* Fix miri problems by assuming alignment is 1 in rust
* Removed is_aligned fn from rust verifier.
* Add back is_aligned, but make it w.r.t. buffer[0]
* touch unused variable
* touch unused variable
* +nightly
* Move Rust miri testing into its own docker
* fix bash
* missing one endian conversion
* fix endianness2
* format stuff
Co-authored-by: Casper Neo <cneo@google.com>
This commit fixes handling of default and NULL `key` fields in `Parser::ParseVector` (#5928).
The JSON generator updated. It outputs `key` fields even if the `--force-defaults` option is inactive.
Additional test cases for `key` added.
* [idl_parser] Improve stack overflow protection
Add stack overflow protection for Flexbuffer and nested Flatbuffer parsers.
Replaces the `Recurse()` method by the new ParseDepthGuard RAII class.
* Remove move operator from Parser.
It was wrong decision to add move ctor and assignment into Parser class.
These operators will make it extremely difficult to add constant or reference fields in the future.
* Remove ';' from definition of FLATBUFFERS_DELETE_FUNC
* Format code
* Make this PR compatible with MSVC2010 (it doesn't support inherited ctor)
* [idl_parser] Check the range of explicitly set field's id value
The explicitly set `id` attribute should be a non-negative value of the `voffset_t` type.
* Format FieldIdentifierTest()
* Rebuild the way swift handles structs from scratch
* Updates docs, and sample binary
* Replaces InMemory to Mutable
* Migrates docs from inmemory
* use inline for some functions
* Renamed Mutable objects
* Updates documentation
* Add vectorNumElements attribute to Builder for simpler vector creation.
This adds a default to EndVector which should simplify its use.
* Update tutorial to reflect new default arg in Python EndVector.
* Remove optional argument to Python EndVector.
* Add generated files.
* Unset Builder.vectorNumElems when not in use.
* Added missing EndTable() call to VerifyObject()
VerifyObject called VerifyTableStart() but not EndTable(). This made Verifier::VerifyComplexity() increase depth_ with each table, not with the depth of tables.
https://groups.google.com/forum/#!topic/flatbuffers/OpxtW5UFAdg
* Added Check to VerifyAlignment
https://stackoverflow.com/questions/59376308/flatbuffers-verifier-returns-false-without-any-assertion-flatbuffers-debug-veri
* Add GetStringView (Convenience function to get string_view from a String returning an empty string_view on null pointer) like GetString, GetCstring
* flatc should warn, when an attribute is attached more than once.
flatc.exe -b duplicate.fbs
warning: duplicate.fbs(5, 36): warning: attribute already found: priority
duplicate.fbs:
namespace MyGame;
attribute "priority";
table Monster (priority:1, priority:2) {
}
root_type Monster;
* flatc should support --binary --schema with optional scalar fields.
This fixes 'error: Optional scalars are not yet supported in at least one the of the specified programming languages.' when calling flatc.exe --binary --schema with a schema containing optional scalars.
* Generate nullable properties in C# object-based API for optional scalars.
tests\generate_code.bat extended to test this.
Ran tests\generate_code.bat
Ran tests\Flatbuffers.Test\NetTest.bat
* %TEST_BASE_FLAGS% replaced with --gen-object-api in generate_code.bat, because only this is part of this PR. Added this same flag to generate_code.sh
* generate_code.bat and generate_code.sh changed to only test c# with object based api.
* Fix C/C++ Create<Type>Direct with sorted vectors
If a struct has a key the vector has to be sorted. To sort the vector
you can't use "const".
* Changes due to code review
* Improve code readability
* Add generate of JSON schema to string to lib
* option indent_step is supported
* Remove unused variables
* Fix break in test
* Fix style to be consistent with rest of the code
* Add option --no-warnings to inhibit all warnings
* Fix order of member initialization
* Add documentation for --no-warnings
* Updated comments and fixed a fundemental type error.
* bump rust flatbuffers semver
* Initial commit with verifier, need to clean up
* Verifier tested. Needs clean up and refactoring.
* Display for InvalidFlatbuffer and better errors for strings
* SimpleToVerify, some refactoring
* Combined VerifierType TableAccessorFuncBody into FollowType
* scrub todos
* Update Rust get_root functions.
There are 6 variants, with verifier options, default verifier options
and no verification "fast".
* Rename root fns
* inline
* Update to use thiserror
* fix for bad compiler
* improve error formatting
* Replace multiply with saturating_multiply
* saturating adds too
* Add docs disclaiming experimental verification system
Co-authored-by: Casper Neo <cneo@google.com>
Prior to this commit the default C++ code generator was `c++0x`.
A code generated with `c++0x` code-gen might have a vulnerability (undefined behavior) connected evolution of enums in a schema. This UB could break the backward compatibility if previously generated code casts an unknown enumerator to enum type that knows nothing about future enumerators added to the schema.
The main differences between `c++0x` and `c++11`:
- generated enums use explicitly declared underlying type;
- generated object-API tables don't declare default ctor() explicitly, instead of it default data member initializers are generated.
Please use `flatc --cpp-std c++0x` option for backward compatibility with old compilers.
- add a new method ParseJson to minimize failures during fuzzing
- add default (conditional) move-constructor for Parser
- add a new monster_fuzzer
- switch fuzzers to C++17 and `test/cpp17` generated code
* Add --require-explicit-ids to require explicit ids
We just got bit by a well intentioned developer forgetting that field
order by default is the field index. 3 people missed it in review.
I'm looking at ways to make it harder to mess up. We are requesting
that developers explicitly id all fields in tables. Automatic (opt in
for others) enforcement of this will help the effort succeed. This
patch adds a command line flag which lets the user require ids on all
fields in tables.
* Added docs to Compiler.md as well
* idl_gen_json_schema.cpp: Changed generation of array element types
#6175
* idl_gen_json_schema.cpp: Simplified indent generation as suggested by @vglavnyy
#6175
* Remove _POSIX_C_SOURCE and _XOPEN_SOURCE definitions when compiling on OpenBSD
* Only define _POSIX_C_SOURCE and _XOPEN_SOURCE for mingw/cygwin platforms
* Only define POSIX statements for mingw/cygwin/qnx platforms
* Added missing EndTable() call to VerifyObject()
VerifyObject called VerifyTableStart() but not EndTable(). This made Verifier::VerifyComplexity() increase depth_ with each table, not with the depth of tables.
https://groups.google.com/forum/#!topic/flatbuffers/OpxtW5UFAdg
* Added Check to VerifyAlignment
https://stackoverflow.com/questions/59376308/flatbuffers-verifier-returns-false-without-any-assertion-flatbuffers-debug-veri
* Add GetStringView (Convenience function to get string_view from a String returning an empty string_view on null pointer) like GetString, GetCstring
* flatc should warn, when an attribute is attached more than once.
flatc.exe -b duplicate.fbs
warning: duplicate.fbs(5, 36): warning: attribute already found: priority
duplicate.fbs:
namespace MyGame;
attribute "priority";
table Monster (priority:1, priority:2) {
}
root_type Monster;
* flatc should support --binary --schema with optional scalar fields.
This fixes 'error: Optional scalars are not yet supported in at least one the of the specified programming languages.' when calling flatc.exe --binary --schema with a schema containing optional scalars.
* Refactor idl_gen_rust to a ForAllX continuation pattern.
* Removed unneeded SetValue and updated sample rust gencode
* Make Rust flatbuffers print right
* Generated code and removed unnecessary trait constraint
* bumped rust version. Release required
* removed an unwrap in Rust Debug-print unions
* Tested formatting flatbuffers in rust.
* Set float precision in flaky debug-print test
* impl Debug for structs too
Co-authored-by: Casper Neo <cneo@google.com>
* Updated SupportsAdvancedUnionFeatures to look out for string
* Mass refactoring to use BASE_TYPE helper functions.
Co-authored-by: Casper Neo <cneo@google.com>
* Rework enums in rust.
They're now a unit struct, rather than an enum. This is a
backwards incompatible change but the previous version had UB
and was also backwards incompatible so...
* Update and test sample rust flatbuffers
* Use bitflags crate to properly support rust enums.
Previously, the bitflags attribute was just ignored. This is a breaking change
as the bitflgs API is not like a normal rust enum (duh).
* variant_name() -> Option<_>
* repr transparent
* Reexport bitflags from flatbuffers
* Make bitflags constants CamelCase, matching normal enums
* Deprecate c-style associated enum constants
Co-authored-by: Casper Neo <cneo@google.com>
* Fix C/C++ Create<Type>Direct with sorted vectors
If a struct has a key the vector has to be sorted. To sort the vector
you can't use "const".
* Changes due to code review
* Improve code readability
* Add generate of JSON schema to string to lib
* option indent_step is supported
* Remove unused variables
* Fix break in test
* Fix style to be consistent with rest of the code
* Add option to not generate direct copy methods.
The direct copy methods generated by flatc utilize std::vector which isn't allowed on some embedded systems. Permit users of the compiler to not generate these methods so they don't have to be stubbed out.
* Update docs for no-cpp-direct-copy option.
* Fixed rust nested flatbuffers for tables other than self
* replaced lifetimes
* Use WrapInNameSpace and also update samples
Co-authored-by: Casper Neo <cneo@google.com>
* GRPC implementation for Typescript
* Fixes a couple of issues
* Finished implementing the typescript support for grpc
* Updated generated code
* Fixes CI
* Moved C++ to optional_scalars2 and added some tests.
Also deleted unused optional_scalars_generated.lobster
* Fixed whitespece in C++ gencode & fixed BUILD file
* Moved C++ onto optional_scalars2 in the .bat file
Co-authored-by: Casper Neo <cneo@google.com>
UnpackTo copies vector elements one-by-one, which can be very inefficient depending on the quality of the compiler optimizations performed. This change updates the operation for vectors of bytes that aren't enums to use 'std::copy', which is usually highly optimized.
vectors of types that are more than one byte can't be optimized in this way because of the endianness of the serialized bytes vs. the target architecture endianness.
vectors of enums can't be optimized because they are required to be static_cast into the appropriate enum type when stored in the vector.
vectors of bools can be optimized in most cases, but since the standard
allows std::vector<bool> template specialization for space-savings,
std::copy doesn't work on every implementation (looking at you
Microsoft). Thus, this optimization is skipped for vector<bool>.
For a specific example, this improves the latency of unpacking large buffers on the Hexagon DSP by about 10x.
Co-authored-by: Matthew Markwell <markwell@google.com>
