Fix a few issues with PipeStream #1399
Conversation
Apply similar treatment to PipeStream as sshnet#1322 did to ShellStream PipeStream now behaves much more Stream-like. In particular, it performs partial reads (instead of blocking until a certain amount of data is available), blocks until data is available (instead of returning 0 prematurely) and removes the Stream-unlike properties `BlockLastReadBuffer` and `MaxBufferLength`. Sadly I gave up trying to make a benchmark compatible with all the quirks of the previous implementation, but a dumb throughput test (reading and writing simultaneously) shows about 5.2GB/s with this implementation compared to 140MB/s previously. Some cleanup of its usage in the library followed.
| // command.Result (also returned from EndExecute) consumes OutputStream, | ||
| // which we've already read from, so Result will be empty. | ||
| // TODO consider the suggested changes in https://github.com/sshnet/SSH.NET/issues/650 | ||
|
|
||
| //Assert.AreEqual(expectedResult, actualResult); | ||
| //Assert.AreEqual(expectedResult, command.Result); | ||
|
|
||
| // For now just assert the current behaviour. | ||
| Assert.AreEqual(0, actualResult.Length); | ||
| Assert.AreEqual(0, command.Result.Length); |
There was a problem hiding this comment.
I was initially planning on getting this PR in, then getting a release out. Now I am thinking of addressing #650 beforehand by taking the break changing string Execute() to void Execute() in order to stop Execute from consuming OutputStream.
That way, we can get the CancelAsync changes (#1345), this PR with its fixes and subtle behaviour changes, and the source-breaking changes wrapped up in one release. Probably easier for consumers.
|
Thank you |
|
Thank you for fixing all these problems! Is there any estimation, in which release is the fix going to be included? |
|
I have a couple more fixes/changes I'd like to make to SshCommand before the next release, and then we can make one then. So I would say June, but the last time I answered that question I said May and it won't be May :-) |
Apply similar treatment to PipeStream as #1322 did to ShellStream
PipeStream now behaves much more Stream-like. In particular, it performs partial reads (instead of blocking until a certain amount of data is available), blocks until data is available (instead of returning 0 prematurely) and removes the Stream-unlike properties
BlockLastReadBufferandMaxBufferLength.Sadly I gave up trying to make a benchmark compatible with all the quirks of the previous implementation, but a dumb throughput test (reading and writing simultaneously) shows about 5.2GB/s with this implementation compared to 140MB/s previously.
Some cleanup of its usage in the library followed.
contributes to #650 (does not make any of the suggested source breaking changes - still worth considering)
closes #440
closes #164
closes #12
closes #1045
closes #142
of existing PRs:
closes #1070
closes #924
closes #374
closes #13