- From: Daniel Murphy <notifications@github.com>
- Date: Thu, 05 Nov 2026 15:28:06 -0800
- To: w3c/manifest <manifest@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
- Message-ID: <w3c/manifest/pull/1255/review/5793563318@github.com>
@dmurph commented on this pull request.
> @@ -1878,6 +1878,18 @@ <h3>
</p>
</aside>
</section>
+ <section>
+ <h3>
+ Processing a document without a manifest link
+ </h3>
+ <p>
+ A [=document=] can also have no manifest link. The user agent MAY
+ still [=processing a manifest|process a manifest=] for that
+ [=document=], passing its [=Document/URL=] as both |document
+ URL:URL| and |manifest URL:URL|, an empty [=byte sequence=] as
Can manifest URL be null? since it's never actually referenced in the algorithm with no json, it shouldn't matter?
> @@ -1878,6 +1878,18 @@ <h3>
</p>
</aside>
</section>
+ <section>
+ <h3>
+ Processing a document without a manifest link
+ </h3>
+ <p>
+ A [=document=] can also have no manifest link. The user agent MAY
+ still [=processing a manifest|process a manifest=] for that
+ [=document=], passing its [=Document/URL=] as both |document
+ URL:URL| and |manifest URL:URL|, an empty [=byte sequence=] as
I wonder if, instead of calling this out this way, we create a 'note' section saying what the processed manifest looks like for an example page with no manifest link.
--
Reply to this email directly or view it on GitHub:
https://github.com/w3c/manifest/pull/1255#pullrequestreview-5793563318
You are receiving this because you are subscribed to this thread.
Message ID: <w3c/manifest/pull/1255/review/5793563318@github.com>
Received on Thursday, 5 November 2026 23:28:11 UTC