please help with cloud bikeshed issue?

[Hodges, Jeff]

I'm sorry but I cannot determine what exact problem cloud bikeshed is having with processing the attached index.bs file.

Here's what is happening..
% Total % Received % Xferd Average Speed Time Time Time Current

Dload Upload Total Spent Left Speed

100 117k 0 573 100 116k 239 50073 0:00:02 0:00:02 --:--:-- 50095


-rw-r--r-- 1 jehodges 110028724 573 Aug 28 08:58 errors.txt


Error running preprocessor, returned code: 1.

\033[1;96mWARNING:\033[0m The following <var>s were only used once in the document:

'authenticatorData'

If these are not typos, please add them to the 'Ignored Vars' metadata.

\033[1;33mLINK ERROR:\033[0m No 'dfn' refs found for 'clientdatajson'.

\033[1;33mLINK ERROR:\033[0m No 'dfn' refs found for 'authenticatordata'.

\033[1;31mFATAL ERROR:\033[0m Couldn't find target document section sample-scenarios:

<a href="#sample-scenarios" data-section=""></a>

\033[7;31m ✘ \033[0m Did not generate, due to fatal errors

I am pretty sure that cloud bikeshed processed this input file (ie at commit 2b72ddf) just fine early last week, and I note that you made a commit to https://github.com/tabatkins/bikeshed "5 days ago" and wonder whether said changes are causing this?

thanks,

=JeffH

[Tab Atkins Jr.]

Nonsensical errors suddenly popping up throughout the spec usually
indicate that one of the handful of transformations that operate on
the raw text of the document (before HTML parsing) is misfiring, and
yeah, that's what happened here. I'm trying to get better compat with
CommonMark, and that means dealing with both inline code spans and
fenced code blocks. My code to ignore the start of code blocks was
misfiring, and as a result large portions of your spec were being
treated as escaped code rather than working HTML. ^_^

Fixing now.

[Hodges, Jeff]

On 8/30/16, 11:53 AM, "Tab Atkins Jr." <***@gmail.com<mailto:***@gmail.com>> wrote:
Nonsensical errors suddenly popping up throughout the spec usually
indicate that one of the handful of transformations that operate on
the raw text of the document (before HTML parsing) is misfiring, and
yeah, that's what happened here. I'm trying to get better compat with
CommonMark, and that means dealing with both inline code spans and
fenced code blocks. My code to ignore the start of code blocks was
misfiring, and as a result large portions of your spec were being
treated as escaped code rather than working HTML. ^_^

Fixing now.

great, thanks.

=JeffH

[Hodges, Jeff]

On 8/30/16, 3:33 PM, "Hodges, Jeff" <***@paypal.com<mailto:***@paypal.com>> wrote:
On 8/30/16, 11:53 AM, "Tab Atkins Jr." <***@gmail.com<mailto:***@gmail.com>> wrote:
Nonsensical errors suddenly popping up throughout the spec usually
indicate that one of the handful of transformations that operate on
the raw text of the document (before HTML parsing) is misfiring, and
yeah, that's what happened here. I'm trying to get better compat with
CommonMark, and that means dealing with both inline code spans and
fenced code blocks. My code to ignore the start of code blocks was
misfiring, and as a result large portions of your spec were being
treated as escaped code rather than working HTML. ^_^

Fixing now.

great, thanks.

and I note that as of yesterday afternoon, the cloud bikeshed correctly processed our webauthn index.bs input file, thanks again!

=JeffH