Schema suggestions/requests/errors

[Jeremy Sylvestre]

Some observations that arose during this summer's editing when attempting schema validation. Most of these work/render fine without being in the schema, but adding them to the schema would reduce the number of extraneous jing errors during validation.

Let me know if you would like some/all of this posted as an issue on github.

Schema suggested/requested additions

* Allow multiple

  frontmatter/bibinfo/website

⟶ I include links to both the book's "landing page" and github repository in the colophon. I suppose that one would argue that there should already be a link to the latter from the former (and there is for my books), making a link to the latter from the colophon unnecessary. But I would counter that as an open-source document, there should be a link to the source code within the document itself.

* Allow both of

  prelude/aside
  postlude/aside

* In the ExampleLike specification, change
  (BlockStatement)+
  to
  (BlockStatement | Case)+
⟶ Example calculations/arguments often need to be broken up into cases.

* Allow

  quantity/mag/m

⟶ Example: π rad

(Not sure if this one would currently render OK.)

* Allow

  docinfo/macros/@component

⟶ Sometimes a macro needs a different "workaround" definition for a specific output target.

* Allow

  book/subtitle/@component
⟶ Example: full version versus abridged version, with no subtitle on the full version.

* Allow

  idx/@component

⟶ Sometimes the index path/hierarchy for an entry is different when a version excludes material that contains related idx entries.

Schema Error

* xref/@text

(https://pretextbook.org/doc/guide/html/topic-cross-referencing.html)

⟶ Valid @text values "type-local-title" and "type-global-title" are missing from the list in the specification of XrefTextStyle

[Rob Beezer]

Thanks very much for the great suggestions. Some comments:

* Alex J is the keeper of units, so I'll let him see about that.

* I do not think #case was ever designed for #example, just for PROOF-LIKE. An
#example can have a #solution which can have a #proof which can have #case.
That sounds like the way to go to me. I can imagine an argument for
example/case but I think that is really something else - an example is not meant
to be a proof/argument.

Would it be too much to ask if you could make individual issues for these? I
don't really have a good way to track them. Or a single issue with "task"
(which I have seen, but never created myself.)

Thanks,
Rob

On 8/12/25 07:55, Jeremy Sylvestre wrote:
> Some observations that arose during this summer's editing when attempting schema
> validation. Most of these work/render fine without being in the schema, but
> adding them to the schema would reduce the number of extraneous jing errors
> during validation.
>
> Let me know if you would like some/all of this posted as an issue on github.
>

> *Schema suggested/requested additions*

> *Schema Error*
>
> * xref/@text
> (https://pretextbook.org/doc/guide/html/topic-cross-referencing.html <https://
> pretextbook.org/doc/guide/html/topic-cross-referencing.html>)

> ⟶ Valid @text values "type-local-title" and "type-global-title" are missing from
> the list in the specification of XrefTextStyle
>
>

> --
> You received this message because you are subscribed to the Google Groups
> "PreTeXt development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email
> to pretext-dev...@googlegroups.com <mailto:pretext-
> dev+uns...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/pretext-dev/
> CALMYOndxr9K0buopmcnX-1ZxeWeHx%2BsM1huhz%2BoU-8Zh4JgVEA%40mail.gmail.com
> <https://groups.google.com/d/msgid/pretext-dev/
> CALMYOndxr9K0buopmcnX-1ZxeWeHx%2BsM1huhz%2BoU-8Zh4JgVEA%40mail.gmail.com?
> utm_medium=email&utm_source=footer>.

[Jeremy Sylvestre]

Rob: thanks for replying. More below.

On Thu, 14 Aug 2025 at 17:27, 'Rob Beezer' via PreTeXt development <prete...@googlegroups.com> wrote:

*  I do not think #case was ever designed for #example, just for PROOF-LIKE.  An
#example can have a #solution which can have a #proof which can have #case.
That sounds like the way to go to me.  I can imagine an argument for
example/case but I think that is really something else - an example is not meant
to be a proof/argument.

I'm thinking of something like:

Example 1. Compute the characteristic polynomial and eigenvalues of matrix A = ...

Solution. [computations arriving at λ = ±1]

Example 2. Compute a basis for each of the eigenspaces of the matrix from Example 1.

Solution.

Case λ = 1. [computations]

Case λ = −1. [more computations]

^^^ Not a proof or argument.

 

Would it be too much to ask if you could make individual issues for these?

I can do that, definitely not too much to ask.

Cheers,

Jeremy S.