A field with key attribute must always be written on the message so it
can be looked up by key. There is a edge case where inserting a key
field with same value as default would prevent it to be written on
the message and later cannot be found when searched by key.
Any string type that is supported by CreateString(), e.g.
const char* or string_view will now also work.
Signed-off-by: Henner Zeller <hzeller@google.com>
* Adds implementation flag for swift
Forces internal flag when using @_implementationOnly in swift
Fixes access type for verifier functions & encoder functions
Updates generated code
* Addresses PR comments & adds a code gen dir within the swift tests
* Adds test case for no-include
* Fixes code gen script
Removes prefix
* Started to migrate to target_compile_options
* combined compile options together. Added Mac CI builds
* remove arm build (not supported). Fixed old-style-casts
* moved to using a ProjectConfig interface library to specify options
* remove the explicit CMAKE_CXX_STANDARD
* code gen flexbuffer verifier
* remove verify nested flexbuffers from flexbuffers
* made function static, and placed higher in file
* moved function to own header
* Add overloads to Add/Put for ArraySegment and IntPtr
In order to allow using code to reduce memory allocations, add overloads to ByteBuffer's and FlatBuffersBuilder's Put/Add methods that take ArraySegment<T> or IntPtr respectively.
Also, adaptions to the c# code generator in flatc to emit corresponding CreateVectorBlock() overloads
* Add missing files generated with generate_code.py
The previous commit changed the C# code generate, but didn't contain the updated generated test files.
* Incorporate review findings
(1) Adhere to 80 characters limit.
(2) In FlatBufferBuilder.Add(IntPtr,int), move zero length check topmost and add sanity check against negative input
* Started applying Namer to Java.
- Java didn't previously have keyword escaping
- Added prefixes and suffixes to the Namer methods
- TODO: migrate previous namer applications to using pre/suffixes
- Java methods / functions are interesting, it's mostly camel case
except when it involves a struct/enum name. That section is Keep case
- I changed the casing for some internal arguments/variables. This
violates the "don't change genfiles" rule that I've been using but it
shouldn't break user code.
- LegacyJavaMethod2 is interesting. Basically, Java has a "mixed" case
convention where it's camel case, except for the type/variant name
itself, which is keep case. So a type foo_bar would become getfoo_bar
instead of getFooBar.
* small fix
* Namer for Namespaces
* removed unused parameter, add const everywhere
* Remove unused argument
* More unused args
* Use mutable reference out parameters
* Made more strings const and inlined const empty strings
* remove do not submit
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
* [TS] Fix reserved words as arguments (#6955)
* [TS] Fix generation of reserved words in object api (#7106)
* [TS] Fix generation of object api
* [TS] Fix MakeCamel -> ConvertCase
* [C#] Fix collision of field name and type name
* [TS] Add test for struct of struct of struct
* Update generated files
* Add missing files
* [TS] Fix query of null/undefined fields in object api
* Add example for type field name collision
* Apply Namer to Dart.
- Also refactor idl_gen_dart a bit
- to use more const and references
- out parameters should be the last argument
* Add keyword test
* minor fixes
* fix merge
* extra 's'
* move dart keyord into dart dir
* Address comments
* Use $ for escaping keywords
* Outparameters for namespace_map
* Escape dollar in toString
* Escape dollar in toString2
* Use UpperCamelCase for types and variants
* try to fix ToString
* namer Type fixes
* Remove path prefixing in imports
* gen code
Co-authored-by: Casper Neo <cneo@google.com>
* Handle invalid root offset
* Handle vtable offset invalidation
* Added script generator. Add more cases through vtable ref table size
* review responses
* vtable offset validation
* Moved padding insertion to the end. Tests invalid field lenghts
* table offsets validated. Added type after field
* validate string length
* add todo
* invalid vector length
* invalid structs
* general cleanup
* reworded invalid offsets
* example for vector of structs
* invalid vector of tables
* invalid vector of strings
* invalid vector of scalars
* vector of unions
* validate union type value
* invalid vector union type values
* 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
* [TS] Fix reserved words as arguments (#6955)
* [TS] Fix generation of reserved words in object api (#7106)
* [TS] Fix generation of object api
* [TS] Fix MakeCamel -> ConvertCase
* [TS] Add test for struct of struct of struct
* Update generated files
* Add missing files
* [TS] Fix query of null/undefined fields in object api
The headline here is adding a flatbuffer_ts_library rule for generating
typescript code in bazel. This entails some non-trivial other changes,
but ideally none are user-visible.
In particular:
* Added a --ts-flat-file flag that generates a single *_generated.ts
file instead of separate files for each typescript type. This makes
bazel much happier.
* Import the bazel rules_nodejs stuff needed to support building
typescript in bazel
* Move flatbuffers.ts to index.ts because I wasn't sure how to make
bazel comprehend the "main" attribute of the package.json. Happy
to take another stab at figuring that out if really needed.
* Fix another couple keyword escaping spots in typescript...
* Add Object to the list of reserved keywords.
* Properly escape keywords in type names.
* Properly escape keywords in enum names.
* Properly escape keywords in enum field names.
* Fix 64-bit default numeric enum values in typescript
If you had a default value that wasn't a valid enum value (e.g., a zero
if you used a bit_flag setting, like you get with AdvancedFeatures
in reflection.fbs), we weren't using BigInt.
* Run generate_code.py
* [DART] Handle deprecated fields & invalid enum defaults
* Update .NET test
* [Rust] Add length checks to arrays and vectors.
The previous behavior allowed for out of bounds access in
the public API (#7011).
* bump semver and test warning
Co-authored-by: Casper Neo <cneo@google.com>
* Unified name case conversion to single method
* Convert bfbs_gen to use ConvertCase
* convert rust to use ConvertCase
* Convert idl_parser to use ConvertCase
* Convert MakeScreamingCamel to ConvertCase
* Replaced MakeCamel with ConvertCase
* minor fixes
I'm not seeing the reason why we didn't attempt to support transitive
dependencies for flatbuffer_cc_library, and the current setup makes
having to propagate new dependencies to all of their recursive
dependents obnoxious.
* [go] always write required types
* support optional scalars in go
* generate optional_scalars and monster_test
* restore original behavior for non-optional scalars
* add tests
* Change Rust generated file defaults
After #6731, flatc changed its default behavior
for generating rust code to fix some importing issues.
This was a breaking change which invlidated the patch release,
`flatc 2.0.5` (#7081). This PR reverses the default so we can
release a patch update. However, does break Rust users who work at
HEAD.
* Bump flatc patch version (2.0.6)
Co-authored-by: Casper Neo <cneo@google.com>
* fix for rust build
* Rust: Implement Serialize on generated types
For debugging convenience it is really handy to be able to dump out
types as another format (ie: json). For example, if we are logging a
type to a structured logging system, or even printing it out in a
structured way to the console.
Right now we handle this by shelling out to `flatc` which is not ideal;
by implementing Serialize on the generated types we can use any of the
Serializer-implementing packages for our structured debug output.
* clang-format
* Make the flatbuffers Rust crate only have an optional dependency on the `serde` packages.
* fix warning
* fix rust test build
* Oh yeah this needs to be initialized
* fix toml syntax
* code review feedback
* rebuild test data
Augment the C++ generator to emit a C++ copy constructor and a by-value
copy assignment operator. This is enabled by default when the C++
standard is C++11 or later. These additional functions are only emitted
for objects that need it, typically tables containing other tables.
These new functions are declared in the object table type and are
defined as inline functions after table declarations.
When these new functions are declared, a user-defined
explicitly-defaulted default constructor and move constructor are also
emitted.
The copy assignment operator uses the copy-and-swap idiom to provide
strong exception safety, at the expense of keeping 2 full table copies
in memory temporarily.
fixes#5783
* Add --warnings-as-errors to flatc compiler.
With this option set, flatc will return an error on parsing if
any warnings occurred.
* Add unit test for opts.warnings_as_errors.
* Change explicit option setting to default.
