xenia/home
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Refactor find_diff_rec and detect kept prefix/suffixes early

Browse Source 
Certain cases, like diffing an empty buffer with a big buffer, were
very slow, now this should get better as we will directly detect the
matching eol at the end of both buffers, end then immediatly detect
we need to add the rest of the big buffer.

We still are too slow on some general diff when there is a lot of
differences.
This commit is contained in:
Maxime Coste 2017-07-17 13:30:11 +02:00
parent 388ada8142
commit d90cd6de77
1 changed files with 23 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
src/diff.hh
View File

@ -96,6 +96,9 @@ struct Diff
inline void append_diff(Vector<Diff>& diffs, Diff diff) inline void append_diff(Vector<Diff>& diffs, Diff diff)
{ {
if (diff.len == 0)
return;
if (not diffs.empty() and diffs.back().mode == diff.mode if (not diffs.empty() and diffs.back().mode == diff.mode
and (diff.mode != Diff::Add or and (diff.mode != Diff::Add or
diffs.back().posB + diffs.back().len == diff.posB)) diffs.back().posB + diffs.back().len == diff.posB))
@ -109,7 +112,21 @@ void find_diff_rec(Iterator a, int offA, int lenA,
Iterator b, int offB, int lenB, Iterator b, int offB, int lenB,
int* V1, int* V2, Equal eq, Vector<Diff>& diffs) int* V1, int* V2, Equal eq, Vector<Diff>& diffs)
{ {
if (lenA > 0 and lenB > 0) int prefix_len = 0;
while (lenA > 0 and lenB > 0 and a[offA] == b[offB])
++offA, ++offB, --lenA, --lenB, ++prefix_len;
int suffix_len = 0;
while (lenA != 0 and lenB != 0 and a[offA + lenA - 1] == b[offB + lenB - 1])
--lenA, --lenB, ++suffix_len;
append_diff(diffs, {Diff::Keep, prefix_len, 0});
if (lenA == 0)
append_diff(diffs, {Diff::Add, lenB, offB});
else if (lenB == 0)
append_diff(diffs, {Diff::Remove, lenA, 0});
else
{ {
auto middle_snake = find_middle_snake(a + offA, lenA, b + offB, lenB, V1, V2, eq); auto middle_snake = find_middle_snake(a + offA, lenA, b + offB, lenB, V1, V2, eq);
kak_assert(middle_snake.u <= lenA and middle_snake.v <= lenB); kak_assert(middle_snake.u <= lenA and middle_snake.v <= lenB);
@ -119,8 +136,7 @@ void find_diff_rec(Iterator a, int offA, int lenA,
b, offB, middle_snake.y, b, offB, middle_snake.y,
V1, V2, eq, diffs); V1, V2, eq, diffs);
if (int len = middle_snake.u - middle_snake.x) append_diff(diffs, {Diff::Keep, middle_snake.u - middle_snake.x, 0});
append_diff(diffs, {Diff::Keep, len, 0});
find_diff_rec(a, offA + middle_snake.u, lenA - middle_snake.u, find_diff_rec(a, offA + middle_snake.u, lenA - middle_snake.u,
b, offB + middle_snake.v, lenB - middle_snake.v, b, offB + middle_snake.v, lenB - middle_snake.v,
@ -131,7 +147,6 @@ void find_diff_rec(Iterator a, int offA, int lenA,
if (middle_snake.d == 1) if (middle_snake.d == 1)
{ {
const int diag = middle_snake.x - (middle_snake.add ? 0 : 1); const int diag = middle_snake.x - (middle_snake.add ? 0 : 1);
if (diag != 0)
append_diff(diffs, {Diff::Keep, diag, 0}); append_diff(diffs, {Diff::Keep, diag, 0});
if (middle_snake.add) if (middle_snake.add)
@ -139,14 +154,11 @@ void find_diff_rec(Iterator a, int offA, int lenA,
else else
append_diff(diffs, {Diff::Remove, 1, 0}); append_diff(diffs, {Diff::Remove, 1, 0});
} }
if (int len = middle_snake.u - middle_snake.x) append_diff(diffs, {Diff::Keep, middle_snake.u - middle_snake.x, 0});
append_diff(diffs, {Diff::Keep, len, 0});
} }
} }
else if (lenB > 0)
append_diff(diffs, {Diff::Add, lenB, offB}); append_diff(diffs, {Diff::Keep, suffix_len, 0});
else if (lenA > 0)
append_diff(diffs, {Diff::Remove, lenA, 0});
} }
template<typename Iterator, typename Equal = std::equal_to<typename std::iterator_traits<Iterator>::value_type>> template<typename Iterator, typename Equal = std::equal_to<typename std::iterator_traits<Iterator>::value_type>>