Inline Span static methods in Scanner builder

[keith-turner]

To build a scanner over a row, something like the following needs to be done.

 try (Snapshot snap = client.newSnapshot()) {
     CellScanner scanner =  snap.scanner().over(Span.prefix("d:")).build();
}

It would be nice to be able to do something like following instead.

 try (Snapshot snap = client.newSnapshot()) {
     CellScanner scanner =  snap.scanner().overPrefix("d:").build();
 }

This fluent style would make writing code in an IDE with code completion more seamless. To accomplish this I think the following methods should be added to ScannerBuilder. The implementations of these methods should call the existing static methods in Span.

interface ScannerBuilder {
  //implementations of following should call corresponding Span.exact() methods
  default ScannerBuilder over(Bytes row) {over(Span.exact(row));}
  default ScannerBuilder over(CharSequence row) {...}
  default ScannerBuilder over(Bytes row, Column col){...}
  default ScannerBuilder over(CharSequence row, , Column col){...}

  //implementations of following should call corresponding Span.prefix() methods
  default ScannerBuilder overPrefix(Bytes row) {over(Span.prefix(row));}
  default ScannerBuilder overPrefix(CharSequence row){...}
  default ScannerBuilder overPrefix(Bytes row, Column col){...}
  default ScannerBuilder overPrefix(CharSequence row, , Column col){...}
}

The java docs for the new methods should have a @since tag and link to the javadoc in the corresponding Span methods.

All existing test should be updated to use the new methods. For example ScannerIT, FluoIT, WorkerIT, and WeakNotificationIT should be updated to use the new methods.

[mingscraft]

@keith-turner I can have a go on this issue, Thanks

[keith-turner]

Sounds good @codeandplay I look forward to your PR. If you have any questions, feel free to ask here.

[keith-turner]

@codeandplay did you have any questions about this issue?

[kennethmcfarland]

If this is up for grabs I'd like to take a shot, if not please keep me in mind for the next help wanted as it was very nice working with you.

[mingscraft]

Hi Keith, Sorry for the delay of the fix, I have been held up with other work last week. I will be probably the same for the this week and next. If @kenneth would like to take over, I am happy if he want to have a go. @keith sorry again about this.

…

On 27 September 2017 at 21:13, Kenneth Paul McFarland < ***@***.***> wrote: If this is up for grabs I'd like to take a shot, if not please keep me in mind for the next help wanted as it was very nice working with you. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#923 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AZdFH28L4oW94gmkcbgk0cuvAQgH2n-gks5smqx2gaJpZM4PbKKF> .

[keith-turner]

Thanks @kpm1985 . Looks like its open, if you want to give it a shot.

[kennethmcfarland]

I will take it and thanks for letting me adopt this issue.

…

On Sep 27, 2017 7:37 PM, "Keith Turner" ***@***.***> wrote: Thanks @kpm1985 <https://github.com/kpm1985> . Looks like its open, if you want to give it a shot. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#923 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Acg-LHv5c0hnoOoLu0Vd2fYK3R1OVohkks5smwZXgaJpZM4PbKKF> .

[AshB2108]

Is the issue taken or can I take it ?

[kennethmcfarland]

I talked to Keith, if you want it, take it AshishPB.

…

On Thu, Oct 5, 2017 at 8:57 AM, AshishPB ***@***.***> wrote: Is the issue taken or can I take it ? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#923 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Acg-LMyKTP0iOamEj6v8tl9HqEM_WN7Xks5spPyEgaJpZM4PbKKF> .

[kennethmcfarland]

@jeremy I have vacated this issue but others have shown interest. Please file a pull request with WIP to let everyone know your have claimed and are working on this issue. I hope I have phrased this correctly.

…

On Sat, Oct 7, 2017 at 3:44 PM, Jeremy Valenzuela ***@***.***> wrote: Hey folks, new guy here. Please let me know if this one is up for grabs. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#923 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/Acg-LOA6RGvT_BVHp4GyQrdopnbXmln8ks5sp_7GgaJpZM4PbKKF> .

[jkosh44]

I've seen this pass through a bunch of hands, so I'm gonna hop on line. If this frees up I'd like to give it a shot.

[keith-turner]

I am glad to see all of the interest. I am not sure what the best solution is when multiple people would like to work on the same issue. I think the simplest thing to do is just directly ask the person who expressed interest if they are currently working on the issue or plan to shortly. For my part I can keep looking through the code and finding things that need addressing and open more issues. I just opened #938 as another help wanted issue.

I was thinking we could take the approach that Fluo takes and use optimistic locking (first person to get a PR accepted), but that seems a bit anti-social so I vetoed that idea in my head. I think its better to just discuss, although I know that can be a bit annoying sometimes because you have to wait 24 to 48 hours for a response. I am not sure how to speed up the process and make it civil, so personally I would like to see a civil process. If anyone has any suggestions, would be glad to hear them.

@AshB2108 are you currently working on this or do you plan to work on it in the near term?

[jkosh44]

Don't mean to steal this from anyone. Just saying if it opens up again I'd like to work on it.

[keith-turner]

Don't mean to steal this from anyone

I hope I didn't give that impression with my last post. I didn't think you were attempting to do that. I was just airing my thoughts based on activity I have seen on multiple issues posted to helpwanted. I was not actually responding to anything in particular on this issue, until the last sentence.

if it opens up again I'd like to work on it.

That is one thing I was trying to determine, is it open?

[jkosh44]

Gotcha, might as well give @AshB2108 some time to respond

[ctubbsii]

In my personal opinion, I think you can just start. You can always put a link to your personal branch, so others can collaborate with you on what will eventually be just a single pull request to the upstream Apache Fluo repository. Git being a distributed version control system is nice that way.

[ctubbsii]

As long as you've made a good faith effort to collaborate (including accepting feedback), nobody can fault you for "sniping" a particular bug, in my opinion.