You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/algorithm.md
+6-6Lines changed: 6 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,12 +16,12 @@
16
16
-`seq::unique`: removed duplicates from a range in a stable way. This is very similar to `std::unique` except that the range does not need to be sorted. It uses a hash table under the hood to find duplicate values. A custom hash function and comparison function can be passed for custom types.
17
17
18
18
-`seq::net_sort` and `seq::net_sort_size`: "new" generic stable sorting algorithm that is used everywhere within the seq library. `seq::net_sort` is a merge sort algorithm with the following specificities:
19
-
- Bottom-up merging instead of the more traditional top-down approach,
20
-
- Small blocks of 8 elements are sorted using a sorting network,
21
-
- Bidirectional merging is used for relocatable types,
22
-
- Ping-pong merge is used to merge 4 sorted ranges,
23
-
- Can work without allocating memory through a (potentially null) user provided buffer,
24
-
- Also works on bidirectional iterators.
19
+
- Bottom-up merging instead of the more traditional top-down approach,
20
+
- Small blocks of 8 elements are sorted using a sorting network,
21
+
- Bidirectional merging is used for relocatable types,
22
+
- Ping-pong merge is used to merge 4 sorted ranges,
23
+
- Can work without allocating memory through a (potentially null) user provided buffer,
24
+
- Also works on bidirectional iterators.
25
25
26
26
If provided buffer is one of `seq::default_buffer`, `seq::medium_buffer`, `seq::small_buffer` or `seq::tiny_buffer`, this function will try to allocate memory.
0 commit comments