This is a collection of solutions to the Advent of Code 2024 programming implemented in the C programming language.
For this project, we focus on efficient algorithms and include formal pseudocode for each solution.
Day 9: My original algorithm for Day 9(a) and Day 9(b) had space complexity
exceeding
Day 18: The union–find-based technique for Day 18(b) is inspired by this Reddit conversation.
Day 14(b), Day 17(b), Day 24(b): By design, there is no generalized heuristic-free algorithm for Day 14(b), Day 17(b), and Day 23(b). These reverse-engineering problems that require assumptions about the input.
Day 20: The generalized solution for Day 20(a) and Day 20(b) is inefficient.
Day 21: My favorite solution for Day 21(a) and Day 21(b) comes from the insights at expectancyViolation/aoc_2024. I have no improvements to offer.
Day 22(b): The generalized solution for Day 22(b) is inefficient.
These programs are cross-platform, having been tested on Windows and Linux operating systems.
Important: There is no error-checking. An invalid input usually results in a segmentation violation.
Important: I have tested each program on the example test cases and two
puzzle input files - not a large sample size! I have avoided assumptions about
the input, sometimes even at the cost of performance. However, to avoid memory
allocation, all buffers have fixed sizes. Ensure that all buffers (see #define
statements) have sufficient capacity before running. Not doing so could result
in a stack smashing, segmentation violation, or worse.
I am working within the following constraints to ensure high code quality.
- Final solutions must be implemented in the C programming language following the C99 standard, with the following non-standard assumptions:
- asssume that all characters are ASCII-encoded.
- Bounds checking is not required for data structures whose capacity is defined by a macro.
- The return values of all C standard library functions must be checked, except for those returned from the following:
fprintfmemcpymemmovememsetprintf