refactor(storage): fix confusing naming #18701
Conversation
zwang28
added
the
ci/run-backwards-compat-tests
proto/hummock.proto
Outdated
| @@ -50,7 +50,7 @@ message OverlappingLevel { | |||
| message Level { | |||
| uint32 level_idx = 1; | |||
| LevelType level_type = 2; | |||
| repeated SstableInfo table_infos = 3; | |||
| repeated SstableInfo sstable_infos = 3; | |||
There was a problem hiding this comment.
These fields are persisted in meta store. Is it safe to rename to break the backward compatibility?
I think changing the in memory struct in hummock_sdk should be enough
There was a problem hiding this comment.
Is it safe to rename to break the backward compatibility?
Renaming is safe because protobuf identifies fields by their number. I've tested it as well.
There was a problem hiding this comment.
For metadata stored in etcd backend, we use the protobuf encoding, but for those stored in sql backend, which are encoded as json, I think the fields are identified by their names.
There was a problem hiding this comment.
blob rather than json is used in sql backend, so backward compatibility is retained.
But since we rarely use pb type directly any longer, this PR will keep pb names unchanged.
I hereby agree to the terms of the RisingWave Labs, Inc. Contributor License Agreement.
What's changed and what's your intention?
This PR simply renames some fields from
table_infotosstable_info.There're some fields of
SstableInfotype namedtable_info. The naming is misleading, regarding whether thetablerefers to sstable or state table.As this PR modifies numerous files, it should be merged only when no other complex open PRs are pending, to avoid conflicts.
Checklist
./risedev check(or alias,./risedev c)Documentation
Release note
If this PR includes changes that directly affect users or other significant modifications relevant to the community, kindly draft a release note to provide a concise summary of these changes. Please prioritize highlighting the impact these changes will have on users.