Add sorted and list.sort

[ghost]

No description provided.

[codecov-io]

Codecov Report

Merging #81 into master will increase coverage by 0.7%.
The diff coverage is 67.96%.

@@ Coverage Diff @@## master #81 +/- ##
=========================================
+ Coverage 68.65% 69.36% +0.7% 
=========================================
Files 59 60 +1 Lines 10525 10778 +253 =========================================
+ Hits 7226 7476 +250 + Misses 2790 2775 -15 - Partials 509 527 +18

Impacted Files	Coverage Δ
builtin/builtin.go	80.17% <100%> (+0.49%)	⬆️
py/list.go	46.71% <62.92%> (+11.58%)	⬆️
repl/repl.go	100% <0%> (ø)	⬆️
py/range_repr110.go	70% <0%> (ø)
py/type.go	52.36% <0%> (+0.81%)	⬆️
py/internal.go	41.78% <0%> (+0.91%)	⬆️
py/sequence.go	54.83% <0%> (+2.15%)	⬆️
py/method.go	61.32% <0%> (+2.83%)	⬆️
py/float.go	35.23% <0%> (+2.85%)	⬆️
... and 4 more

---

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 eb115a9...5f7a289. Read the comment docs.

[corona10]

@Tim-St Hi, Thanks for the contribution.
I know implementing sort for python interpreter(tim sort) is very hard.
Would you like to add the test about the list for your change?
The test should be run at here.
https://github.com/go-python/gpython/blob/master/py/tests/list.py
Might be coverage will be increased.

[ghost]

@corona10 Ok, I will add some test cases.
There is a PR at golang that implements Block Sort, which will improve the stable sort, once it's merged. Timsort isn't wanted in Go because it uses O(n) space currently.

[corona10]

@Tim-St Thanks for the information!

[corona10]

@ncw@sbinet Can you take a look at this PR?

[ghost]

I will add one more thing.

a= [1,3,2]
assertlist.sort(a) isNoneasserta== [1, 2, 3]

should work. Currently this syntax doesn't work for list.append etc too.

[ncw]

This looks like excellent work :-)

If we could get the coverage of the patch up to above the threshold that would be perfect!

I didn't see anything which I thought needed changing :-)

[ghost]

@ncw Thanks! I'd like to change it to get a higher score, but I don't understand how "codecov" measures the results, and there are no hints on the page which parts should be changed :\

[ncw]

@ncw Thanks! I'd like to change it to get a higher score, but I don't understand how "codecov" measures the results, and there are no hints on the page which parts should be changed :\

:-)

The output is a bit cryptic

The easiest way to improve the coverage is to use the go coverage tools locally to see the bits which haven't been covered in the code you've added. Cover those and the coveralls score will increase!

[ghost]

Hm, when I run this cover tool, most of the code in list.go is marked red, not only my code. I don't know how to improve this, maybe someone can show me how to get this right.

[corona10]

This looks like excellent work :-)
If we could get the coverage of the patch up to above the threshold that would be perfect!
I didn't see anything which I thought needed changing :-

@Tim-St cc @ncw
If there no issue with the current code without code coverage.
We can go to merge it.

[corona10]

@Tim-St
Thanks for your contribution!