[
https://issues.apache.org/jira/browse/TINKERPOP-2363?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18119719#comment-18119719
]
ASF GitHub Bot commented on TINKERPOP-2363:
-------------------------------------------
bitflicker64 commented on PR #3672:
URL: https://github.com/apache/tinkerpop/pull/3672#issuecomment-5851089407
Started porting this to master and hit something worth flagging first.
The port was easy: same `IntIO` bug in `graphbinaryV4.py`, same fix. But
when I went to write a reproducer for the PR description, mine didn't fail. On
master, `client.submit` turns the parameters dict into GremlinLang text
(`['x':3000000000]`) before serialization, and the server's grammar reads that
as a Long. Traversal args work the same way. So the failure that sent me here,
a JanusGraph load passing server-assigned ids back as bindings, doesn't happen
on master.
The old code is still reachable through request options:
`batchSize=3000000000` still raises `struct.error` on current master.
The original failure is still present on 3.7.x and 3.8.x, and `3.8-dev` has
the same unfixed `IntIO`. Would 3.8-dev work, merging forward to master?
> GraphBinary and numerics for Python
> -----------------------------------
>
> Key: TINKERPOP-2363
> URL: https://issues.apache.org/jira/browse/TINKERPOP-2363
> Project: TinkerPop
> Issue Type: Improvement
> Components: python
> Affects Versions: 3.4.6
> Reporter: Stephen Mallette
> Priority: Critical
>
> After TINKERPOP-2360 we've started down the path of coercing python numerics
> to Java type ranges. GraphBinary doesn't do that and therefore switching
> serializers from GraphSON to GraphBinary will have the potential to result in
> error if the python numeric overflows the Java integer space. In addition, we
> don't support {{BigInteger}} serialization apparently so that remains a
> problem. This ticket should just sort out GraphBinary and numbers all at once.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
