* CMake: Fix target options
* update CMake package version after protocol sync
* Update protocol to 1.59.0
* protocol_gen: change OneOf to `object` type
The DAP spec introduces ambiguities with its particular uses of OneOf,
which means that we can't deserialize the variants generated from it.
Just set OneOf to an `object` type, like godap does.
Add options to link against external JSON libraries
This adds two options to the CMake config:
`CPPDAP_USE_EXTERNAL_NLOHMANN_JSON_PACKAGE` and `CPPDAP_USE_EXTERNAL_RAPIDJSON_PACKAGE`
If either of these is set, instead of using the json submodule, the build will be configured by using `find_package()` and linking against the specified JSON library.
In C++11, shared_ptr has an explicit operator bool which
means that a shared_ptr can't be implicitly converted to a bool.
adding an explicit cast is a valid fix to the code.
This was entirely broken:
* The `Server` `Impl` used a `do{}while(false)` block, which never attempted to accept another connection after the first connection closed (#69)
* The `Server` `Impl` could deadlock with the mutex being locked by both the thread calling `isRunning()` and `stopWithLock()` waiting on `thread.join()`.
* `Socket::accept()` didn't check that the returned socket was valid, and could return a `ReaderWriter` that would just error on IO.
* `Socket::accept()` could deadlock on macOS as `shutdown()` can seemingly fail to unblock an accept call. This has been worked around by calling `close()` outside of the mutex write-lock. This introduces a potential race, but I'm not sure of a better solution right now.
Fixes: #69
CMake installation target has been broken for years.
cppdap is only internally used and tested as a library that is included as a sub-module of another project.
In order for a package to be installable, it really needs a stable ABI / API, which we've made no explicit guarantees about.
cppdap is small enough that it can easily be included in a third_party directory of another project.
Remove the broken, unmaintained installation rules.
Fixes: #54
Classify 'advanced' APIs from the 'basic' API.
Move `bind()` up top, which is the typical way to use the Session.
Comment on the other APIs that you should probably just use `bind()`.
Document all the members that are now public.
Fixes: #63
Schema top-level definitions, such as `InvalidatedAreas` and `SteppingGranularity` were being emitted as empty structures, when they were actually enumerators.
Re-work protocol_gen.go to emit these correctly.
Also bumps the protocol to DAP version 1.46.0
When there is partial message in reader, popping out the
characters lead to parse errors on the subsequent attempt
to parse. To avoid this, the last matched characters'
index is stored during parsing and reset on error.
Bind API is split into connect and startProcessingMessages calls.
This enables users to directly call connect and manage processing
messages on user threads.
bindNoThread in conjunction with OnDataAvailable will provide
users with a choice of thread to process requests on. This is
useful when the user relies on single threaded message loop based
design to avoid locking.
`license-checker` is a tool that verifies each file has contains a permitted license header.
See https://github.com/ben-clayton/license-checker for more information.
Also add missing licenses to presubmit scripts.
If any of the `CPPDAP_?SAN` sanitizers are enabled, then the build would fail as the compile flags would collide with hard-coded use of address sanitizer for the fuzzer.
Don't hard code address sanitizer for the fuzzer - just use the CMake flag options.
Define `JSON_NOEXCEPTION` to avoid raising exceptions.
Exceptions are usually disabled for Google projects (https://google.github.io/styleguide/cppguide.html#Exceptions).
Also pass `false` to the `allow_exceptions` parameter of `nlohmann::json::parse()`.
Issue identified by @kuafuwang in #26.
`DAP_IMPLEMENT_STRUCT_TYPEINFO_EXT` and `DAP_STRUCT_TYPEINFO_EXT` are two new flavors of `DAP_IMPLEMENT_STRUCT_TYPEINFO` and `DAP_STRUCT_TYPEINFO` that allow you to derive message types.
This involved a bit of reworking on the serializer interfaces.
Added test.
Issue: #32
