Opened 6 years ago
Last modified 5 years ago
#5770 task new
Deprecate IntNStringReceiver.recvd
Reported by: | therve | Owned by: | therve |
---|---|---|---|
Priority: | low | Milestone: | |
Component: | core | Keywords: | |
Cc: | Branch: |
branches/deprecated-recvd-5770
branch-diff, diff-cov, branch-cov, buildbot |
|
Author: | therve |
Description
It was kept for compatibility purpose when the latest optimization was introduced, it should be deprecated.
Change History (7)
comment:1 Changed 6 years ago by
Summary: | Deprecate IntNStringReceiver.recv → Deprecate IntNStringReceiver.recvd |
---|
comment:2 Changed 5 years ago by
Author: | → therve |
---|---|
Branch: | → branches/deprecated-recvd-5770 |
comment:3 Changed 5 years ago by
Owner: | set to therve |
---|
comment:4 Changed 5 years ago by
Keywords: | review added |
---|---|
Owner: | therve deleted |
Ready for review, build results here: http://buildbot.twistedmatrix.com/boxes-supported?branch=/branches/deprecated-recvd-5770
comment:5 Changed 5 years ago by
Keywords: | review removed |
---|---|
Owner: | set to therve |
Thanks for fixing this.
Looks fine, modulo some minor points:
- I think that these pyflakes errors are spurious, but if not, fix it.
clearBuffer
's docstring really should be a lot better. What is "the current buffer"? It might be best to describe this in terms of the behavior ofstringReceived
anddataReceived
.# new protocol: luckily it's giving that with a handy method
is not a very useful comment. Perhaps ifclearBuffer
's docstring explained what it did, this comment could simply be eliminated.
Please address to your satisfaction and merge.
comment:6 Changed 5 years ago by
Do we not care that clearing a buffer provides a smaller feature set than being able to assign to recvd
? It might be nice to see a unit test that does the same thing as test_recvdChanged
without using a deprecated API. And it seems like we're just discarding the feature of being able to inspect the current buffer?
comment:7 Changed 5 years ago by
Also, doesn't the set-case warning point to the wrong code? The test asks for deprecations pointed at test_recvdChanged
, but that's not where the code that writes to recvd
is.
(In [37980]) Branching to 'deprecated-recvd-5770'