thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-24 16:42:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

refine a little bit

Browse Source
This commit is contained in:
Han Fei 2023-03-14 18:15:42 +01:00
parent 01be209e43
commit 076d33bb03
2 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/Common/OptimizedRegularExpression.cpp
View File

@ -386,7 +386,7 @@ finish:
bool next_is_trivial = true;
pos = analyzeImpl(regexp, pos, required_substring, next_is_trivial, next_alternatives);
/// For xxx|xxx|xxx, we only conbine the alternatives and return a empty required_substring.
if (next_alternatives.empty())
if (next_alternatives.empty() || shortest_alter_length(next_alternatives) < required_substring.literal.size())
{
global_alters.push_back(required_substring);
}

6
src/Common/tests/gtest_optimize_re.cpp
View File

@ -37,8 +37,10 @@ TEST(OptimizeRE, analyze)
test_f(R"(abc(de|xyz|(\{xx\}))fg)", "abc", {"abcdefg", "abcxyzfg", "abc{xx}fg"});
test_f("abc(abc|fg)?xyzz", "xyzz");
test_f("abc(abc|fg){0,1}xyzz", "xyzz");
test_f("abc(abc|fg)xyzz|bcdd?k|bc(f|g|h?)z", "", {"abcabcxyzz", "abcfgxyzz", "bcd", "bcfz", "bcgz", ""});
test_f("abc(abc|fg)xyzz|bc(dd?x|kk?y|(f))k|bc(f|g|h?)z", "", {"abcabcxyzz", "abcfgxyzz", "bcd", "bck", "bcfk", "bcfz", "bcgz", ""});
test_f("abc(abc|fg)xyzz|bcdd?k|bc(f|g|h?)z", "", {"abcabcxyzz", "abcfgxyzz", "bcd", "bc"});
test_f("abc(abc|fg)xyzz|bc(dd?x|kk?y|(f))k|bc(f|g|h?)z", "", {"abcabcxyzz", "abcfgxyzz", "bcd", "bck", "bcfk", "bc"});
test_f("((?:abc|efg|xyz)/[a-zA-Z0-9]{1-50})(/?[^ ]*|)", "", {"abc/", "efg/", "xyz/"});
test_f(R"([Bb]ai[Dd]u[Ss]pider(?:-[A-Za-z]{1,30})(?:-[A-Za-z]{1,30}|)|bingbot|\bYeti(?:-[a-z]{1,30}|)|Catchpoint(?: bot|)|[Cc]harlotte|Daumoa(?:-feedfetcher|)|(?:[a-zA-Z]{1,30}-|)Googlebot(?:-[a-zA-Z]{1,30}|))", "", {"pider-", "bingbot", "Yeti-", "Yeti", "Catchpoint bot", "Catchpoint", "harlotte", "Daumoa-feedfetcher", "Daumoa", "Googlebot-", "Googlebot"});
test_f("abc|(:?xx|yy|zz|x?)def", "", {"abc", "def"});
test_f("abc|(:?xx|yy|zz|x?){1,2}def", "", {"abc", "def"});
}