fix: retry using remote offset

[frankyn]

Note: @JesseLovelace will be driving this PR to completion as I'll be OOO over the next two weeks. Thanks Jesse!

• Ensure the tests and linter pass
• Code coverage does not decrease

Fixes #603 ☕️

[yoshi-code-bot]

Suggested change

	expect(storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	eq(false))).andThrow(exception);
	expect(storageRpcMock.getCurrentUploadOffset(eq(UPLOAD_ID))).andReturn(0L);
	expect(storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	expect(
	storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	eq(false)))
	.andThrow(exception);
	expect(
	storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	eq(false)))
	.andReturn(null);

[yoshi-code-bot]

Suggested change

	expect(storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	eq(false))).andThrow(exception);
	expect(storageRpcMock.getCurrentUploadOffset(eq(UPLOAD_ID))).andReturn(10L);
	expect(storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(10),
	eq(10L),
	eq(MIN_CHUNK_SIZE - 10),
	expect(
	storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(0),
	eq(0L),
	eq(MIN_CHUNK_SIZE),
	eq(false)))
	.andThrow(exception);
	expect(
	storageRpcMock.writeWithResponse(
	eq(UPLOAD_ID),
	capture(capturedBuffer),
	eq(10),
	eq(10L),
	eq(MIN_CHUNK_SIZE - 10),
	eq(false)))
	.andReturn(null);

[codecov]

Codecov Report

Merging #604 (d987287) into master (ea25492) will decrease coverage by 0.35%.
The diff coverage is 38.80%.

@@             Coverage Diff              @@
##             master     #604      +/-   ##
============================================
- Coverage     64.68%   64.32%   -0.36%     
+ Complexity      629      621       -8     
============================================
  Files            32       32              
  Lines          5255     5317      +62     
  Branches        510      520      +10     
============================================
+ Hits           3399     3420      +21     
- Misses         1697     1735      +38     
- Partials        159      162       +3     

Impacted Files	Coverage Δ	Complexity Δ
...om/google/cloud/storage/spi/v1/HttpStorageRpc.java	1.55% <0.00%> (-0.04%)	2.00 <0.00> (ø)
...va/com/google/cloud/storage/spi/v1/StorageRpc.java	66.23% <ø> (ø)	0.00 <0.00> (ø)
...ogle/cloud/storage/testing/StorageRpcTestBase.java	94.11% <0.00%> (-1.89%)	48.00 <0.00> (ø)
...ava/com/google/cloud/storage/BlobWriteChannel.java	68.00% <61.90%> (-6.61%)	11.00 <1.00> (+1.00)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ea25492...d987287. Read the comment docs.

[JesseLovelace]

The coverage miss seems to be due to getCurrentUploadOffset in HttpStorageRpc not being tested. This is part of way bigger issue, that entire file doesn't seem to have any coverage. I'm going to go ahead and merge this as is and it should be a follow up issue to get coverage for HttpStorageRpc

[littleK0i]

I suspect, this fix did not go so well.

I am now getting following exceptions sometimes:

com.google.cloud.storage.StorageException: java.lang.NullPointerException
Stack trace:
com.google.cloud.storage.StorageException.translateAndThrow(StorageException.java:81)
com.google.cloud.storage.BlobWriteChannel.flushBuffer(BlobWriteChannel.java:180)
com.google.cloud.BaseWriteChannel.flush(BaseWriteChannel.java:112)
com.google.cloud.BaseWriteChannel.write(BaseWriteChannel.java:139)
java.base/java.nio.channels.Channels.writeFullyImpl(Channels.java:74)
java.base/java.nio.channels.Channels.writeFully(Channels.java:97)
java.base/java.nio.channels.Channels$1.write(Channels.java:172)

I think, at very least we should have a better error message.

I use version 1.113.4. I'll try to go back to 1.113.3 and see how it goes.

Thank you.

[frankyn]

Thanks for letting us know, @wildraid!

Could you share a snippet of code that you used which caused this error?

[frankyn]

@wildraid released a new version with a fix for the issue you raised. Thanks for calling it out!

[littleK0i]

@frankyn , great news! Thanks for your help!