On 2017-07-25 01:15:56 +1200, Tim Uckun wrote:
> I don't like the approach with a large increment. It would mean complicated
> logic to see if you filled the gap and then update all the other peers if you
> did. It sounds like the re-order is going to be expensive no matter what. My
> primary concern are race conditions though. What if two or more users are
> trying to update the hierarchy either by inserts or updates? I can definitely
> see a situation where we have issues transactions trip over each other.
You could add a unique index over (parent, sequence_number). That way
two transactions won't be able to add a node with the same sequence
number under the same parent. You will have to handle duplicate key
errors, though.
hp
--
_ | Peter J. Holzer | we build much bigger, better disasters now
|_|_) | | because we have much more sophisticated
| | | hjp@hjp.at | management tools.
__/ | http://www.hjp.at/ | -- Ross Anderson <https://www.edge.org/>