Here is another tricky one. From the spec:
“cs:group may carry the delimiter attribute to separate its child
elements, as well as affixes and display attributes (applied to the
output of the group as a whole) and formatting attributes (transmitted
to the enclosed elements).”
My interpretation of this was that each child node inherits the
formatting options of cs:group — that they apply to each child node
individually and not on the output of all the nodes as a whole.
But this test seems to contradict this:
The referenced style is:
And here I am interested in the “vol-page” macro:
As you can see there is even a comment highlighting this very issue. The
results call for a result like “3,” like the comment suggests,
but I would have expected this to still render as “3,” because of
the fact that the spec says a cs:group’s formatting attributes are
transmitted to the enclosed elements.
I suspect my interpretation of ‘transmitted to’ is wrong; on the other
hand, the rendering implied by the test case seems like there is no
special treatment of cs:group formatting attributes, so why would this
be mentioned in the spec at all?
I’m pretty sure this has been discussed at some point, since the
language in the spec is rather specific, but I couldn’t find that
discussion. Frank, do you know from the top of your head?
Sorry for replying so late. I’ve stumbled across the thread while looking
for something else.
“Transmitted” must mean what Sylvester assumes that it means, but that’s
obviously not what the test require. I think that clause should just be
removed from the spec.
At some point I may have thought that italics etc. should be set
exclusively on the rendering elements, but as far as I can tell (thinking
about it now) the only effect of that would have been to reduce flexibility.
huh - missed this, too. I agree with Frank. We want to keep the
current behavior of citeproc-js and fix the specs.
Thanks for the clarification! I think Frank is completely right that
this actually reduces flexibility, especially in the context of macros.
I’ll remove the ‘feature’ from citeproc-ruby, too.
signature.asc (181 Bytes)
FYI I just created a pull request for this:
This would assume that formatting attributes are still valid (this also means there should be no change to the schema necessary) and applied to the group as a whole (I assume this is how citeproc-js would treat the situation now?).
signature.asc (203 Bytes)