Derive C

Elevator Pitch

When using C for complex projects, the lack of generics is frustrating:

• tradeoff performance and store runtime type information (e.g. size of items in a vector)
• tradeoff usability and restrict sizes of types, use unions (e.g. vector of only 8 byte items, user must cast to their own type, debuggers unaware of real type)
• tradeoff readability & development speed by outsourcing data structure generation to another language (cmake calling python to generate headers)
• tradeoff tooling for formatting, linting & intellisense with complex preprocessor macros to generate data structures

Derive-C aims to avoid these, and to get an experience close to simple templates in C++:

• templates are edittable (with intellisense), lintable, formattable independent of their usage.
• instances of templates should be clangd friendly / show types & completions
• no performance tradeoff
• easy to debug with in GDB

See examples in ./examples

Use

In a CMakeLists.txt

include(FetchContent)
 
FetchContent_Declare(
    derive-c
    GIT_REPOSITORY https://github.com/OliverKillane/derive-C
    GIT_TAG <Chosen derive-c version>
)
FetchContent_MakeAvailable(derive-c)

Develop

For development, the clang toolchain is recommended.

nix-shell toolchain/clang.nix # from repo root

cmake -S . -B build -GNinja

cd build
ninja
ctest -j
ninja format
ninja lint
ninja docs
ninja coverage

Then when opening vscode from the nix-shell, the correct clangd & library paths are used:

code .

For using infer, infer must be installed separately (it is not yet packaged with nix - see here)

• Build with clang

  cmake -S . -B build -DEXTERNALS=Off
  infer run --compilation-database build/compile_commands.json --bufferoverrun --liveness --pulse
  infer explore

• Statically detects generic bugs (e.g. use after free, buffer overrun, integer overflow)

TODO

In development, remaining tasks:

• Fix remaining infer-detected casting issues
• Finish gdb pretty printers
• Increase coverage for hashmap
• Regression benchmarks
• compare & optimise hashmap versus: ankerl

container
    vector 
        static  // current staticvec
        dynamic // normal vector
    map
        linear     // linear search map
        decomposed // current hashmap
        ankerl     // new ankerl hashmap
        btree      // btreemap
    queue
        circular // current: circular buffer
        deque    // current: double ended queue
    arena
        basic // current implementation
        generational
utils
    option // current option
    result
    string
        short
        intern
    ptr
        count
        observe // notified 
        span
    closure
        dynamic
        small
algorithms
    sort
        binarysort
    search
        binarysearch
    hash
        murmur
        integer
allocs
    std
    staticbump
    test
    alloc
test
    mock

References

• moving-fast-with-software-verification