Bug#20683959 LOAD DATA INFILE IGNORES A SPECIFIC ROW SILENTLY UNDER D…

…B CHARSET IS UTF8

READ_INFO::read_field has logic for reallocating the internal buffer
if it is too small. This did not work correctly for multibyte charsets.

Fix: If we are out of space, then push the current character on the
input stack, and break out of while loop, to reach realloc code.

Also: Ensure that we report an error before returning an error status.

mysql-test/r/loaddata.result

@@ -564,3 +564,30 @@ SET @@sql_mode= @old_mode;
 Warnings:
 Warning	3090	Changing sql mode 'NO_AUTO_CREATE_USER' is deprecated. It will be removed in a future release.
 DROP TABLE t1;
+#
+# Bug#20683959 LOAD DATA INFILE IGNORES A SPECIFIC ROW SILENTLY
+#              UNDER DB CHARSET IS UTF8
+#
+CREATE DATABASE d1 CHARSET latin1;
+USE d1;
+CREATE TABLE t1 (val TEXT);
+LOAD DATA INFILE '../../std_data/bug20683959loaddata.txt' INTO TABLE t1;
+SELECT COUNT(*) FROM t1;
+COUNT(*)
+1
+SELECT HEX(val) FROM t1;
+HEX(val)
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
+CREATE DATABASE d2 CHARSET utf8;
+USE d2;
+CREATE TABLE t1 (val TEXT);
+LOAD DATA INFILE '../../std_data/bug20683959loaddata.txt' INTO TABLE t1;
+SELECT COUNT(*) FROM t1;
+COUNT(*)
+1
+SELECT HEX(val) FROM t1;
+HEX(val)
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
+DROP TABLE d1.t1, d2.t1;
+DROP DATABASE d1;
+DROP DATABASE d2;

mysql-test/std_data/bug20683959loaddata.txt

@@ -0,0 +1 @@
+"RT @niouzechun: 私はハッピーエンド厨なんだけれど、アンハッピーエンドや惨劇的なお話をあまり積極的に摂取しない理由は、現実の人生は基本的にうまくいかないし不平等だし理不尽だし辛い

mysql-test/t/loaddata.test

@@ -670,3 +670,27 @@ eval LOAD DATA LOCAL INFILE '$MYSQLTEST_VARDIR/mysql' REPLACE INTO TABLE t1
 
 SET @@sql_mode= @old_mode;
 DROP TABLE t1;
+
+--echo #
+--echo # Bug#20683959 LOAD DATA INFILE IGNORES A SPECIFIC ROW SILENTLY
+--echo #              UNDER DB CHARSET IS UTF8
+--echo #
+
+CREATE DATABASE d1 CHARSET latin1;
+USE d1;
+CREATE TABLE t1 (val TEXT);
+LOAD DATA INFILE '../../std_data/bug20683959loaddata.txt' INTO TABLE t1;
+SELECT COUNT(*) FROM t1;
+SELECT HEX(val) FROM t1;
+
+CREATE DATABASE d2 CHARSET utf8;
+USE d2;
+CREATE TABLE t1 (val TEXT);
+LOAD DATA INFILE '../../std_data/bug20683959loaddata.txt' INTO TABLE t1;
+SELECT COUNT(*) FROM t1;
+SELECT HEX(val) FROM t1;
+
+
+DROP TABLE d1.t1, d2.t1;
+DROP DATABASE d1;
+DROP DATABASE d2;

sql/sql_load.cc

@@ -864,7 +864,7 @@ read_fixed_length(THD *thd, COPY_INFO &info, TABLE_LIST *table_list,
     */
     if (validate_default_values_of_unset_fields(thd, table))
     {
-      read_info.error= 1;
+      read_info.error= true;
       break;
     }
 
@@ -1021,7 +1021,7 @@ read_sep_field(THD *thd, COPY_INFO &info, TABLE_LIST *table_list,
     */
     if (validate_default_values_of_unset_fields(thd, table))
     {
-      read_info.error= 1;
+      read_info.error= true;
       break;
     }
 
@@ -1113,7 +1113,7 @@ read_sep_field(THD *thd, COPY_INFO &info, TABLE_LIST *table_list,
     }
 
     if (thd->is_error())
-      read_info.error= 1;
+      read_info.error= true;
 
     if (read_info.error)
       break;
@@ -1284,7 +1284,7 @@ read_xml_field(THD *thd, COPY_INFO &info, TABLE_LIST *table_list,
     */
     if (validate_default_values_of_unset_fields(thd, table))
     {
-      read_info.error= 1;
+      read_info.error= true;
       break;
     }
 
@@ -1494,8 +1494,8 @@ READ_INFO::READ_INFO(File file_par, uint tot_length, const CHARSET_INFO *cs,
   stack=stack_pos=(int*) sql_alloc(sizeof(int)*length);
 
   if (!(buffer=(uchar*) my_malloc(key_memory_READ_INFO,
-                                  buff_length+1,MYF(0))))
-    error=1; /* purecov: inspected */
+                                  buff_length+1, MYF(MY_WME))))
+    error= true; /* purecov: inspected */
   else
   {
     end_of_buff=buffer+buff_length;
@@ -1506,7 +1506,7 @@ READ_INFO::READ_INFO(File file_par, uint tot_length, const CHARSET_INFO *cs,
     {
       my_free(buffer); /* purecov: inspected */
       buffer= NULL;
-      error=1;
+      error= true;
     }
     else
     {
@@ -1712,7 +1712,10 @@ int READ_INFO::read_field()
       GET_MBCHARLEN(read_charset, chr, ml);
       if (ml == 0)
       {
-        error= 1;
+        *to= '\0';
+        my_error(ER_INVALID_CHARACTER_STRING, MYF(0),
+                 read_charset->csname, buffer);
+        error= true;
         return 1;
       }
 
@@ -1745,6 +1748,12 @@ int READ_INFO::read_field()
           PUSH(*--to);
         chr= GET;
       }
+      else if (ml > 1)
+      {
+        // Buffer is too small, exit while loop, and reallocate.
+        PUSH(chr);
+        break;
+      }
       *to++ = (uchar) chr;
     }
     /*
@@ -1753,7 +1762,7 @@ int READ_INFO::read_field()
     if (!(new_buffer=(uchar*) my_realloc(key_memory_READ_INFO,
                                          (char*) buffer,buff_length+1+IO_SIZE,
 					MYF(MY_WME))))
-      return (error=1);
+      return (error= true);
     to=new_buffer + (to-buffer);
     buffer=new_buffer;
     buff_length+=IO_SIZE;