RE: PR #137 and #138

Weird, I see that too.

Commenting out this line in .gitattributes:
* binary diff

Makes things work properly again. For some reason the default is not being overridden by the more specific *.bs rule. The weird thing is that "git diff master" does work without this modification, so in some corner of git's crazy inner workings it does understand that this is a text file. Maybe someone more knowledgeable about git can find a way to debug this further.

For now, in the interest of making forward progress, I would:

-          Modify .gitattributes to comment out the line above

-          Do the merge

-          Revert the change to .gitattributes

From: Rolf Lindemann [mailto:rlindemann@noknok.com]
Sent: Tuesday, July 12, 2016 1:08 AM
To: Vijay Bharadwaj <vijaybh@microsoft.com>; 'Rolf Lindemann' <rlindemann@noknok.com>
Cc: public-webauthn@w3.org
Subject: AW: PR #137 and #138

Hi Vijay,

I would like to do the merge but git merge is treating index.bs as *binary* file and hence doesn't do anything helpful:
>git merge master -v
>warning: Cannot merge binary files: index.bs (HEAD vs. master)

.gitattributes exists and contains the line
*.bs              cleantext

Do you get the same message when trying the merge?

Kind regards,
  Rolf
Von: Vijay Bharadwaj [mailto:vijaybh@microsoft.com]
Gesendet: Dienstag, 12. Juli 2016 01:42
An: Rolf Lindemann
Cc: public-webauthn@w3.org<mailto:public-webauthn@w3.org>
Betreff: PR #137 and #138

Rolf,

It looks like you have at least two LGTMs on each of these, but the PRs have conflicts that need to be resolved before they can be merged. What are the next steps here?

It seems like #138 can be merged as is, and #137 can be merged modulo Adam's question on the AAGUID. Perhaps an AAGUID field could also be added to the TPM attestation format? The Android attestation format would not seem to need such an addition as it already carries all the information that one would need inside the SafetyNet response.