py3: incorporate a tab completion function from sagenb

[fchapoton]

to continue cutting the links to sagenb

CC: @embray @jdemeyer

Component: python3

Author: Frédéric Chapoton

Branch/Commit: dc2e6f0

Reviewer: John Palmieri

Issue created by migration from https://trac.sagemath.org/ticket/24998

[fchapoton]

Commit: 9bf9f05

[fchapoton]

Branch: u/chapoton/24998

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

cdfe056

incorporate one function from sagenb

[sagetrac-git]

Changed commit from 9bf9f05 to cdfe056

[fchapoton]

comment:3

and bot is morally green

[fchapoton]

comment:4

ping ?

[jhpalmieri]

comment:5

How about some doctests for completions?

[sagetrac-git]

Changed commit from cdfe056 to 039799b

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

039799b

adding a doctest

[fchapoton]

comment:7

done

[fchapoton]

comment:8

and bot is morally green again

[embray]

comment:9

Why not go ahead and make this Python 3 compatible as well (currently I think it isn't)? Otherwise +1 from me.

[fchapoton]

comment:10

Could you precise in which way it is not py3 compatible ? I do not feel very qualified to make it so, by the way. I am quite lost in what exactly is using unicode and what else is using bytes..

[embray]

comment:11

Just try testing it on Python 3, I mean. For example, this will break:

            v += [x for x in __builtins__.keys() if x[:n] == s] 

Actually the line is broken on Python 2 as well--there is no __builtins__.keys(). Maybe it means vars(__builtins__)? I'm not sure though. I haven't read the code that closely.

[sagetrac-git]

Changed commit from 039799b to da202c8

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

da202c8

simplify the completion method

[fchapoton]

comment:13

Here is a tentative of simplification. Now pyflakes is happy.

[fchapoton]

comment:14

and bot is morally green again

[jhpalmieri]

comment:15

I'm happy with it.

[fchapoton]

comment:16

Should I remove the unused option ?

[jhpalmieri]

comment:17

Would that require deprecation?

[fchapoton]

comment:18

No, because we are not removing the original method in sagenb, just making inside sage a minimal version suited for our purposes.

[fchapoton]

comment:19

So would you give a positive review in the present state, or should I remove the ignored argument ?

[jhpalmieri]

comment:20

It seems best to remove the ignored argument, but I don't feel strongly about it.

[sagetrac-git]

Changed commit from da202c8 to dc2e6f0

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

9afe56f	Merge branch 'u/chapoton/24998' in 8.2.rc1
dc2e6f0	trac 24998 removing ignored argument

[fchapoton]

comment:22

The ignored argument has now been removed.

[jhpalmieri]

comment:23

Okay, looks good.

[jhpalmieri]

Reviewer: John Palmieri

[vbraun]

Changed branch from u/chapoton/24998 to dc2e6f0