[WebAuthn] Support makeCredential for virtual HID authenticators

[[e-mail]]

^

[Radar WebKit Bug Importer]

<rdar://problem/90393676>

[[e-mail]]

Created attachment 454898 [details]
Patch

[Brent Fulgham]

Comment on attachment 454898 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=454898&action=review

r=me

> Source/WebKit/UIProcess/WebAuthentication/Virtual/VirtualAuthenticatorUtils.mm:41
> +    int8_t flags = 0b01000000; // AT set

Nit: It might be nice to define constants for these values:

constexpr int8_t ATset = 0b01000000;
... etc...

> Source/WebKit/UIProcess/WebAuthentication/Virtual/VirtualAuthenticatorUtils.mm:45
> +        flags = flags | 0b00000100;

Then you could explain what this one means :-)

> Source/WebKit/UIProcess/WebAuthentication/Virtual/VirtualCredential.h:39
> +    bool isVerificationRequired;

Nit: Putting the bools together at the end of the struct can allow better packing of the structure and reduce memory use.

> Source/WebKit/UIProcess/WebAuthentication/Virtual/VirtualHidConnection.cpp:230
> +            Vector<uint8_t> buffer;

Is buffer ever large enough that it's worth hinting the size of the buffer before appending using the 'reserveCapacity' method?

buffer.reserveCapacity(payload.size() + 1)

Scanning through the code, there might be a number of places in the WebAuthn code where this could be helpful.

[[e-mail]]

Created attachment 454915 [details]
Patch for landing

[EWS]

ChangeLog entry in Source/WebCore/ChangeLog contains OOPS!.

[[e-mail]]

Created attachment 454983 [details]
Patch for landing

[EWS]

Committed r291423 (248550@main): <https://commits.webkit.org/248550@main>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 454983 [details].