Implement asynchronous support in ODataJsonLightBatchWriter

[gathogojr]

Issues

This pull request is in partial fulfilment of issue #2019.

Description

Implement asynchronous support in ODataJsonLightBatchWriter

To prevent a breaking change, I added the following 6 protected virtual asynchronous methods to the public abstract ODataBatchWriter class as asynchronous wrappers over the synchronous methods.

• WriteStartBatchImplementationAsync
• WriteEndBatchImplementationAsync
• WriteStartChangesetImplementationAsync
• WriteEndChangesetImplementationAsync
• CreateOperationRequestMessageImplementationAsync
• CreateOperationResponseMessageImplementationAsync

These virtual asynchronous methods are then invoked from the public asynchronous methods. For example;

public Task WriteStartBatchAsync()
{
    // ...
    return TaskUtils.GetTaskForSynchronousOperation(this.WriteStartBatchImplementation);
}

protected abstract void WriteEndBatchImplementation();

becomes

public Task WriteStartBatchAsync()
{
    // ...
    return this.WriteStartBatchImplementationAsync();
}

protected abstract void WriteEndBatchImplementation();

protected virtual Task WriteStartBatchImplementationAsync()
{
    return TaskUtils.GetTaskForSynchronousOperation(this.WriteStartBatchImplementation);
}

This I believe should guarantee asynchronous methods in the abstract base class will work as before unless overridden in a derived class.
It also allows asynchronous support to be implemented by overriding the protected methods in the derived class(es).

Checklist (Uncheck if it is not completed)

• Test cases added
• Build and test with one-click build and test script passed

Additional work necessary

If documentation update is needed, please add "Docs Needed" label to the issue and provide details about the required document change in the issue.

[KenitoInc]

LGTM

[habbes]

Out of curiosity, do you know why ContinueWith is not used here instead?

[habbes]

Or just using await on the second statement, i.e.

await this.WriteEndBatchImplementationAsync();
await this.FlushAsync();

What else is FollowOnSuccessWithTask doing behind the scenes?

[gathogojr]

You can think of FollowOnSuccessWithTask as a "syntactic" sugar for ContinueWith, although it's a little more than that. We actually use ContinueWith in the method. The method is designed to propagate failures if the antecedent task fails. If WriteEndBatchImplementationAsync throws an ODataException, we propagate that to the caller - not an AggregateException like ContinueWith would do.
The delegate passed to FollowOnSuccessWithTask will not be executed if the antecedent task fails. Specifically in this case, we do not wish to execute FlushAsync if WriteEndBatchImplementationAsync fails.
The method is defined in TaskUtils class just in case you'd like to study the implementation further.

[habbes]

Out of curiosity, what's the difference between FollowOnSuccessWithTask and FollowOnSuccessWith

[habbes]

Are there async versions of this.FinishWriteStartChangeSet()?

[gathogojr]

FollowOnSuccessWithTask supports delegate that return another Task instance while FollowOnSuccessWith doesn't. As an illustration consider the following two usages:

Task.Factory.StartNew(() =>
{
    // Do work
    return new Random().Next(1, 100);
}).FollowOnSuccessWith((antecedentTask) =>
{
    Console.WriteLine("Result: {0}", antecedentTask.Result);
});

// vs.

Task.Factory.StartNew(() =>
{
    // Do work
    // return new Random().Next(1, 100);
}).FollowOnSuccessWithTask((antecedentTask) =>
{
    return Task.Factory.StartNew(() => Console.WriteLine("Result: {0}", antecedentTask.Result));
});

[gathogojr]

There's no async version of FinishWriteStartChangeSet. The private method currently doesn't do any writing. It's an implementation detail so it'd be easy to introduce an async equivalent if the synchronous version evolved to doing any writing.

[habbes]

Could you remind me what the synchronous argument does again? I remember it from a previous PR but don't remember what it was used for. I think it was used in the dispose methods, but I'm not quite sure.

[gathogojr]

Reference to #2109, ODataWriteStream previously had no way of telling whether we're writing synchronously or asynchronously. We address that gap by introducing this argument such that when the Dispose method is invoked, either of StreamDisposed or StreamDisposedAsync is called depending on the value passed for the argument.
This approach is necessitated by the fact that we support NET45 and NETSTANDARD1_1 both of which do not support IAsyncDisposable that would make it feasible to use DisposeAsync in asynchronous scenarios.

[habbes]

What does FollowOnFaultWith do? Could this be safely transformed to something like:

try {
    await this.FlushAsynchronously();
} catch {
    this.SetState(BatchWriterState.Error);
}

[gathogojr]

FollowOnFaultWith is designed such that the Action delegate is invoked only when the antecedent task fails (i.e. task is in a Faulted state)

[gathogojr]

It could be transformed the way you demonstrated. But instead of littering the code with such try...catch I think it's neater to encapsulate that in a helper method

[pull-request-quantifier-deprecated]

This PR has 935 quantified lines of changes. In general, a change size of upto 200 lines is ideal for the best PR experience!

---

Quantification details

Label      : Extra Large
Size       : +889 -46
Percentile : 97.83%

Total files changed: 6

Change summary by file extension:
.cs : +889 -46

Change counts above are quantified counts, based on the PullRequestQuantifier customizations.

Why proper sizing of changes matters

Optimal pull request sizes drive a better predictable PR flow as they strike a
balance between between PR complexity and PR review overhead. PRs within the
optimal size (typical small, or medium sized PRs) mean:

• Fast and predictable releases to production:
  • Optimal size changes are more likely to be reviewed faster with fewer
    iterations.
  • Similarity in low PR complexity drives similar review times.
• Review quality is likely higher as complexity is lower:
  • Bugs are more likely to be detected.
  • Code inconsistencies are more likely to be detetcted.
• Knowledge sharing is improved within the participants:
  • Small portions can be assimilated better.
• Better engineering practices are exercised:
  • Solving big problems by dividing them in well contained, smaller problems.
  • Exercising separation of concerns within the code changes.

What can I do to optimize my changes

• Use the PullRequestQuantifier to quantify your PR accurately
  • Create a context profile for your repo using the context generator
  • Exclude files that are not necessary to be reviewed or do not increase the review complexity. Example: Autogenerated code, docs, project IDE setting files, binaries, etc. Check out the Excluded section from your prquantifier.yaml context profile.
  • Understand your typical change complexity, drive towards the desired complexity by adjusting the label mapping in your prquantifier.yaml context profile.
  • Only use the labels that matter to you, see context specification to customize your prquantifier.yaml context profile.
• Change your engineering behaviors
  • For PRs that fall outside of the desired spectrum, review the details and check if:
    • Your PR could be split in smaller, self-contained PRs instead
    • Your PR only solves one particular issue. (For example, don't refactor and code new features in the same PR).

How to interpret the change counts in git diff output

• One line was added: +1 -0
• One line was deleted: +0 -1
• One line was modified: +1 -1 (git diff doesn't know about modified, it will
  interpret that line like one addition plus one deletion)
• Change percentiles: Change characteristics (addition, deletion, modification)
  of this PR in relation to all other PRs within the repository.

---

Was this comment helpful? 👍  :ok_hand:  :thumbsdown: (Email)
Customize PullRequestQuantifier for this repository.