Details
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
19 | Why not making this iterator? I think it will make the code more readable. Plus it will still work with the regular pointers. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
19 | Pointers will not be enough here, but iterators probably require more boilerplate on caller side. Not sure. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
35 | templates don't need inline to avoid multiple definitions |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
19 | What I am proposing is this: This will work both with regular pointers and STL integrators depending on the instantiation of the template. For example if you are using a container you can pass a back_insert_iterator. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
19 | Problem is we don't have STL in compiler-rt, so for anything but pointer you to define one, which is a class with at least deference and increment operators vs just a lambda. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
19 | I don't have strong preference about Iterator vs lambda. Both will work. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
37 | Nit: output value should be at the end of the list and a pointer according to g3 styleguide. Up to you. |
compiler-rt/lib/sanitizer_common/sanitizer_leb128.h | ||
---|---|---|
37 | The bit about pointers is no longer true: … while non-optional output and input/output parameters should usually be references (which cannot be null). See https://google.github.io/styleguide/cppguide.html#Inputs_and_Outputs. |
Why not making this iterator? I think it will make the code more readable. Plus it will still work with the regular pointers.