Remove obsolete _MSC_VER check #10642

Prior to this commit the _MSC_VER macro was being used regardless of being defined due to a misplaced #endif.

Thanks. However I investigated and we can remove this check completely:

• MSVC++ 8.0 _MSC_VER == 1400 (Visual Studio 2005) (source https://stackoverflow.com/questions/70013/how-to-detect-if-im-compiling-code-with-visual-studio-2008)
• C++11 is supported since MSVC 2010: https://msdn.microsoft.com/en-us/library/hh567368.aspx
• Compiling without C++11 support is no longer possible.

@laanwj Thanks for the clarification! If I understand it correctly the new diff should be:

diff --git a/src/init.cpp b/src/init.cpp
index ed76953..e213958 100644
--- a/src/init.cpp
+++ b/src/init.cpp
@@ -844,8 +844,6 @@ bool AppInitBasicSetup()
     // Turn off Microsoft heap dump noise
     _CrtSetReportMode(_CRT_WARN, _CRTDBG_MODE_FILE);
     _CrtSetReportFile(_CRT_WARN, CreateFileA("NUL", GENERIC_WRITE, 0, NULL, OPEN_EXISTING, 0, 0));
-#endif
-#if _MSC_VER >= 1400
     // Disable confusing "helpful" text message on abort, Ctrl-C
     _set_abort_behavior(0, _WRITE_ABORT_MSG | _CALL_REPORTFAULT);
 #endif

In other words keeping the #ifdef _MSC_VER but removing the #if _MSC_VER >= 1400. Looks correct? :-)

Yes, indeed. That retains the behavior to not do any of that stuff for non-MSVC compilers (e.g. mingw). (which I think is the intent? that code is ancient)

@laanwj Updated accordingly :-)

Travis fails at p2p-segwit.py in one of the seven triggered builds. Seems totally unrelated. Non-determinism?

Failed test:

2017-06-21 12:40:32.262000 TestFramework (INFO): Testing detection of non-standard P2WSH witness
2017-06-21 12:40:33.627000 TestFramework (ERROR): Assertion failed
Traceback (most recent call last):
  File "/home/travis/build/bitcoin/bitcoin/build/bitcoin-i686-pc-linux-gnu/test/functional/test_framework/test_framework.py", line 152, in main
    self.run_test()
  File "/home/travis/build/bitcoin/bitcoin/build/bitcoin-i686-pc-linux-gnu/test/functional/p2p-segwit.py", line 1945, in run_test
    self.test_non_standard_witness()
  File "/home/travis/build/bitcoin/bitcoin/build/bitcoin-i686-pc-linux-gnu/test/functional/p2p-segwit.py", line 1822, in test_non_standard_witness
    self.std_node.test_transaction_acceptance(p2wsh_txs[0], True, False, b'bad-witness-nonstandard')
  File "/home/travis/build/bitcoin/bitcoin/build/bitcoin-i686-pc-linux-gnu/test/functional/p2p-segwit.py", line 81, in test_transaction_acceptance
    assert_equal(self.last_message["reject"].reason, reason)
  File "/home/travis/build/bitcoin/bitcoin/build/bitcoin-i686-pc-linux-gnu/test/functional/test_framework/util.py", line 418, in assert_equal
    raise AssertionError("not(%s)" % " == ".join(str(arg) for arg in (thing1, thing2) + args))
AssertionError: not(b'bad-txns-vout-negative' == b'bad-witness-nonstandard')

Not sure if it's worrying but that must be unrelated. I've restarted the travis build.

Test failure caused by attempting to spend a negative value in p2p_segwit: https://travis-ci.org/bitcoin/bitcoin/jobs/245346398#L3109

CTransaction(nVersion=1 vin=[CTxIn(prevout=COutPoint(hash=6d7956f3bdf3f5c646146311bcf255305fcfbc7ccc60e6d70be216491278a54c n=0) scriptSig= nSequence=0)] vout=[CTxOut(nValue=-1.99996597 scriptPubKey=0014b472a266d0bd89c13706a4132ccfb16f7c3b9fcb)] ...

I've seen this failure once or twice before from Travis. Looks like the test isn't verifying that the utxo it uses for splitting is large enough, though I'm not sure how it could end up negative. @jl2012 I think you added this test, mind having a look?