There are two significant gaps:
1) the "none" result does not distinguish between a record with a policy
"p=none" vs. the lack of a record. These are not the same as authentication
should be done for a record with p=none and results reported back to the
sender.
2) no analysis of the "org" domain
As a suggestion, take a look at https://github.com/linkedin/dmarc-msys and
particularly at these 4 important functions:
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L242
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L261
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L326
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L342
LUA and python are pretty similar so you should be able to easily leverage the
capabilities. It would be great to have a fully compliant DMARC module in PiPy
1) the "none" result does not distinguish between a record with a policy
"p=none" vs. the lack of a record. These are not the same as authentication
should be done for a record with p=none and results reported back to the
sender.
2) no analysis of the "org" domain
As a suggestion, take a look at https://github.com/linkedin/dmarc-msys and
particularly at these 4 important functions:
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L242
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L261
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L326
https://github.com/linkedin/dmarc-msys/blob/master/dmarc.lua#L342
LUA and python are pretty similar so you should be able to easily leverage the
capabilities. It would be great to have a fully compliant DMARC module in PiPy