#2716: tuple optimization

[levBagryansky]

Relates to #2716
This pr makes tuple.at faster. Testing time decreased by 2 times on my laptop.
Such test became more than 10 times faster:

#Test
[] > just-long-test
  seq > @
    *
      TRUE
      TRUE
      TRUE
      TRUE
      TRUE

---

PR-Codex overview

This PR adds new tests and a method at-without-checks to the tuple object in the eo-runtime.

Detailed summary

• Added tests for at-without-checks method with first and last element
• Added at-without-checks method to tuple object for retrieving elements without index validation

✨ Ask PR-Codex anything about this PR by commenting with /codex {your question}

[levBagryansky]

@maxonfjvipon please check

[maxonfjvipon]

@levBagryansky i actually may be any object, it's not mandatory i to be int. It can be for example

[] > my-i
  [x] > lt
    TRUE > @

And it will work. So I would say that it's expected i has lt attribute and we don't check its value

[levBagryansky]

@maxonfjvipon i should be int. Otherwise the object does not have sense

[maxonfjvipon]

@levBagryansky It should not and we can't guarantee it. All we need that i has attribute lt with 1 free attribute. That's it

[levBagryansky]

What will be (* 1 2 3) .at my-int here? Not ub but something like that, because then the method guarantees nothing

[levBagryansky]

@maxonfjvipon Also I did not write that i should be int. i must be less than length and greater or equal 0. That is all. If you object is less than length and greater or equal to 0 then you can use this method.

[maxonfjvipon]

@levBagryansky but x.lt y is not supposed to mean "x is less than y"

[levBagryansky]

Supposed to be. Actually it is abstraction. We have polymorphism here.

[levBagryansky]

We can remove it if it seems confusing

[maxonfjvipon]

@levBagryansky I would propose something like this for the description:

"Takes element from the tuple without index validation.
Here i must be an object that can be comparable with int using lt method."

[levBagryansky]

@maxonfjvipon ok

[levBagryansky]

@yegor256 please check

[levBagryansky]

@yegor256 please check

[yegor256]

@rultor merge

[yegor256]

@rultor status

[rultor]

@rultor merge

@yegor256 OK, I'll try to merge now. You can check the progress of the merge here

[rultor]

@rultor status

@yegor256 This is what's going on here:

• request 1970587478 is in processing, command is merge
• request has 6 parameter(s):
• fork: [email protected]:levBagryansky/eo.git
• fork_branch: 2716_tuple-optimization-in-eo
• head: [email protected]:objectionary/eo.git
• head_branch: master
• pull_id: 2898
• pull_title: #2716: tuple optimization
• build is not running

More information about Rultor commands you can get here.

[rultor]

@rultor merge

@yegor256 Done! FYI, the full log is here (took me 24min)