Below is the list of changes that have just been committed into a local
5.0 repository of guilhem. When guilhem 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
1.1988 05/09/21 15:04:57 guilhem@stripped +1 -0
WL#1012: changing some TODO items to DONE :)
wl1012-review-pending-comments.txt
1.8 05/09/21 15:04:53 guilhem@stripped +10 -4
Some TODO becomes DONE :)
# 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: guilhem
# Host: gbichot3.local
# Root: /home/mysql_src/mysql-5.0-wl1012
--- 1.7/wl1012-review-pending-comments.txt 2005-09-20 23:12:15 +02:00
+++ 1.8/wl1012-review-pending-comments.txt 2005-09-21 15:04:53 +02:00
@@ -1881,8 +1881,7 @@
[G2DONE] on slave. Slave just stopped without saying why. So I created a bug
[G2DONE] report BUG#13042 and fixed it (added a message in the error log).
-[G2TODO] we could save thd in the event, like in Query_log_event, to
-[G2TODO] avoid current_thd.
+[G2DONE] As we save thd in the event, I removed current_thd.
> + /*
> + Now we have opened and locked the table. Now we can change the state of
> + this instance and update the values.
@@ -2270,6 +2269,7 @@
[G2TODO] Serg writes "why is that?"
+[G2TODO] Mats then explained and Serg asked that Mats puts a comment about it.
> + table->file->ha_rnd_init(1);
@@ -3559,7 +3559,10 @@
[G] I think that now you rely on Xid_log_event for committing, don't
[G] you?).
-[G2TODO]
+[G2DONE] The goal is to know when to call ha_autocommit_or_rollback to
+[G2DONE] commit, if this was an autocommit statement. If it was a
+[G2DONE] transaction, the Xid_log_event is what commits.
+
> #ifdef HAVE_EXPLICIT_TEMPLATE_INSTANTIATION
> template class I_List_iterator<i_string>;
@@ -4563,11 +4566,14 @@
[G] (you will find no Query_log_event or mysql_bin_log around this piece
[G] of code).
+[G2DONE] No problem in fact. check_table_binlog_row_based() returns 0
+[G2DONE] because the table has SYSTEM_TMP_TABLE.
+
[G] Generally by putting binlogging into ha_write_row(), we have chances
[G] to do excess-binlogging (like above), so need to check all ha_*_row()
[G] calls to be sure that we are not.
-[G2TODO]
+[G2DONE] I checked, it looks ok.
> --- 1.253/sql/sql_table.cc 2005-07-12 19:23:28 +02:00
> +++ 1.254/sql/sql_table.cc 2005-07-15 20:33:17 +02:00
| Thread |
|---|
| • bk commit into 5.0 tree (guilhem:1.1988) | guilhem | 21 Sep |