Fix for issue #1199 (regression against scapy<=2.3.2).

[karpierz]

Scapy Version: 2.4.0rc5-64
System: Windows7/Windows10
Python Version: 2.7.14/3.4.4/3.6.4

This PR fixes #1199 (regression against scapy<=2.3.2).
(+ some little obvious code improvement in sndrcv())
Please include it in TODO.

See also already closed #1276

[codecov-io]

Codecov Report

Merging #1282 into master will decrease coverage by <.01%.
The diff coverage is 90.9%.

@@            Coverage Diff             @@
##           master    #1282      +/-   ##
==========================================
- Coverage   84.79%   84.79%   -0.01%     
==========================================
  Files         160      160              
  Lines       38322    38340      +18     
==========================================
+ Hits        32495    32510      +15     
- Misses       5827     5830       +3

Impacted Files	Coverage Δ
scapy/sendrecv.py	79.52% <90.9%> (+1.26%)	⬆️
scapy/layers/tls/record.py	91.2% <0%> (-0.88%)	⬇️
scapy/layers/tls/handshake_sslv2.py	91.95% <0%> (-0.77%)	⬇️
scapy/asn1/ber.py	82% <0%> (-0.29%)	⬇️

[guedou]

Looks ok to me.

[guedou]

@gpotter2 @p-l- I approved this PR as it looks OK to me. Could you also review it and approve it if it is OK too ?

[gpotter2]

Patch is working on single packets, but not lists :/

[karpierz]

@gpotter2 Fixed

[gpotter2]

@guedou You should review it again, changes are big

[gpotter2]

I understand why you removed this try, but this PR itself is already tricky, and it makes it real hard to review + is not needed.
If it’s not bothering you too much, it would be better that you leave the current useless try.

All of those fixes will surely be fixed through the PEPin project, which will come in time

[karpierz]

Done

[gpotter2]

The logic between ˋsrandsr1` is the same. Because networking tests tend to fail more often, it would be good not to have too many doing the same thing :)

[karpierz]

Done

[karpierz]

PR is ready for merge to the master.