You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, we have supported alter one column's type in meta.
But we have't supported it yet in storage.
There are some questions must be answered before going on.
How to do StatsQuery on multi version data?
For example, we want to count sum on one column. But after the column's type changed, maybe we have many different types for related data, how to deal with it.
How to deal with the response in query engine if one column in vertex data have different types. Now we don't deal with it.
How to check if the filter is valid. Now we use the latest version of the schema.
The text was updated successfully, but these errors were encountered:
I think based on the latest schema version is a valid assumption. So if the old version cannot be properly converted to the latest version during the calculation, we should just ignore that row
* Step1 use 3rdparty v3 (vesoft-inc#3554)
* Switched to use third-party v3
1) Switched from HeaderClientChannel to the RocketChannel
(common/thrif/ThriftClientManager-inc.h)
2) Fixed a bug in common/process/ProcessUtils.cpp
3) Fixed all UT
4) 99.9% of changes are mechanical: set_xxx(yyy) ==> xxx_ref() = yyy
5) Only allow empty functions on a single line
* Solved merge conflicts
1) Solved merge conflicts
2) Solved ldap issues
* Fixed lint error
Co-authored-by: Sherman <sherman.ye@vesoft.com>
Currently, we have supported alter one column's type in meta.
But we have't supported it yet in storage.
There are some questions must be answered before going on.
How to do StatsQuery on multi version data?
For example, we want to count sum on one column. But after the column's type changed, maybe we have many different types for related data, how to deal with it.
How to deal with the response in query engine if one column in vertex data have different types. Now we don't deal with it.
How to check if the filter is valid. Now we use the latest version of the schema.
The text was updated successfully, but these errors were encountered: