DOMParser matchTag does not check if nodeType is allowed by content spec of context nodes

[kapouer]

I suppose it could check that node types content spec above parseContext.context allow the a rule's group.

[marijnh]

It doesn't, and isn't supposed to. The placing of the resulting node will take care of this, parse rules are intended to apply regardless of context.

[kapouer]

Well ok, but then something about that fails because it manages to place a type which is not allowed by content spec. E.g. two node types with a "p" tag but one allows only marks without links, and the other with links. Pasting a "p" in such a situation finds the type to be the first one declared, regardless of it being allowed or not in the context content spec.

[marijnh]

Right, that's what the context field in the parse spec is for.

[kapouer]

But then since #875 is a no no, you mean that context consistently fails in prosemirror.

[marijnh]

you mean that context consistently fails in prosemirror.

Does it? It is used when selecting parse rules, which is what this was about, I believe.