List:Commits« Previous MessageNext Message »
From:Martin Skold Date:October 16 2006 8:42am
Subject:bk commit into 5.1 tree (mskold:1.2312)
View as plain text  
Below is the list of changes that have just been committed into a local
5.1 repository of marty. When marty does a push these changes will
be propagated to the main repository and, within 24 hours after the
push, to the public repository.
For information on how to access the public repository
see http://dev.mysql.com/doc/mysql/en/installing-source-tree.html

ChangeSet@stripped, 2006-10-16 10:42:04+02:00, mskold@stripped +5 -0
  Merge mysql.com:/windows/Linux_space/MySQL/mysql-5.0
  into  mysql.com:/windows/Linux_space/MySQL/mysql-5.1
  MERGE: 1.1810.2078.34

  mysql-test/r/ndb_charset.result@stripped, 2006-10-16 10:36:49+02:00, mskold@stripped +0 -2
    Usin g local, will re-generate.
    MERGE: 1.5.1.3

  mysql-test/r/ndb_index_unique.result@stripped, 2006-10-16 10:37:38+02:00, mskold@stripped +0 -5
    Using local, will re-generate
    MERGE: 1.16.1.2

  mysql-test/r/ndb_update.result@stripped, 2006-10-16 10:38:27+02:00, mskold@stripped +0 -1
    Using local, will re-generate
    MERGE: 1.5.1.5

  sql/ha_ndbcluster.cc@stripped, 2006-10-16 10:41:58+02:00, mskold@stripped +1 -2
    Merge
    MERGE: 1.175.53.6

  sql/handler.cc@stripped, 2006-10-16 10:41:59+02:00, mskold@stripped +0 -15
    Merge
    MERGE: 1.167.1.54

# This is a BitKeeper patch.  What follows are the unified diffs for the
# set of deltas contained in the patch.  The rest of the patch, the part
# that BitKeeper cares about, is below these diffs.
# User:	mskold
# Host:	linux.site
# Root:	/windows/Linux_space/MySQL/mysql-5.1/RESYNC

--- 1.358/sql/ha_ndbcluster.cc	2006-10-16 10:42:14 +02:00
+++ 1.359/sql/ha_ndbcluster.cc	2006-10-16 10:42:14 +02:00
@@ -561,7 +561,14 @@ int ha_ndbcluster::ndb_err(NdbTransactio
   if (res == HA_ERR_FOUND_DUPP_KEY)
   {
     if (m_rows_to_insert == 1)
-      m_dupkey= table_share->primary_key;
+    {
+      /*
+	We can only distinguish between primary and non-primary
+	violations here, so we need to return MAX_KEY for non-primary
+	to signal that key is unknown
+      */
+      m_dupkey= err.code == 630 ? table_share->primary_key : MAX_KEY; 
+    }
     else
     {
       /* We are batching inserts, offending key is not available */
Thread
bk commit into 5.1 tree (mskold:1.2312)Martin Skold16 Oct