Hello, Marko, Alexander.

23.04.2019, 15:37, "Marko Mäkelä" <marko.mak...@mariadb.com>:
> Bar, thank you for the revision.
>
> I ran all tests and found a regression for ROW_FORMAT=REDUNDANT, in
> the test innodb.instant_alter_charset,redundant. We should allow any
> extension of VARCHAR maximum length (in bytes) for
> ROW_FORMAT=REDUNDANT. The test attempts to enlarge the column from
> 50*3 to 200*3 bytes. That cannot work for other ROW_FORMAT in InnoDB,
> but it does for ROW_FORMAT=REDUNDANT.

I think it's mostly broken test rather a code regression.

Now there is only one case when field length can be increased due to charset 
change
and it's utf8mb3 -> utf8mb4.

I think the only test case we need now is CHAR(70) UTF8MB3 -> CHAR(70) UTF8MB4
It should work only fro ROW_FORMAT=REDUNDANT

>
> Also, please check the spelling once more. I found "Uncode-1.1" in a comment.
>
> See also the Description of MDEV-18584. I’d want to allow a few more
> instantaneous conversion of CHAR columns in InnoDB.
> I agree that the (table->file->ha_table_flags() &
> HA_EXTENDED_TYPES_CONVERSION) is a bit ugly and could be replaced with
> something more descriptive.
>
> Marko
>
> On Fri, Apr 19, 2019 at 2:24 PM Alexander Barkov <b...@mariadb.com> wrote:
>>  Hi Marko, Eugene,
>>
>>  Thanks for good suggestions. A new version is attached.
>>
>>  See details inline:
>>
>>  On 4/19/19 2:46 PM, Marko Mäkelä wrote:
>>  > Hi Bar,
>>  >
>>  > I see that these bugs were based on wrong assumptions that ASCII or
>>  > UCS2 columns would only contain valid ASCII or UTF-16 data. We should
>>  > have tested these assumptions during the development of MDEV-15564.
>>  > Luckily these were found before the GA release of MariaDB 10.4.
>>
>>  That was my fault. When we had discussions on this topic with
>>  Eugene, I overlooked these problems. Sorry for this.
>>
>>  >
>>  > Here is just a quick note before I will return to work on Tuesday.
>>  >
>>  > Please remove ,algorithm=copy from tests. Instead, use
>>  > --enable_info
>>  > alter table ...
>>  > --disable_info
>>  > and make sure that the table contains at least 1 row when the ALTER
>>  > TABLE is executed. In that way, ALGORITHM=COPY will report that a
>>  > nonzero amount of rows were affected, while the native ALTER would
>>  > report that 0 rows were affected.
>>
>>  I added this into the new tests in the end of instant_alter_charset.test
>>
>>  Note, I didn't add this into original MDEV-15564 tests,
>>  as they mostly use empty tables.
>>
>>  >
>>  > I would also suggest testing with some bad data (non-ASCII chars in
>>  > ASCII column, or Unicode surrogate pairs in UCS2) to see what kind of
>>  > errors would be triggered (if any), or what the converted data would
>>  > look like after ALTER IGNORE TABLE.
>>
>>  This is a very good idea. I've just added "ALTER IGNORE"s
>>  into the new tests.
>>
>>  >
>>  > Please run the commit comment through a spell-checker. I spotted the
>>  > typos "correspoding" and "actuallt".
>>
>>  Done.
>>
>>  >
>>  > Also, please mention in the commit comment that these are regressions
>>  > caused by MDEV-15564. That should make our lifes easier, should we
>>  > ever decide to backport MDEV-15564 to earlier versions.
>>
>>  Done.
>>
>>  >
>>  > I will review the code and tests in detail next week. I hope that
>>  > Eugene can review it as well.
>>  >
>>  > Best regards,
>>  >
>>  > Marko
>>  >
>
> --
> Marko Mäkelä, Lead Developer InnoDB
> MariaDB Corporation

-- 
Eugene


_______________________________________________
Mailing list: https://launchpad.net/~maria-developers
Post to     : maria-developers@lists.launchpad.net
Unsubscribe : https://launchpad.net/~maria-developers
More help   : https://help.launchpad.net/ListHelp

Reply via email to