storages: add missing return #9103
Conversation
selsta
force-pushed
the
missing-return
branch
from
d980fae
to
df92430
Compare
|
Thanks @selsta . I just left a minor comment about the possibility of adding |
selsta
force-pushed
the
missing-return
branch
from
df92430
to
562bdb8
Compare
| @@ -231,6 +231,7 @@ namespace epee | |||
| default: | |||
| CHECK_AND_ASSERT_THROW_MES(false, "unknown entry_type code = " << type); | |||
| } | |||
| return read_ae<int8_t>(); // dummy return to avoid compiler warning | |||
There was a problem hiding this comment.
Do you think it is a good idea to use unreachable here (in addition to this dummy return)?
monero/src/crypto/variant4_random_math.h
Line 73 in ac02af9
There was a problem hiding this comment.
After a brief discussion with selsta in IRC, we decided to not code more complicated. We might add it later.
| @@ -322,6 +323,7 @@ namespace epee | |||
| default: | |||
| CHECK_AND_ASSERT_THROW_MES(false, "unknown entry_type code = " << ent_type); | |||
| } | |||
| return read_se<int8_t>(); // dummy return to avoid compiler warning | |||
There was a problem hiding this comment.
Same comment as previous one.
There was a problem hiding this comment.
Same as a comment on previous issue.
After a brief discussion with selsta in IRC, we decided to not code more complicated. We might add it later.
|
Agreed that unreachable would be good to add (as a broadly available macro, because such cases are common), otherwise the new default behavior could hide logic errors. As it really should be available project-wide, I don't object to making the change in a later PR. |
Fixes compilation as we treat some warnings as errors as of #7481.
Closes #8622