From e3cd20538887c037faea1f93cdda32763f7fc7fb Mon Sep 17 00:00:00 2001
From: Andrew Hutchings <andrew@linuxjedi.co.uk>
Date: Tue, 11 Jun 2019 10:37:04 +0100
Subject: [PATCH] MCOL-1968 Fix UTF char/varchar min/max handling

If the first byte of a char/varchar was > 0x80 then it will break the
min/max values for an extent during cpimport. This patch makes the
min/max compare unsigned and only switches to signed when storing.

In addition send all the LDI / INSERT...SELECT data to cpimport, not
truncated. Let cpimport figure out the truncation point.
---
 dbcon/mysql/ha_calpont_dml.cpp         | 18 ++++++++++--------
 writeengine/bulk/we_bulkloadbuffer.cpp | 11 +++++++----
 2 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/dbcon/mysql/ha_calpont_dml.cpp b/dbcon/mysql/ha_calpont_dml.cpp
index c2dc36e47..a9404c223 100755
--- a/dbcon/mysql/ha_calpont_dml.cpp
+++ b/dbcon/mysql/ha_calpont_dml.cpp
@@ -860,7 +860,13 @@ int ha_calpont_impl_write_batch_row_(uchar *buf, TABLE* table, cal_impl_if::cal_
 				}
 				case CalpontSystemCatalog::VARCHAR:
 				{
-					if (nullVal && (ci.columnTypes[colpos].constraintType != CalpontSystemCatalog::NOTNULL_CONSTRAINT))
+                    size_t length;
+                    if (ci.utf8)
+                        length = (ci.columnTypes[colpos].colWidth * 3);
+                    else
+                        length = ci.columnTypes[colpos].colWidth;
+
+                    if (nullVal && (ci.columnTypes[colpos].constraintType != CalpontSystemCatalog::NOTNULL_CONSTRAINT))
 					{
 						fprintf(ci.filePtr, "%c", ci.delimiter);
 						if (!ci.utf8)
@@ -901,6 +907,7 @@ int ha_calpont_impl_write_batch_row_(uchar *buf, TABLE* table, cal_impl_if::cal_
 								dataLength = *(uint16_t*) buf;
 								buf = buf + 2 ;
 							}
+                            length = dataLength;
                             escape.assign((char*)buf, dataLength);
                             boost::replace_all(escape, "\\", "\\\\");
 							fprintf(ci.filePtr, "%c%.*s%c%c", ci.enclosed_by, (int)escape.length(), escape.c_str(), ci.enclosed_by, ci.delimiter); 							
@@ -917,8 +924,7 @@ int ha_calpont_impl_write_batch_row_(uchar *buf, TABLE* table, cal_impl_if::cal_
 								dataLength = *(uint16_t*) buf;
 								buf = buf + 2 ;
 							}
-							if ( dataLength > ci.columnTypes[colpos].colWidth)
-								dataLength = ci.columnTypes[colpos].colWidth;
+                            length = dataLength;
 
                             escape.assign((char*)buf, dataLength);
                             boost::replace_all(escape, "\\", "\\\\");
@@ -926,11 +932,7 @@ int ha_calpont_impl_write_batch_row_(uchar *buf, TABLE* table, cal_impl_if::cal_
 							fprintf(ci.filePtr, "%c%.*s%c%c", ci.enclosed_by, (int)escape.length(), escape.c_str(), ci.enclosed_by, ci.delimiter); 
 						}
 					}
-					//buf += ci.columnTypes[colpos].colWidth;
-					if (ci.utf8)
-						buf += (ci.columnTypes[colpos].colWidth * 3);
-					else
-						buf += ci.columnTypes[colpos].colWidth;
+					buf += length;
 					
 					break;
 				}
diff --git a/writeengine/bulk/we_bulkloadbuffer.cpp b/writeengine/bulk/we_bulkloadbuffer.cpp
index b1902e98f..7326f987b 100644
--- a/writeengine/bulk/we_bulkloadbuffer.cpp
+++ b/writeengine/bulk/we_bulkloadbuffer.cpp
@@ -524,13 +524,16 @@ void BulkLoadBuffer::convert(char *field, int fieldLength,
             }
 
             // Swap byte order before comparing character string
-            int64_t binChar = static_cast<int64_t>( uint64ToStr(
-                    *(reinterpret_cast<uint64_t*>(charTmpBuf)) ) );
+            // Compare must be unsigned
+            uint64_t compChar = uint64ToStr( *(reinterpret_cast<uint64_t*>(charTmpBuf)) );
+            int64_t binChar = static_cast<int64_t>( compChar );
 
             // Update min/max range
-            if (binChar < bufStats.minBufferVal)
+            uint64_t minVal = static_cast<uint64_t>( bufStats.minBufferVal );
+            uint64_t maxVal = static_cast<uint64_t>( bufStats.maxBufferVal );
+            if (compChar < minVal)
                 bufStats.minBufferVal = binChar;
-            if (binChar > bufStats.maxBufferVal)
+            if (compChar > maxVal)
                 bufStats.maxBufferVal = binChar;
 
             pVal = charTmpBuf;