Backcompat conflict

[aaronpk]

I am not 100% sure this is a bug in the implementation or a bug in the spec, so let me know if you can tell. My new h-review posts also have backcompat hReview classes. So far, the Python parser is the only one that gets them even remotely correct. (Node parser issue, PHP parser issue)

The only thing that I didn't expect to see in the Python-parsed version was this extra h-card as a child element. It's coming from the vcard inside the h-entry, which I added there to indicate the reviewer for hReview.

• Original post: https://aaronparecki.com/2016/12/15/16/dropvox
• python parsed

Based on the mf2 wiki, I'm not totally clear whether this should be ignored or not: http://microformats.org/wiki/microformats2-parsing#note_backward_compatibility_details However, it was surprising to me that it's there, and I don't want it to show up in the mf2 parsed version.

[gRegorLove]

An mf2 parser should ignore the reviewer vcard because it's mf1 inside an mf2 root. To make that reviewer information appear in an mf2 parser, a publisher should add p-author h-card to that same element.

This is an issue in php-mf2 as well, which I'm working on currently.

[kartikprabhu]

fixed in #94