[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #5557 [Tor Bridge]: Tor helpers' launch error message contains useless whitespace
#5557: Tor helpers' launch error message contains useless whitespace
------------------------+---------------------------------------------------
Reporter: asn | Owner:
Type: defect | Status: new
Priority: normal | Milestone: Tor: 0.2.3.x-final
Component: Tor Bridge | Version:
Keywords: | Parent:
Points: | Actualpoints:
------------------------+---------------------------------------------------
Comment(by asn):
I spent too many hours on this bug today. coding different fixes and
realizing that they are all ugly.
I ended up preferring these two:
a) Refactor `format_helper_exit_status()` to create the string from left-
to-right, instead of right-to-left. The problem here is the hex encode
function, since an integer-to-hex-string (or an `itoa()`) function is kind
of messy (it needs a `reverse()`). I was thinking that maybe a simpler
solution would be better while in the signal-handler realm.
b) Use `eat_whitespace()` to eat through the useless whitespace, and print
the correct substring. This seems like the nicest solution, except from
the fact that `eat_whitespace()` calls `tor_assert()` which is not signal-
handler-safe. I'm thinking of making `eat_whitespace_impl()` with an extra
boolean argument (say, `sig_safe`) which skips the `tor_assert()` if set.
Then I can macro-ify `eat_whitespace_impl()` to `eat_whitespace()` and
`eat_whitespace_signal_safe()`.
Nick, any opinions or other ideas?
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/5557#comment:3>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs