Merge branch 'master' into no-bad-tests

Author: alexey-milovidov

ci/jobs/scripts/check_style/aspell-ignore/en/aspell-dict.txt

@@ -1961,6 +1961,7 @@ grpcio
 gtest
 gtid
 gzip
+gunzip
 gzipped
 hadoop
 halfMD
@@ -2345,6 +2346,8 @@ namespace
 namespaces
 natively
 nats
+ncat
+netcat
 ness
 nestjs
 netloc

ci/jobs/scripts/check_style/check_cpp.sh

@@ -327,8 +327,8 @@ find $ROOT_PATH/{src,base,programs,utils} -name '*.h' -or -name '*.cpp' |
     echo "If an exception has LOGICAL_ERROR code, there is no need to include the text 'Logical error' in the exception message, because then the phrase 'Logical error' will be printed twice."
 
 PATTERN="allow_";
-DIFF=$(comm -3 <(grep -o "\b$PATTERN\w*\b" $ROOT_PATH/src/Core/Settings.cpp | sort -u) <(grep -o -h "\b$PATTERN\w*\b" $ROOT_PATH/src/Databases/enableAllExperimentalSettings.cpp $ROOT_PATH/utils/check-style/experimental_settings_ignore.txt | sort -u));
-[ -n "$DIFF" ] && echo "$DIFF" && echo "^^ Detected 'allow_*' settings that might need to be included in src/Databases/enableAllExperimentalSettings.cpp" && echo "Alternatively, consider adding an exception to utils/check-style/experimental_settings_ignore.txt"
+DIFF=$(comm -3 <(grep -o "\b$PATTERN\w*\b" $ROOT_PATH/src/Core/Settings.cpp | sort -u) <(grep -o -h "\b$PATTERN\w*\b" $ROOT_PATH/src/Databases/enableAllExperimentalSettings.cpp $ROOT_PATH/ci/jobs/scripts/check_style/experimental_settings_ignore.txt | sort -u));
+[ -n "$DIFF" ] && echo "$DIFF" && echo "^^ Detected 'allow_*' settings that might need to be included in src/Databases/enableAllExperimentalSettings.cpp" && echo "Alternatively, consider adding an exception to ci/jobs/scripts/check_style/experimental_settings_ignore.txt"
 
 # Don't allow the direct inclusion of magic_enum.hpp and instead point to base/EnumReflection.h
 find $ROOT_PATH/{src,base,programs,utils} -name '*.cpp' -or -name '*.h' | xargs grep -l "magic_enum.hpp" | grep -v EnumReflection.h | while read -r line;

ci/jobs/scripts/check_style/check_typos.sh

@@ -7,8 +7,8 @@ ROOT_PATH="."
 #FIXME: check all (or almost all) repo
 codespell \
     --skip "*generated*,*gperf*,*.bin,*.mrk*,*.idx,checksums.txt,*.dat,*.pyc,*.kate-swp,*obfuscateQueries.cpp,d3-*.js,*.min.js,*.sum,${ROOT_PATH}/utils/check-style/aspell-ignore" \
-    --ignore-words "${ROOT_PATH}/utils/check-style/codespell-ignore-words.list" \
-    --exclude-file "${ROOT_PATH}/utils/check-style/codespell-ignore-lines.list" \
+    --ignore-words "${ROOT_PATH}/ci/jobs/scripts/check_style/codespell-ignore-words.list" \
+    --exclude-file "${ROOT_PATH}/ci/jobs/scripts/check_style/codespell-ignore-lines.list" \
     --quiet-level 2 \
     "$ROOT_PATH"/{src,base,programs,utils} \
     $@ | grep -P '.' \

utils/check-style/codespell-ignore-lines.list renamed to ci/jobs/scripts/check_style/codespell-ignore-lines.list

@@ -224,7 +224,7 @@ endif()
 
 if (CHECK_LARGE_OBJECT_SIZES)
     add_custom_command(TARGET clickhouse POST_BUILD
-        COMMAND "${CMAKE_SOURCE_DIR}/utils/check-style/check-large-objects.sh" "${CMAKE_BINARY_DIR}")
+        COMMAND "${CMAKE_SOURCE_DIR}/utils/check-large-objects.sh" "${CMAKE_BINARY_DIR}")
 endif ()
 
 if (SPLIT_DEBUG_SYMBOLS)

utils/check-style/codespell-ignore-words.list renamed to ci/jobs/scripts/check_style/codespell-ignore-words.list

@@ -771,7 +771,13 @@ ZooKeeper::multiImpl(const Coordination::Requests & requests, Coordination::Resp
 
     if (future_result.wait_for(std::chrono::milliseconds(args.operation_timeout_ms)) != std::future_status::ready)
     {
-        impl->finalize(fmt::format("Operation timeout on {} {}", Coordination::OpNum::Multi, requests[0]->getPath()));
+        auto & request = *requests[0];
+        impl->finalize(fmt::format(
+            "Operation timeout on {} of {} requests. First ({}): {}",
+            Coordination::OpNum::Multi,
+            requests.size(),
+            demangle(typeid(request).name()),
+            request.getPath()));
         return {Coordination::Error::ZOPERATIONTIMEOUT, ""};
     }
 

utils/check-style/experimental_settings_ignore.txt renamed to ci/jobs/scripts/check_style/experimental_settings_ignore.txt

@@ -15,12 +15,27 @@ namespace ErrorCodes
     extern const int BAD_ARGUMENTS;
 }
 
-static const re2::RE2 range_regex(R"({([\d]+\.\.[\d]+)})"); /// regexp for {M..N}, where M and N - non-negative integers
-static const re2::RE2 enum_regex(R"({([^{}*,]+[^{}*]*[^{}*,])})"); /// regexp for {expr1,expr2,expr3}, expr's should be without "{", "}", "*" and ","
+namespace
+{
+struct Regexps
+{
+    static const Regexps & instance()
+    {
+        static Regexps regexps;
+        return regexps;
+    }
+
+    /// regexp for {M..N}, where M and N - non-negative integers
+    re2::RE2 range_regex{R"({([\d]+\.\.[\d]+)})"};
+
+    /// regexp for {expr1,expr2,expr3}, expr's should be without "{", "}", "*" and ","
+    re2::RE2 enum_regex{R"({([^{}*,]+[^{}*]*[^{}*,])})"};
+};
+}
 
 bool containsRangeGlob(const std::string & input)
 {
-    return RE2::PartialMatch(input, range_regex);
+    return RE2::PartialMatch(input, Regexps::instance().range_regex);
 }
 
 bool containsOnlyEnumGlobs(const std::string & input)
@@ -67,8 +82,8 @@ std::string makeRegexpPatternFromGlobs(const std::string & initial_str_with_glob
         std::string_view matched_range;
         std::string_view matched_enum;
 
-        auto did_match_range = RE2::PartialMatch(input, range_regex, &matched_range);
-        auto did_match_enum = RE2::PartialMatch(input, enum_regex, &matched_enum);
+        auto did_match_range = RE2::PartialMatch(input, Regexps::instance().range_regex, &matched_range);
+        auto did_match_enum = RE2::PartialMatch(input, Regexps::instance().enum_regex, &matched_enum);
 
         /// Enum regex matches ranges, so if they both match and point to the same data,
         /// it is a range.
@@ -78,7 +93,7 @@ std::string makeRegexpPatternFromGlobs(const std::string & initial_str_with_glob
         /// We matched a range, and range comes earlier than enum
         if (did_match_range && (!did_match_enum || matched_range.data() < matched_enum.data()))
         {
-            RE2::FindAndConsume(&input, range_regex, &matched);
+            RE2::FindAndConsume(&input, Regexps::instance().range_regex, &matched);
             std::string buffer(matched);
             oss_for_replacing << escaped_with_globs.substr(current_index, matched_range.data() - escaped_with_globs.data() - current_index - 1) << '(';
 
@@ -122,7 +137,7 @@ std::string makeRegexpPatternFromGlobs(const std::string & initial_str_with_glob
         /// We matched enum, and it comes earlier than range.
         else if (did_match_enum && (!did_match_range || matched_enum.data() < matched_range.data()))
         {
-            RE2::FindAndConsume(&input, enum_regex, &matched);
+            RE2::FindAndConsume(&input, Regexps::instance().enum_regex, &matched);
             std::string buffer(matched);
 
             oss_for_replacing << escaped_with_globs.substr(current_index, matched.data() - escaped_with_globs.data() - current_index - 1) << '(';