get(i)) ||
calls.Add(call->name, call)) {
delete call;
return false;
}
}
}
if (!DeserializeAttributes(parser, service->attributes())) return false;
DeserializeDoc(doc_comment, service->documentation());
return true;
}
Offset<reflection::Enum> EnumDef::Serialize(FlatBufferBuilder *builder,
const Parser &parser) const {
std::vector<Offset<reflection::EnumVal>> enumval_offsets;
for (auto it = vals.vec.begin(); it != vals.vec.end(); ++it) {
enumval_offsets.push_back((*it)->Serialize(builder, parser));
}
const auto qualified_name = defined_namespace->GetFullyQualifiedName(name);
const auto name__ = builder->CreateString(qualified_name);
const auto vals__ = builder->CreateVector(enumval_offsets);
const auto type__ = underlying_type.Serialize(builder);
const auto attr__ = SerializeAttributes(builder, parser);
const auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
std::string decl_file_in_project = declaration_file ? *declaration_file : "";
const auto file__ = builder->CreateSharedString(decl_file_in_project);
return reflection::CreateEnum(*builder, name__, vals__, is_union, type__,
attr__, docs__, file__);
}
bool EnumDef::Deserialize(Parser &parser, const reflection::Enum *_enum) {
name = parser.UnqualifiedName(_enum->name()->str());
for (uoffset_t i = 0; i < _enum->values()->size(); ++i) {
auto val = new EnumVal();
if (!val->Deserialize(parser, _enum->values()->Get(i)) ||
vals.Add(val->name, val)) {
delete val;
return false;
}
}
is_union = _enum->is_union();
if (!underlying_type.Deserialize(parser, _enum->underlying_type())) {
return false;
}
if (!DeserializeAttributes(parser, _enum->attributes())) return false;
DeserializeDoc(doc_comment, _enum->documentation());
return true;
}
flatbuffers::Offset<
flatbuffers::Vector<flatbuffers::Offset<reflection::KeyValue>>>
EnumVal::SerializeAttributes(FlatBufferBuilder *builder,
const Parser &parser) const {
return SerializeAttributesCommon(attributes, builder, parser);
}
bool EnumVal::DeserializeAttributes(
Parser &parser, const Vector<Offset<reflection::KeyValue>> *attrs) {
return DeserializeAttributesCommon(attributes, parser, attrs);
}
Offset<reflection::EnumVal> EnumVal::Serialize(FlatBufferBuilder *builder,
const Parser &parser) const {
const auto name__ = builder->CreateString(name);
const auto type__ = union_type.Serialize(builder);
const auto attr__ = SerializeAttributes(builder, parser);
const auto docs__ = parser.opts.binary_schema_comments
? builder->CreateVectorOfStrings(doc_comment)
: 0;
return reflection::CreateEnumVal(*builder, name__, value, type__, docs__,
attr__);
}
bool EnumVal::Deserialize(Parser &parser, const reflection::EnumVal *val) {
name = val->name()->str();
value = val->value();
if (!union_type.Deserialize(parser, val->union_type())) return false;
if (!DeserializeAttributes(parser, val->attributes())) return false;
DeserializeDoc(doc_comment, val->documentation());
return true;
}
Offset<reflection::Type> Type::Serialize(FlatBufferBuilder *builder) const {
size_t element_size = SizeOf(element);
if (base_type == BASE_TYPE_VECTOR && element == BASE_TYPE_STRUCT &&
struct_def->bytesize != 0) {
// struct_def->bytesize==0 means struct is table
element_size = struct_def->bytesize;
}
return reflection::CreateType(
*builder, static_cast<reflection::BaseType>(base_type),
static_cast<reflection::BaseType>(element),
struct_def ? struct_def->index : (enum_def ? enum_def->index : -1),
fixed_length, static_cast<uint32_t>(SizeOf(base_type)),
static_cast<uint32_t>(element_size));
}
bool Type::Deserialize(const Parser &parser, const reflection::Type *type) {
if (type == nullptr) return true;
base_type = static_cast<BaseType>(type->base_type());
element = static_cast<BaseType>(type->element());
fixed_length = type->fixed_length();
if (type->index() >= 0) {
bool is_series = type->base_type() == reflection::Vector ||
type->base_type() == reflection::Array;
if (type->base_type() == reflection::Obj ||
(is_series && type->element() == reflection::Obj)) {
if (static_cast<size_t>(type->index()) < parser.structs_.vec.size()) {
struct_def = parser.structs_.vec[type->index()];
struct_def->refcount++;
} else {
return false;
}
} else {
if (static_cast<size_t>(type->index()) < parser.enums_.vec.size()) {
enum_def = parser.enums_.vec[type->index()];
} else {
return false;
}
}
}
return true;
}
flatbuffers::Offset<
flatbuffers::Vector<flatbuffers::Offset<reflection::KeyValue>>>
Definition::SerializeAttributes(FlatBufferBuilder *builder,
const Parser &parser) const {
return SerializeAttributesCommon(attributes, builder, parser);
}
bool Definition::DeserializeAttributes(
Parser &parser, const Vector<Offset<reflection::KeyValue>> *attrs) {
return DeserializeAttributesCommon(attributes, parser, attrs);
}
/************************************************************************/
/* DESERIALIZATION */
/************************************************************************/
bool Parser::Deserialize(const uint8_t *buf, const size_t size) {
flatbuffers::Verifier verifier(reinterpret_cast<const uint8_t *>(buf), size);
bool size_prefixed = false;
if (!reflection::SchemaBufferHasIdentifier(buf)) {
if (!flatbuffers::BufferHasIdentifier(buf, reflection::SchemaIdentifier(),
true))
return false;
else
size_prefixed = true;
}
auto verify_fn = size_prefixed ? &reflection::VerifySizePrefixedSchemaBuffer
: &reflection::VerifySchemaBuffer;
if (!verify_fn(verifier)) { return false; }
auto schema = size_prefixed ? reflection::GetSizePrefixedSchema(buf)
: reflection::GetSchema(buf);
return Deserialize(schema);
}
bool Parser::Deserialize(const reflection::Schema *schema) {
file_identifier_ = schema->file_ident() ? schema->file_ident()->str() : "";
file_extension_ = schema->file_ext() ? schema->file_ext()->str() : "";
std::map<std::string, Namespace *> namespaces_index;
// Create defs without deserializing so references from fields to structs and
// enums can be resolved.
for (auto it = schema->objects()->begin(); it != schema->objects()->end();
++it) {
auto struct_def = new StructDef();
struct_def->bytesize = it->bytesize();
struct_def->fixed = it->is_struct();
struct_def->minalign = it->minalign();
if (structs_.Add(it->name()->str(), struct_def)) {
delete struct_def;
return false;
}
auto type = new Type(BASE_TYPE_STRUCT, struct_def, nullptr);
if (types_.Add(it->name()->str(), type)) {
delete type;
return false;
}
}
for (auto it = schema->enums()->begin(); it != schema->enums()->end(); ++it) {
auto enum_def = new EnumDef();
if (enums_.Add(it->name()->str(), enum_def)) {
delete enum_def;
return false;
}
auto type = new Type(BASE_TYPE_UNION, nullptr, enum_def);
if (types_.Add(it->name()->str(), type)) {
delete type;
return false;
}
}
// Now fields can refer to structs and enums by index.
for (auto it = schema->objects()->begin(); it != schema->objects()->end();
++it) {
std::string qualified_name = it->name()->str();
auto struct_def = structs_.Lookup(qualified_name);
struct_def->defined_namespace =
GetNamespace(qualified_name, namespaces_, namespaces_index);
if (!struct_def->Deserialize(*this, *it)) { return false; }
if (schema->root_table() == *it) { root_struct_def_ = struct_def; }
}
for (auto it = schema->enums()->begin(); it != schema->enums()->end(); ++it) {
std::string qualified_name = it->name()->str();
auto enum_def = enums_.Lookup(qualified_name);
enum_def->defined_namespace =
GetNamespace(qualified_name, namespaces_, namespaces_index);
if (!enum_def->Deserialize(*this, *it)) { return false; }
}
if (schema->services()) {
for (auto it = schema->services()->begin(); it != schema->services()->end();
++it) {
std::string qualified_name = it->name()->str();
auto service_def = new ServiceDef();
service_def->defined_namespace =
GetNamespace(qualified_name, namespaces_, namespaces_index);
if (!service_def->Deserialize(*this, *it) ||
services_.Add(qualified_name, service_def)) {
delete service_def;
return false;
}
}
}
advanced_features_ = schema->advanced_features();
if (schema->fbs_files())
for (auto s = schema->fbs_files()->begin(); s != schema->fbs_files()->end();
++s) {
for (auto f = s->included_filenames()->begin();
f != s->included_filenames()->end(); ++f) {
IncludedFile included_file;
included_file.filename = f->str();
files_included_per_file_[s->filename()->str()].insert(included_file);
}
}
return true;
}
std::string Parser::ConformTo(const Parser &base) {
for (auto sit = structs_.vec.begin(); sit != structs_.vec.end(); ++sit) {
auto &struct_def = **sit;
auto qualified_name =
struct_def.defined_namespace->GetFullyQualifiedName(struct_def.name);
auto struct_def_base = base.LookupStruct(qualified_name);
if (!struct_def_base) continue;
std::set<FieldDef *> renamed_fields;
for (auto fit = struct_def.fields.vec.begin();
fit != struct_def.fields.vec.end(); ++fit) {
auto &field = **fit;
auto field_base = struct_def_base->fields.Lookup(field.name);
const auto qualified_field_name = qualified_name + "." + field.name;
if (field_base) {
if (field.value.offset != field_base->value.offset)
return "offsets differ for field: " + qualified_field_name;
if (field.value.constant != field_base->value.constant)
return "defaults differ for field: " + qualified_field_name;
if (!EqualByName(field.value.type, field_base->value.type))
return "types differ for field: " + qualified_field_name;
} else {
// Doesn't have to exist, deleting fields is fine.
// But we should check if there is a field that has the same offset
// but is incompatible (in the case of field renaming).
for (auto fbit = struct_def_base->fields.vec.begin();
fbit != struct_def_base->fields.vec.end(); ++fbit) {
field_base = *fbit;
if (field.value.offset == field_base->value.offset) {
renamed_fields.insert(field_base);
if (!EqualByName(field.value.type, field_base->value.type))
return "field renamed to different type: " + qualified_field_name;
break;
}
}
}
}
// deletion of trailing fields are not allowed
for (auto fit = struct_def_base->fields.vec.begin();
fit != struct_def_base->fields.vec.end(); ++fit) {
auto &field_base = **fit;
// not a renamed field
if (renamed_fields.find(&field_base) == renamed_fields.end()) {
auto field = struct_def.fields.Lookup(field_base.name);
if (!field) {
return "field deleted: " + qualified_name + "." + field_base.name;
}
}
}
}
for (auto eit = enums_.vec.begin(); eit != enums_.vec.end(); ++eit) {
auto &enum_def = **eit;
auto qualified_name =
enum_def.defined_namespace->GetFullyQualifiedName(enum_def.name);
auto enum_def_base = base.enums_.Lookup(qualified_name);
if (!enum_def_base) continue;
for (auto evit = enum_def.Vals().begin(); evit != enum_def.Vals().end();
++evit) {
auto &enum_val = **evit;
auto enum_val_base = enum_def_base->Lookup(enum_val.name);
if (enum_val_base) {
if (enum_val != *enum_val_base)
return "values differ for enum: " + enum_val.name;
}
}
}
return "";
}
} // namespace flatbuffers