Skip to content

executor: Fix LIKE with non-ASCII patterns on unsigned-char platforms #11038

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ti-chi-bot merged 2 commits into from
Aug 14, 2026
+76 −16
Merged

executor: Fix LIKE with non-ASCII patterns on unsigned-char platforms #11038

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
50a5c20
executor: Fix LIKE with non-ASCII patterns on unsigned-char platforms
ChangRui-Ryan Aug 12, 2026
7948ea5
executor: Avoid LIKE performance regression on AArch64
ChangRui-Ryan Aug 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
49 changes: 33 additions & 16 deletions dbms/src/TiDB/Collation/Collator.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -134,44 +134,60 @@ template <typename Collator>
void Pattern<Collator>::tryCompileAsciiCi(const std::string & pattern, char escape)
{
is_ascii_ci_pattern = false;
const auto unsigned_escape = static_cast<unsigned char>(escape);
// Can't handle non-ASCII escape
if (escape < 0)
if (unsigned_escape >= weight_ascii_ci.size())
{
return;
}
ascii_ci_pattern.clear();
ascii_ci_pattern.reserve(pattern.length());

auto last_tp = MatchType::Match;
for (size_t i = 0; i < pattern.length(); i++)
{
auto c = pattern[i];
// Can't handle non-ASCII character
if (c < 0)
{
return;
}
auto c = static_cast<unsigned char>(pattern[i]);
MatchType tp;

if (c == escape)
if (c == unsigned_escape)
{
tp = MatchType::Match;
if (i < pattern.length() - 1)
{
// use next char to match
c = pattern[++i];
c = static_cast<unsigned char>(pattern[++i]);
}
else
{
// use `escape` to match
}
}
else if (c == '_')
{
tp = MatchType::One;
}
else if (c == '%')
{
if (i > 0 && pattern[i - 1] == '%')
// Only keep one '%' for continuous unescaped '%'s
if (last_tp == MatchType::Any)
{
continue;
}
tp = MatchType::Any;
}
else
{
tp = MatchType::Match;
}

// Can't handle non-ASCII characters
if (c >= weight_ascii_ci.size())
{
return;
}

ascii_ci_pattern.push_back(weight_ascii_ci[c]);
last_tp = tp;
}
is_ascii_ci_pattern = true;
}
Expand Down Expand Up @@ -300,12 +316,13 @@ int Pattern<Collator>::tryMatchAsciiCi(const char * s, size_t length) const
}
else
{
// Can't handle non-ASCII escape
if (s[str_idx] < 0)
// Can't handle non-ASCII characters
if (static_cast<unsigned char>(s[str_idx]) >= weight_ascii_ci.size())
{
return -1;
}
if ((match_types[p_idx] == Match && weight_ascii_ci[s[str_idx]] == ascii_ci_pattern[p_idx])
if ((match_types[p_idx] == Match
&& weight_ascii_ci[static_cast<unsigned char>(s[str_idx])] == ascii_ci_pattern[p_idx])
|| match_types[p_idx] == One)
{
p_idx++;
Expand Down Expand Up @@ -340,13 +357,13 @@ int Pattern<Collator>::tryMatchAsciiCi(const char * s, size_t length) const
{ // Fast forward to the first match position
while (str_idx < length)
{
// Can't handle non-ASCII escape
if (s[str_idx] < 0)
// Can't handle non-ASCII characters
if (static_cast<unsigned char>(s[str_idx]) >= weight_ascii_ci.size())
{
return -1;
}

if (weight_ascii_ci[s[str_idx]] != ascii_ci_pattern[p_idx_after_any])
if (weight_ascii_ci[static_cast<unsigned char>(s[str_idx])] != ascii_ci_pattern[p_idx_after_any])
{
str_idx = ++backtrack_idx;
}
Expand Down
43 changes: 43 additions & 0 deletions dbms/src/TiDB/tests/gtest_tidb_collator.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -164,6 +164,11 @@ const typename CollatorCases::PatternCase CollatorCases::pattern_cases[] = {
{"中国", {false, false, false, false, false, false, false}},
{"中国中", {true, true, true, true, true, true, true}},
{"", {false, false, false, false, false, false, false}}}},
{"%报告%",
{{"报告", {true, true, true, true, true, true, true}},
{"测试报告完成", {true, true, true, true, true, true, true}},
{"报表", {false, false, false, false, false, false, false}},
{"", {false, false, false, false, false, false, false}}}},
{"A%",
{{"a", {false, false, true, false, true, true, false}},
{"aaa", {false, false, true, false, true, true, false}},
Expand Down Expand Up @@ -275,6 +280,44 @@ const typename CollatorCases::PatternCase CollatorCases::pattern_cases[] = {
{"1234567910", {true, true, true, true, true, true, true}},
},
},
{
"a\\%%b", // escaped '%' followed by wildcard '%'
{
{"a%b", {true, true, true, true, true, true, true}},
{"a%foob", {true, true, true, true, true, true, true}},
{"a%x", {false, false, false, false, false, false, false}},
{"afoob", {false, false, false, false, false, false, false}},
},
},
{
"a%%b", // test consecutive '%' deduplication - two '%'s
{
{"ab", {true, true, true, true, true, true, true}},
{"axb", {true, true, true, true, true, true, true}},
{"a123b", {true, true, true, true, true, true, true}},
{"a", {false, false, false, false, false, false, false}},
},
},
{
"a%%%b", // test consecutive '%' deduplication - three '%'s
{
{"ab", {true, true, true, true, true, true, true}},
{"axb", {true, true, true, true, true, true, true}},
{"a456b", {true, true, true, true, true, true, true}},
{"", {false, false, false, false, false, false, false}},
},
},
{
"a\\%%%b", // escaped '%' followed by two wildcard '%'s (dedup to one)
{
{"a%b", {true, true, true, true, true, true, true}},
{"a%xb", {true, true, true, true, true, true, true}},
{"a%123b", {true, true, true, true, true, true, true}},
{"a%x", {false, false, false, false, false, false, false}},
{"a%123c", {false, false, false, false, false, false, false}},
{"ab", {false, false, false, false, false, false, false}},
},
},
{
"%__", // test match from end
{
Expand Down