Hi, Alexander! On Nov 28, Alexander Barkov wrote:
Hello Sergei,
On 11/28/2016 05:47 PM, Sergei Golubchik wrote:
Hi, Alexander!
diff --git a/sql/sql_load.cc b/sql/sql_load.cc index af4b251..6a473ce 100644 --- a/sql/sql_load.cc +++ b/sql/sql_load.cc @@ -79,6 +79,81 @@ class READ_INFO { NET *io_net; int level; /* for load xml */
+ +#if MYSQL_VERSION_ID >= 100200 +#error This 10.0 and 10.1 specific fix should be removed in 10.2. +#error Fix read_mbtail() to use my_charlen() instead of my_charlen_tmp()
I don't quite undrestand that. Should the compete patch be removed when merging with 10.2? Or only my_charlen_tmp be replaced with my_charlen ?
10.2 already has read_mbtail() that uses my_charlen(), so it seems as if the whole patch should be null-merged into 10.2. But then, why didn't you put the whole new read_mbtail() under #else?
Thanks for the review!
Correct, this entire patch should be null-merged into 10.2. Thanks for the suggestion! Let's put read_mbtail() under #else as well.
The modified version is attached.
ok to push. there are a couple of not #ifdef'ed one-liners, but they should be easy to notice and fix during the merge. not an issue. Regards, Sergei Chief Architect MariaDB and security@mariadb.org