[tor-bugs] #18362 [Core Tor/Tor]: Tor could use a generic 'handle' implementation.

[Tor Bug Tracker & Wiki]

#18362: Tor could use a generic 'handle' implementation.
-------------------------------------------------+-------------------------
 Reporter:  nickm                                |          Owner:  nickm
     Type:  enhancement                          |         Status:
 Priority:  High                                 |  needs_revision
Component:  Core Tor/Tor                         |      Milestone:  Tor:
 Severity:  Normal                               |  0.2.9.x-final
 Keywords:  modularity, TorCoreTeam201604, tor-  |        Version:
  modularity                                     |     Resolution:
Parent ID:                                       |  Actual Points:
 Reviewer:  dgoulet                              |         Points:  small
                                                 |        Sponsor:
                                                 |  SponsorS-can
-------------------------------------------------+-------------------------

Comment (by cypherpunks):

 So, this implementation is totally broken for multithreaded use (I thought
 this was the main use-case). I found several other synchonisation errors,
 but seeing how the whole thing would probably need rework, it doesn't seem
 like you'd be particularly interested, are you?

 The only other thing I could add is that `_new()` and `_clear()` are
 missing a `tor_assert(object)` at the start. And that maybe you should be
 more clear in the documentation about the critical importance of calling
 `_clear` before destroying the object.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/18362#comment:15>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online