| 1 |
On Thursday 17 October 2013 05:47:49 Alexander Berntsen wrote: |
| 2 |
> pym/portage/bin/prepstrip.py: |
| 3 |
> -You import sys, shlex, _copyxattr and xattr but never use them |
| 4 |
|
| 5 |
i've run a linter on the files now |
| 6 |
|
| 7 |
> -On line 86, I would change the indentation to be |
| 8 |
|
| 9 |
i think existing style is correct. it's certainly more |
| 10 |
readable/maintainable/mutable. |
| 11 |
|
| 12 |
> -On line 98, you have too many newlines before the method |
| 13 |
|
| 14 |
fixed |
| 15 |
|
| 16 |
> -On line 175 you catch an exception "as e" but never use the 'e' |
| 17 |
|
| 18 |
fixed |
| 19 |
|
| 20 |
> -The following lines are > 79 chars: |
| 21 |
|
| 22 |
i've run a linter on the files now |
| 23 |
|
| 24 |
> -The inline comment on line 486 should have two spaces |
| 25 |
|
| 26 |
fixed |
| 27 |
|
| 28 |
> -In Prepstrip(), why do you |
| 29 |
|
| 30 |
i like the existing style better rather than returning early. makes |
| 31 |
adding/removing sections cleaner (which i've done during development). |
| 32 |
|
| 33 |
> - Also, instead of |
| 34 |
> for path, reason in stripped: |
| 35 |
> if not reason: |
| 36 |
> print(...) |
| 37 |
> else: |
| 38 |
> print(...) |
| 39 |
> why not use a ternary? |
| 40 |
|
| 41 |
inlining formats/tuples with a ternary doesn't seem like it'd really be |
| 42 |
cleaner. |
| 43 |
print((' %-*s # %s' % (align, path, reason)) if reason |
| 44 |
else (' %s' % path), file=out) |
| 45 |
that's pretty busy and not easy to track. |
| 46 |
|
| 47 |
> pym/portage/util/prallel.py: |
| 48 |
|
| 49 |
as the top of the file documents, this is imported from an external tree. i'm |
| 50 |
not making changes here in general. for python3 compat, i'm merging those to |
| 51 |
the upstream copy. |
| 52 |
|
| 53 |
stylewise, it's as expected |
| 54 |
|
| 55 |
> -Background.Wait(): almost the whole thing is in a Try, which is a |
| 56 |
> bit ehh, and if you don't want to rewrite that than maybe at least |
| 57 |
> either remove the finally or put the return inside the finally |
| 58 |
|
| 59 |
it is a bit hairy, but it's making a hairy system work. it's making sure |
| 60 |
output is captured in parallel and replied sanely rather than interleaving it |
| 61 |
all. similar to handling of exceptions. |
| 62 |
|
| 63 |
> Oh, and you use camelCase for functions and methods. In python we use |
| 64 |
> underscores. I don't know if there's some convention outside of emerge |
| 65 |
> (which is the part of Portage I'm most familiar with) to not use |
| 66 |
> underscore. If not, I would prefer it if we were consistent and used |
| 67 |
> underscores everywhere. |
| 68 |
|
| 69 |
i'll fix the new code |
| 70 |
-mike |
Archives