RadioTap flags support + Dot11 FCS&improvements

[gpotter2]

This PR:

• fixes Signal strength (RadioTap.dbm_antsignal) not working #1379 by improving RadioTap fields support. This required the implementation of ReversePadField (RadioTap padding is reversed), which itself required a pkt._tmp_dissect_pos)
• Supports FCS in Dot11
• introducing pkt._tmp_dissect_pos. It is a variable which contains the size of the current layer, at the time that it is used, and can only be used during the dissection. It is very useful in length_from functions, as it gets what has been exactly dissected when the lambda gets executed. It is changing dynamically
• merging contrib/ppi with dot11: the PPI() layer was implemented twice.

[codecov-io]

Codecov Report

Merging #1381 into master will increase coverage by 0.07%.
The diff coverage is 90.62%.

@@            Coverage Diff             @@
##           master    #1381      +/-   ##
==========================================
+ Coverage    85.2%   85.27%   +0.07%     
==========================================
  Files         174      174              
  Lines       40378    40609     +231     
==========================================
+ Hits        34403    34630     +227     
- Misses       5975     5979       +4

Impacted Files	Coverage Δ
scapy/contrib/ppi_geotag.py	90% <100%> (ø)	⬆️
scapy/packet.py	76.41% <100%> (+0.08%)	⬆️
scapy/contrib/ppi_cace.py	90.38% <100%> (-0.19%)	⬇️
scapy/fields.py	90.88% <100%> (+0.05%)	⬆️
scapy/layers/bluetooth4LE.py	89.58% <100%> (ø)	⬆️
scapy/layers/ppi.py	84.09% <100%> (ø)
scapy/layers/dot11.py	90.4% <86.04%> (-0.54%)	⬇️
scapy/layers/ntp.py	91.51% <0%> (-0.27%)	⬇️
scapy/asn1/ber.py	82.28% <0%> (+0.28%)	⬆️
... and 3 more

[calebmadrigal]

This seems to be working to me (tested on both Ubuntu Linux and macOS 10.13). Any idea when we can get scapy 2.4.1? I'd love to use this as soon as possible in my project (https://github.com/calebmadrigal/trackerjacker).

[guedou]

Original PPI author?

[gpotter2]

From the contributor modules

[guedou]

I meant to change "PPI author:" to "Original PPI author:".

[guedou]

I restarted the failing test.

[p-l-]

I think PPI related stuff should be handled in a dedicated layer file, not in dot11, as it is not specific to dot11. What do you think?

[p-l-]

Could you limit what is imported here?

[gpotter2]

@p-l- Updated: moved PPI to ppi.py

[p-l-]

Can you rebase against current master? We have a conflict due to removed tests.

[guedou]

Looks good. @p-l- are the changes OK?

[guedou]

@gpotter2 this PR conflicts.

[gpotter2]

Fixed conflicts

[calebmadrigal]

@gpotter2 Any ETA on a 2.4.1 release? It'd be great prior to DEF CON 2018! :)

[gpotter2]

Oh, you’re not the first one wanting 2.4.1 for DEF CON :) I have created a project a few days ago, but I doubt it will be ready for August...

Why do you need 2.4.1 ? Can’t you just use the dev build ? You can even install it via pip with git (lookup)

[calebmadrigal]

@gpotter2 I basically just need the stuff in this PR. I could use a dev build for personal use, but I'd like a nice clean pip "scapy==2.4.1" for dependency management.

[gpotter2]

We still have stuff to check :/
https://github.com/secdev/scapy/projects/2

Other maintainers don’t have much time to check all of those, as currently preparing for BlackHat.

For DEF CON you should be able to use the dev release, we’ll try to provide 2.4.1 ASAP

[guedou]

I am not sure that v2.4.1 will be released before the conference =\ If we cannot do it, you can try to specify a git commit in requirements.txt