Check for punctuation

src/Common/parseRemoteDescription.cpp

@@ -52,20 +52,8 @@ static bool parseNumber(const String & description, size_t l, size_t r, size_t &
 }
 
 
-/* Parse a string that generates shards and replicas. Separator - one of two characters | or ,
- *  depending on whether shards or replicas are generated.
- * For example:
- * host1,host2,...      - generates set of shards from host1, host2, ...
- * host1|host2|...      - generates set of replicas from host1, host2, ...
- * abc{8..10}def        - generates set of shards abc8def, abc9def, abc10def.
- * abc{08..10}def       - generates set of shards abc08def, abc09def, abc10def.
- * abc{x,yy,z}def       - generates set of shards abcxdef, abcyydef, abczdef.
- * abc{x|yy|z} def      - generates set of replicas abcxdef, abcyydef, abczdef.
- * abc{1..9}de{f,g,h}   - is a direct product, 27 shards.
- * abc{1..9}de{0|1}     - is a direct product, 9 shards, in each 2 replicas.
- */
-std::vector<String>
-parseRemoteDescription(const String & description, size_t l, size_t r, char separator, size_t max_addresses, const String & func_name)
+std::vector<String> parseRemoteDescription(
+    const String & description, size_t l, size_t r, char separator, size_t max_addresses, const String & func_name)
 {
     std::vector<String> res;
     std::vector<String> cur;

src/Common/parseRemoteDescription.h

@@ -3,7 +3,7 @@
 #include <vector>
 namespace DB
 {
-/* Parse a string that generates shards and replicas. Separator - one of two characters | or ,
+/* Parse a string that generates shards and replicas. Separator - one of two characters '|' or ','
  *  depending on whether shards or replicas are generated.
  * For example:
  * host1,host2,...      - generates set of shards from host1, host2, ...

src/Core/tests/gtest_settings.cpp

@@ -121,7 +121,7 @@ GTEST_TEST(SettingMySQLDataTypesSupport, SetString)
     ASSERT_EQ(Field("decimal,datetime64"), setting);
 
     // comma with spaces
-    setting = " datetime64 ,    decimal ";
+    setting = " datetime64 ,    decimal "; /// bad punctuation is ok here
     ASSERT_TRUE(setting.changed);
     ASSERT_TRUE(setting.value.isSet(MySQLDataTypesSupport::DECIMAL));
     ASSERT_TRUE(setting.value.isSet(MySQLDataTypesSupport::DATETIME64));
@@ -166,4 +166,3 @@ GTEST_TEST(SettingMySQLDataTypesSupport, SetInvalidString)
     ASSERT_TRUE(setting.changed);
     ASSERT_EQ(0, setting.value.getValue());
 }
-

src/Functions/FunctionsStringHash.cpp

@@ -293,7 +293,7 @@ struct SimHashImpl
 
             // we need to store the new word hash value to the oldest location.
             // for example, N = 5, array |a0|a1|a2|a3|a4|, now, a0 is the oldest location,
-            // so we need to store new word hash into location of a0, then ,this array become
+            // so we need to store new word hash into location of a0, then this array become
             // |a5|a1|a2|a3|a4|, next time, a1 become the oldest location, we need to store new
             // word hash value into location of a1, then array become |a5|a6|a2|a3|a4|
             words[offset] = BytesRef{word_start, length};
@@ -793,4 +793,3 @@ REGISTER_FUNCTION(StringHash)
     factory.registerFunction<FunctionWordShingleMinHashArgCaseInsensitiveUTF8>();
 }
 }
-

src/Interpreters/TransactionLog.cpp

@@ -482,7 +482,7 @@ CSN TransactionLog::finalizeCommittedTransaction(MergeTreeTransaction * txn, CSN
         bool removed = running_list.erase(txn->tid.getHash());
         if (!removed)
         {
-            LOG_ERROR(log , "I's a bug: TID {} {} doesn't exist", txn->tid.getHash(), txn->tid);
+            LOG_ERROR(log, "It's a bug: TID {} {} doesn't exist", txn->tid.getHash(), txn->tid);
             abort();
         }
     }

src/Parsers/Kusto/ParserKQLOperators.h

@@ -31,10 +31,10 @@ protected:
         not_endswith,
         endswith_cs,
         not_endswith_cs,
-        equal, //=~
-        not_equal,//!~
-        equal_cs, //=
-        not_equal_cs,//!=
+        equal, /// =~
+        not_equal, /// !~
+        equal_cs, /// =
+        not_equal_cs, /// !=
         has,
         not_has,
         has_all,
@@ -49,10 +49,10 @@ protected:
         not_hassuffix,
         hassuffix_cs,
         not_hassuffix_cs,
-        in_cs,  //in
-        not_in_cs, //!in
-        in, //in~
-        not_in ,//!in~
+        in_cs, /// in
+        not_in_cs, /// !in
+        in, /// in~
+        not_in, /// !in~
         matches_regex,
         startswith,
         not_startswith,

src/Processors/Formats/Impl/JSONEachRowRowInputFormat.cpp

@@ -236,10 +236,10 @@ bool JSONEachRowRowInputFormat::readRow(MutableColumns & columns, RowReadExtensi
 
 bool JSONEachRowRowInputFormat::checkEndOfData(bool is_first_row)
 {
-    /// We consume , or \n before scanning a new row, instead scanning to next row at the end.
+    /// We consume ',' or '\n' before scanning a new row, instead scanning to next row at the end.
     /// The reason is that if we want an exact number of rows read with LIMIT x
     /// from a streaming table engine with text data format, like File or Kafka
-    /// then seeking to next ;, or \n would trigger reading of an extra row at the end.
+    /// then seeking to next ';,' or '\n' would trigger reading of an extra row at the end.
 
     /// Semicolon is added for convenience as it could be used at end of INSERT query.
     if (!in->eof())

utils/check-style/check-style

@@ -410,3 +410,6 @@ find $ROOT_PATH/{src,programs,utils} -name '*.h' -or -name '*.cpp' | xargs grep
 
 # The stateful directory should only contain the tests that depend on the test dataset (hits or visits).
 find $ROOT_PATH/tests/queries/1_stateful -name '*.sql' -or -name '*.sh' | grep -v '00076_system_columns_bytes' | xargs -I{} bash -c 'grep -q -P "hits|visits" "{}" || echo "The test {} does not depend on the test dataset (hits or visits table) and should be located in the 0_stateless directory. You can also add an exception to the check-style script."'
+
+# Check for bad punctuation: whitespace before comma.
+find $ROOT_PATH/{src,programs,utils} -name '*.h' -or -name '*.cpp' | xargs grep -P --line-number '\w ,' | grep -v 'bad punctuation is ok here' && echo "^ There is bad punctuation: whitespace before comma. You should write it like this: 'Hello, world!'"