Fixing warning C4101: unreferenced local variable #4821

pull ENikS wants to merge 1 commits into bitcoin:master from ENikS:C4101_fix changing 9 files +10 −10

ENikS commented at 7:34 PM on September 2, 2014: contributor

No description provided.
Fixing C4101 warning 4d7d387734
in src/rpcserver.cpp:None in 4d7d387734
```
 627 | @@ -628,7 +628,7 @@ void StartRPCThreads()
 628 |              try {
 629 |                  vEndpoints.push_back(ParseEndpoint(addr, defaultPort));
 630 |              }
 631 | -            catch(boost::system::system_error &e)
 632 | +            catch(...)
```
sipa commented at 7:37 PM on September 2, 2014:

This is not exactly equivalent (haven't verified whether it matters).

ENikS commented at 7:46 PM on September 2, 2014:

'e' is not being used and it is faster because type of exception is not being checked.

sipa commented at 7:47 PM on September 2, 2014:

But now this block will catch all types of exceptions, instead of just boost::system::system_error ones, which may not be the intention.

ENikS commented at 8:04 PM on September 2, 2014:

Well, it might be so but unlikely. Judging from the fact that boost::system_error derived from std::runtime_error it would be safe to assume that intention was to handle any abnormal condition.

gavinandresen commented at 8:16 PM on September 2, 2014:

NAK on replacing with catch (...). We want errors like std::bad_alloc (out of memory) to either be handled by a higher-level catch or to cause an unhandled exception exit.
BitcoinPullTester commented at 7:48 PM on September 2, 2014: none

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/p4821_4d7d387734f5643bf95af8ffc681534b755c558b/ for binaries and test log. This test script verifies pulls every time they are updated. It, however, dies sometimes and fails to test properly. If you are waiting on a test, please check timestamps to verify that the test.log is moving at http://jenkins.bluematt.me/pull-tester/current/ Contact BlueMatt on freenode if something looks broken.
ENikS commented at 9:15 PM on September 2, 2014: contributor

Fair enough, will approach it differently.
ENikS closed this on Sep 2, 2014
ENikS deleted the branch on Sep 2, 2014
laanwj commented at 6:08 AM on September 3, 2014: member

You could remove the e instead of the entire clause?

Agree that this is the wrong solution, never make catch clauses more general than they should be.
ENikS commented at 4:54 AM on September 5, 2014: contributor

@laanwj It was my next step.
DrahtBot locked this on Sep 8, 2021

Contributors