Details
-
Task
-
Status: Open (View Workflow)
-
Minor
-
Resolution: Unresolved
-
None
-
None
Description
I am trying to understand the use of mariadb-binlog as a tool for viewing or processing binary logs.
I think that most of the time when you use such tool, it is for viewing or getting log events between a start point and an end point.
So, for that, you have options:
1. For starting:
- --start-position, for starting at a numeric position, or a GTID list;
- --start-datetime, for starting at a given timestamp;
- --offset for skipping a given number of entries in the log.
2. For ending:
- --stop-position, for ending at a numeric position, or a GTID list;
- --stop-datetime, for ending at a given timestamp.
The problem with these ending options is that it is almost impossible to pick the proper one for achieving the expected ending.
Especially when you started with a GTID list: how to guess a GTID list reasonably close in the future from the starting one for not getting a too long list of events?
-xxxx-datetime is not a proper choice either. With both start and end, how to choose the correct one?
There is an --offset option for starting after a given number of events from the beginning.
Why isn't there a, for example, --count option, for getting a given number of events after the starting point, for example --count=100, and then stop, whatever was the chosen option for starting?
A little like the LIMIT option in SQL queries.
This would make the usage of this tool tremendously simpler.
I have to say that I am considering adding this option myself in the source code and re-compiling, as its implementation seems simple to achieve.
But the problem is that this manipulation would have to be done again with each version update, which complicates things a little.
Regards.
P.S.: Sorry for choosing the "Task" category to post this. If you had a “Feature Request” category, I definitely would have selected that one.