xylaaaaa opened a new pull request, #67470:
URL: https://github.com/apache/doris/pull/67470
### What problem does this PR solve?
Issue Number: None
Related PR: None
Problem Summary: Azure Iceberg data files were coupled to Hadoop ABFS
compatibility paths and Hadoop-shaped credential maps. This change preserves
Azure account/container/object URIs, sends provider-owned native Azure
credentials from FE to BE, and routes FILE_S3/provider=azure ranges through the
native Azure Blob C++ client. SharedKey remains compatible, vended SAS supports
range reads with expiry validation, and native OAuth2 fails closed until a
complete Entra-ID implementation is available. Genuine Fabric OneLake OAuth2
remains on its existing explicit Hadoop path.
### Release note
Azure SAS credentials are passed to the BE native Azure client for Iceberg
data-file access. Native BE OAuth2 remains unsupported and returns a diagnostic
error.
### Check List (For Author)
- Test
- [x] Unit Test
- [x] Manual test (thirdparty dependency recovery and CMake regeneration
verified)
- [ ] Regression test
- Behavior changed:
- [x] Yes. Azure SAS Iceberg data-file access uses the native Azure
client.
- Does this need documentation?
- [x] No.
### Check List (For Reviewer who merge this PR)
- [ ] Confirm the release note
- [ ] Confirm test cases
- [ ] Confirm document
- [ ] Add branch pick label
--
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]
