Set QuickXorHash as checksum to compare file changes #17311
Conversation
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
|
We have Hashes for Files available already: quickXorHash is guaranteed to be available. |
|
Alternatively, use the cTag1-property:
Relying on a implementation detail of the eTag is dangerous. Footnotes |
There was a problem hiding this comment.
Pull Request Overview
This PR implements UUID-based checksum comparison for OneDrive/SharePoint by extracting UUIDs from ETag values that remain constant when moving files, enhancing file synchronization accuracy.
- Modifies
Checksumclass to support UUID extraction from ETag values formatted as"{UUID},version" - Updates OneDrive features to use checksum-based comparison instead of timestamp-only comparison
- Adds comprehensive test coverage for checksum behavior during file operations
Reviewed Changes
Copilot reviewed 21 out of 21 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
core/src/main/java/ch/cyberduck/core/io/Checksum.java |
Enhanced checksum parsing to support Base64 and improved hex string handling with proper validation |
onedrive/src/main/java/ch/cyberduck/core/onedrive/features/GraphAttributesFinderFeature.java |
Added checksum extraction from OneDrive file hashes (QuickXorHash) |
onedrive/src/main/java/ch/cyberduck/core/onedrive/GraphProtocol.java |
Updated comparison service to use checksum-based comparison with fallback to timestamp |
onedrive/src/test/java/ch/cyberduck/core/onedrive/*Test.java |
Added comprehensive test assertions for checksum behavior during write and move operations |
dkocher
changed the title
Co-authored-by: Copilot <[email protected]>
This reverts commit fea39f3
dkocher
force-pushed
the
bugfix/MD-24072
branch
from
cddafca to
46bc3b1
Compare
| if(file != null) { | ||
| final Hashes hashes = file.getHashes(); | ||
| if(hashes != null) { | ||
| attributes.setChecksum(Checksum.parse(hashes.getQuickXorHash())); |
There was a problem hiding this comment.
This relies on quickXorHash always being 20 Bytes, and is misrepresented as SHA1.
There was a problem hiding this comment.
You are right. Full implementation requires us to implement quickXorHash as a ChecksumCompute feature.
Extract UUID from ETag formatted as"{D7751A14-FEA5-4BDA-969E-F8022E58B7E7},9"which remains constant when moving files.