~jonsterling/forester#48: 
Subtrees inherent a parent only if they have a title

When creating an inline subtree with \subtree[name]{content}, the resulting tree seems to automatically infer a \parent property if and only if it has a title. For example, Theorem I.5.5 has no parent, but Corollary I.5.6 does (as can be seen from the context "SGA 1: ... > Étale morphisms > ..." in the heading). I think that all subtrees should have this behaviour by default, but either way it should probably be consistent between them.

Status
REPORTED
Submitter
~thosgood
Assigned to
No-one
Submitted
3 months ago
Updated
a month ago
Labels
needs-design

~jonsterling 3 months ago

Thanks for reporting this; I think I know where to look to solve this.

~jonsterling 3 months ago

Revisiting this, let me clarify something and then we can figure out what the desired behavior is.

  • These subtrees really do have a parent, but I think the issue you are reporting is that this is not rendered as part of the title.
  • However, it is very unclear how the title ought to be rendered in this case. For example, "SGA 1... › ... › Fundamental property of étale morphisms ›" would probably not be desirable, nor would "SGA 1... › ... › Fundamental property of étale morphisms › Untitled".

The the above reason, I think that the current behavior may be the correct one. With that said, I'm definitely open to suggestions for how to better deal with untitled nodes.

~thosgood 3 months ago

Ah, I see. Here's my first suggestion for how the title could render, though I'm not sure it's definitely good. At the moment, if I'm not mistaken, the rendering is of the form

[self.taxon] [self.number]
... > [parent.parent.title] > [parent.title] > [self.title]

I think what could make sense, for both untitled and titled trees (with parents), is to have

... > [parent.parent.title] > [parent.title]
[self.taxon] [self.number]
[self.title]

This way, both of the links that I posted (I.5.5 and I.5.6) would have "SGA 1: Étale covers and the fundamental group › Étale morphisms › Fundamental property of étale morphisms" as their titles, followed by Theorem I.5.5/Corollary I.5.6, and with the latter then being followed by its title ("Extensions of liftings").

But yeah, this would be the desirable rendering for me, but I have no idea how it would affect everybody else!

~jonsterling 3 months ago*

~thosgood, thanks very much for these concrete proposals! Something like what you describe might be better than what we are doing right now. I have to think about it a little bit, because I also need to have a good answer for what the "auto-titled" links should be (e.g. those you write as [[xxx-NNNN]]).

Register here or Log in to comment, or comment via email.