SebastianGruza commented on issue #3090:
URL: https://github.com/apache/hugegraph/issues/3090#issuecomment-5510153135

   @imbajin @kjswaruph — since there has been no activity here for ~6 weeks, 
I'd like to help move this forward. @kjswaruph, if you are still on it, happy 
to coordinate — just say so.
   
   **PR for the interim mitigation is up: #3184.** It applies the same 
`prepareConditionQuery` guard in `HstoreTable.queryByRange` that 
`queryByPrefix` already has (reproducer and analysis in my comment above), with 
unit tests for the guard and for the copy-not-mutate change. No on-disk or 
store-side change; unblocks all sort-key prefix/range queries on 1.7.x and 
master. One thing I changed in both `prepare*` call sites: they used to mutate 
the origin query via `resetConditions`, which core still uses for its own 
filtering after the scan returns — the PR clones instead.
   
   For the versioned sinking codec I'd like to write a short design note before 
any code, covering: (a) a versioned pushed-query blob carrying schema hints 
(`propertyKeyId → (cardinality, dataType)` for **all** keys of the label, so 
the store can skip over values it isn't filtering on) with fallback to 
server-side filtering when the store doesn't understand the version (rolling 
upgrades); (b) the Id equality contract across hugegraph-core/hugegraph-struct 
(normalize to long/string at the boundary); (c) a `Comparable` path in 
`Condition.RelationType.compare` — the gap exists on both sides, the server 
just never exercised it. As shown above, the codec alone turns the crash into 
silently empty results, so (a)–(c) need to land together for the acceptance 
criteria to hold. I'll post the note here for review unless you'd prefer a 
separate issue.
   
   The test-guard flip and the clear/recreate lifecycle (#3141) I'm leaving 
with @kjswaruph; happy to help review or take either if nobody else picks them 
up.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to