Added filepointer entries #885
Conversation
|
Hmmm the pypy took too long.. Can't see my changes should do that ? |
1 similar comment
3 similar comments
3 similar comments
| except ImportError: # pragma: no cover | ||
| # noinspection PyUnresolvedReferences | ||
| from UserDict import DictMixin as MutableMapping | ||
| from UserDict import DictMixin as MutableMapping # pragma: no cover |
There was a problem hiding this comment.
I don't think this "code coverage ignore" is worth it - just take the code coverage report with a grain of salt, understand that it's just a tool, and 100% code coverage doesn't actually guarantee everything is tested anyway.
(Side note: all the coveralls PR comments are annoying. I've seen other projects that have coveralls instead put the score as an extra commit status right next to the CI result, which I prefer.)
There was a problem hiding this comment.
It's worth it when I try to develop and uses hotkeys to jump to next covered part :) But they can be removed if you wishes.
There was a problem hiding this comment.
To be clear, I have no authority here, just trying to offer my opinion (to perhaps help refine things or move them along), without being annoying :)
2 similar comments
1 similar comment
2 similar comments
Refactored a lot of tests Removed the need for a real file Made it more generic Added some test Fixed for 26 Fixed wrapper for raises
5 similar comments
|
@bitprophet will you take a look? I should pass for all version. I rewrote some tests, to be more consistent and remove the need to actually read or write a file. |
| self._system_host_keys.load(filename) | ||
|
|
||
| with open(filename, 'r') as fp: | ||
| self._system_host_keys.load_fp(fp) |
There was a problem hiding this comment.
Since HostKeys.load() does exactly this ("with open() ...") you might as well let it, rather than duplicate that (admittedly tiny amount of) code here. Same argument applies to load_host_keys() below.
| return e.key | ||
| for entry in self._entries: | ||
| if entry.key.get_name() == key: | ||
| return entry.key |
There was a problem hiding this comment.
These diff sections just renaming e to entry are not really needed in this PR IMHO.
Allowing io.StringIO to be used.
Also cleaned up the save, so HostKey's save method is used, like the load was used.
I would also comment on the load before save... If I delete a key and try to save, wouldn't that load the key back and thereby not deleting it ?