Removal of six & styling #1051
Removal of six & styling #1051
Conversation
|
How hard would it be to split these into two different PRs, one for |
|
Well, they are split by commits, but why the worry, they are all tidy-up steps without any functional changes?
|
|
No concern really, should be fine. |
|
All right @ashleysommer & @white-gecko: please review and, if this gets past, we can then look into all the functional change PRs properly. There are lots of edits here but they are all the same thing, many times over, in many files, so you should be able to be happy with the pattern after a few checks... |
| self.depth -= 1 | ||
|
|
||
| return True | ||
|
|
||
| def isValidList(self, l): | ||
| def isValidList(self, l_): |
There was a problem hiding this comment.
Just curious, why wasn't just l OK here? (And if it isn't, the docsstring ought to correspond.)
There was a problem hiding this comment.
Only because flake8 threw up a lot of warnings about possible variable ambiguity for l so there must be a higher-order variable l still within scope that this might clash with.
I suppose I should have thought o rebuild docstrings before putting in the PR. I'll try and do that and add to this.
There was a problem hiding this comment.
the flake8 error regarding l is because it can look like an uppercase i or a number 1. Its generally considered using a single letter variable named l is a bad thing. This solution works to suppress the error, and I think doesn't harm anything.
There was a problem hiding this comment.
I see. In practise I think l is good enough (in spite of flake8), but following that general advice, I think lst would be better, or even if we could establish a naming pattern like rlist (and similarly one for types, like rtype). (Going for rdf_list and rdf_type may seem even better/clearer, unless it becomes unclear whether they would reference an object or the terms themselves (RDF.type and RDF.List).)
Just something to consider going further, nothing critical for this PR.
examples/film.py
Outdated
| @@ -35,7 +35,7 @@ | |||
|
|
|||
| from rdflib import BNode, ConjunctiveGraph, URIRef, Literal, Namespace, RDF | |||
| from rdflib.namespace import FOAF, DC | |||
| from six.moves import input | |||
| from builtins import input | |||
There was a problem hiding this comment.
You don't need to import input from builtins. Anything in builtins is automatically imported into every file by default, you can just use input() without importing anything.
There was a problem hiding this comment.
Thanks @ashleysommer, I've removed that line.
|
@white-gecko can you please review this? |
This PR removes all references to the package six which handles PY2/PY3 differences since we are no longer catering for PY2.
All tests pass locally - I even ran the SPARQLUpdateStore tests!
I've also
blacked all Python files to ensure consistent formatting and addressed all issues fromflake8testing except for broad exceptions & line length warnings.No functionality changes.