Fix query string truncation while dispatching to QE

Execution of a long enough query containing multi-byte characters can cause incorrect truncation of the query string. Incorrect truncation implies an occasional cut of a multi-byte character and (with log_min_duration_statement set to 0 ) subsequent write of an invalid symbol to segment logs. Such broken character present in logs produces problems when trying to fetch logs info from gp_toolkit.__gp_log_segment_ext  table - queries fail with the following error: «ERROR: invalid byte sequence for encoding…». 
This is caused by buildGpQueryString function in `cdbdisp_query.c`, which prepares query text for dispatch to QE. It does not take into account character length when truncation is necessary (text is longer than QUERY_STRING_TRUNCATE_SIZE). 

src/backend/cdb/dispatcher/cdbdisp_query.c

@@ -38,6 +38,7 @@
 #include "utils/session_state.h"
 #include "utils/typcache.h"
 #include "miscadmin.h"
+#include "mb/pg_wchar.h"
 
 #include "cdb/cdbdisp.h"
 #include "cdb/cdbdisp_query.h"
@@ -871,6 +872,8 @@ buildGpQueryString(DispatchCommandQueryParms *pQueryParms,
 	int			total_query_len;
 	char	   *shared_query,
 			   *pos;
+	int			cnt,
+				character_len;
 	MemoryContext oldContext;
 
 	/*
@@ -886,8 +889,16 @@ buildGpQueryString(DispatchCommandQueryParms *pQueryParms,
 	 * Here we only need to determine the truncated size, the actual work is
 	 * done later when copying it to the result buffer.
 	 */
+	cnt = 0;
 	if (querytree || plantree)
-		command_len = strnlen(command, QUERY_STRING_TRUNCATE_SIZE - 1) + 1;
+	{
+		while (cnt < QUERY_STRING_TRUNCATE_SIZE)
+		{
+			character_len = pg_encoding_mblen(GetDatabaseEncoding(), command + cnt);
+			cnt += character_len;
+		}
+		command_len = strnlen(command, cnt - character_len) + 1;
+	}
 	else
 		command_len = strlen(command) + 1;
 

src/test/regress/expected/dispatch_encoding.out

 -- More tests related to dispatching and QD->QE communication.
 --
+-- Test buildGpQueryString of cdbdisp_query.c truncates a query longer than QUERY_STRING_TRUNCATE_SIZE and containing
+-- multi-byte symbols properly
+--
+set log_min_duration_statement to 0;
+create table truncate_test ("колонка 1" int, "колонка 2" int, "колонка 3" int, "колонка 4" int, "колонка 5" int,
+"колонка 6" int, "колонка 7" int, "колонка 8" int, "колонка 9" int, "колонка 10" int, "колонка 11" int,
+"колонка 12" int, "колонка 13" int, "колонка 14" int, "колонка 15" int, "колонка 16" int, "колонка 17" int,
+"колонка 18" int, "колонка 19" int, "колонка 20" int, "колонка 21" int, "колонка 22" int, "колонка 23" int,
+"колонка 24" int, "колонка 25" int, "колонка 26" int, "колонка 27" int, "колонка 28" int, "колонка 29" int,
+"колонка 30" int, "колонка 31" int, "колонка 32" int, "колонка 33" int, "колонка 34" int, "колонка 35" int,
+"колонка 36" int, "колонка 37" int, "колонка 38" int, "колонка 39" int, "колонка 40" int, "особая колонка" int);
+select logdebug from gp_toolkit.__gp_log_segment_ext where logdebug ilike
+'%create table truncate_test%' and logdebug not ilike '%gp_toolkit.__gp_log_segment_ext%' order by logtime desc limit 1;
+                                                     logdebug                                                     
+------------------------------------------------------------------------------------------------------------------
+ create table truncate_test ("колонка 1" int, "колонка 2" int, "колонка 3" int, "колонка 4" int, "колонка 5" int,+
+ "колонка 6" int, "колонка 7" int, "колонка 8" int, "колонка 9" int, "колонка 10" int, "колонка 11" int,         +
+ "колонка 12" int, "колонка 13" int, "колонка 14" int, "колонка 15" int, "колонка 16" int, "колонка 17" int,     +
+ "колонка 18" int, "колонка 19" int, "колонка 20" int, "колонка 21" int, "колонка 22" int, "колонка 23" int,     +
+ "колонка 24" int, "колонка 25" int, "колонка 26" int, "колонка 27" int, "колонка 28" int, "колонка 29" int,     +
+ "колонка 30" int, "колонка 31" int, "колонка 32" int, "колонка 33" int, "колонка 34" int, "колонка 35" int,     +
+ "колонка 36" int, "колонка 37" int, "колонка 38" int, "колонка 39" int, "колонка 40" int, "о
+(1 row)
+
+drop table truncate_test;
+reset log_min_duration_statement;
+--
 -- Test that error messages come out correctly, with non-default
 -- client_encoding. (This test assumes that the regression database does
 -- *not* use latin1, otherwise this doesn't test anything interesting.)

src/test/regress/sql/dispatch_encoding.sql

 -- More tests related to dispatching and QD->QE communication.
 
+--
+-- Test buildGpQueryString of cdbdisp_query.c truncates a query longer than QUERY_STRING_TRUNCATE_SIZE and containing
+-- multi-byte symbols properly
+--
+set log_min_duration_statement to 0;
+create table truncate_test ("колонка 1" int, "колонка 2" int, "колонка 3" int, "колонка 4" int, "колонка 5" int,
+"колонка 6" int, "колонка 7" int, "колонка 8" int, "колонка 9" int, "колонка 10" int, "колонка 11" int,
+"колонка 12" int, "колонка 13" int, "колонка 14" int, "колонка 15" int, "колонка 16" int, "колонка 17" int,
+"колонка 18" int, "колонка 19" int, "колонка 20" int, "колонка 21" int, "колонка 22" int, "колонка 23" int,
+"колонка 24" int, "колонка 25" int, "колонка 26" int, "колонка 27" int, "колонка 28" int, "колонка 29" int,
+"колонка 30" int, "колонка 31" int, "колонка 32" int, "колонка 33" int, "колонка 34" int, "колонка 35" int,
+"колонка 36" int, "колонка 37" int, "колонка 38" int, "колонка 39" int, "колонка 40" int, "особая колонка" int);
+select logdebug from gp_toolkit.__gp_log_segment_ext where logdebug ilike
+'%create table truncate_test%' and logdebug not ilike '%gp_toolkit.__gp_log_segment_ext%' order by logtime desc limit 1;
+drop table truncate_test;
+reset log_min_duration_statement;
+
 --
 -- Test that error messages come out correctly, with non-default
 -- client_encoding. (This test assumes that the regression database does