Added ability to offer a buffer range to a publication #401
Conversation
|
I had some time today to review how this is implemented in Java and modified the C++ version to map more closely to it. Another thing I would like some feedback on is that some of the functions could be improved by removing the end iterator as a parameter. Since the length is calculated in the |
|
There were quite a few conflicts, so I had to rebase. Let me know if there's anything else I need to do to get this pushed through. |
|
Sorry for the delay. I'll look at this tomorrow for a full review. To be honest, not sure I like following the C++ algorithms technique of iterator template params. Not sure it fits here. But I'll review with an open mind. A |
|
No worries on the delay. My main reason for wanting to use an iterator is because I have a use case involving a linked list of buffers and would prefer to not have to allocate an array on each offer. I figured that an iterator-based approach could cover both the linked list and array use cases. Unfortunately, this is a very C++-specific solution, so the code doesn't match exactly with the Java side of things. But I still tried to make it match up the best that I could. If you have any questions or comments, or if you want any examples for unit tests or something, let me know. Thanks. |
|
@nicklauslittle thanks! I spent some time exploring this and it does fairly well. It's a more C++-ishy way than using a native array. And probably better overall. Leaving options for usage with the standard algorithms is quite handy. Nice work! |
|
Glad I could help out! |
I wanted the ability to offer multiple buffers and combine them into a single message. I noticed a similar issue had been brought up recently #(394), so I figured I would share what I've done. Comments are welcome.