Create upgrade_column_info_4_6_5+.sql

[danielgp]

Before submitting pull request, please check that every commit:

• [x ] Has proper Signed-Off-By
• [x ] Has commit message which describes it
• [x ] Is needed on it's own, if you have just minor fixes to previous commits, you can squash them
• [x ] Any new functionality is covered by tests

Bookmark id that is an auto_increment column can only take positive values starting from 0, therefore should be unsigned otherwise half of allowed interval is wasted from start, and this is the fix for that

Signed-off-by: Daniel Popiniuc danielpopiniuc@gmail.com

[codecov-io]

Current coverage is 50.18% (diff: 100%)

Merging #12676 into QA_4_6 will not change coverage

@@             QA_4_6     #12676   diff @@
==========================================
  Files           485        485          
  Lines         81344      81344          
  Methods        2126       2126          
  Messages          0          0          
  Branches          0          0          
==========================================
  Hits          40821      40821          
  Misses        40523      40523          
  Partials          0          0          

Powered by Codecov. Last update 8556b5d...732735d

[danielgp]

#12676 I have added the Sign-Off-By line and now everything should be ok

[nijel]

Why this change? Is int limiting number of your bookmarks?

[danielgp]

@nijel
This change is meant to bring consistency for all "auto_increment" fields in the database which do have the "unassigned" flag except the bookmark one.
Having the "unassigned" flag for an "auto_increment" field is quite normal as the first value will be 1 which is giving a full use of the interval, without it 50% of the interval will be wasted right from the start, see https://dev.mysql.com/doc/refman/5.7/en/integer-types.html .
In conclusion even if the limitation of int(11) without unassigned is quite high, to 2+ billion records, it's more about consistency with all the other "auto_increment" fields, and avoid wasting 50% of the interval right from the start than bringing the limit to 4+ billion.

[nijel]

Okay, as this is rather cleanup than bug fix, it should rather go to master branch instead of QA_4_6.

[danielgp]

@nijel I have modified this merge request against master as instructed, thanks!

[nijel]

Yes, please rebase your changes on master and open PR for that.