On Tue, Dec 3, 2024 at 2:37 PM Roland Winkler <winkler@gnu.org> wrote:
On Mon, Dec 02 2024, Leo Stein wrote:
> I really wish this was more permissive. Looking at a .bib file, we
> have no way of knowing the biblio style that it's going to be set
> with. We also have no way of knowing whether the user is going to
> parse it with bibtex or biber.

The user needs to know whether she wants to use a bib file with BibTeX
or biblatex and use entry types these programs can handle.  Bibtex-mode
cannot be blamed for this.

We are both re-hashing the same points. I will say again that both bibtex and biber+biblatex can handle any types of entries. I think the more flexible approach is to permissive about entry types, and allow bibtex-parse-entry to parse an entry of any type, not just a fixed list of default types from (i) btxdoc.pdf, a piece of documentation from 1988; and (ii) biblatex.pdf, the documentation for a latex package, not the parser, biber, which indeed allows any entry type.
 

> I am still missing something... as far as I can tell, the "dialect" is
> just controlling which entries are valid. Is that right? But this is
> not within the purview of whether we use bibtex, or biber+biblatex. It
> depends on the biblio style that the user wants to use for setting
> their bibliography.

Beyond the defaults documented for BibTeX and biblatex, you are free to
write your own bst style files, and you are free to customize
bibtex-mode to your liking.  Everything we discuss here refers only to
user options of bibtex-mode.

The defaults of bibtex-mode match the defaults specified in the
documentation of BibTeX and biblatex.  It would be confusing for users
of bibtex-mode to deviate from that.

You don't understand why you say it would be confusing. Just like emacs can not read users' minds, I don't see how you can be sure about what would be confusing. There are common entry types, e.g. @software, which are in very wide use with standard bibtex.
 

> I'm happy to hear that there will be future improvements.

The goal is to facilitate the customization of bibtex-mode.  I see no
reason to change the defaults of user variables.

> I sincerely request that parsing of entries be made more permissive —
> not restricted to a list of entry types, or relying on the user to
> make some customizations [I think most users are not going to discover
> that it's possible to customize this].

It is a basic aspect of Emacs that users can customize its behavior.
But Emacs cannot (yet) read the mind of its users and foresee the
customizations they want.

I have heard rumors that reading the users' mind will become a user
option in Emacs 42.  (But I do not know whether this option will be
enabled by default.)

You don't have to try to read my mind — I am trying to make my mind known to the devs by emailing this list. I continue to strongly request that bibtex-parse-entry should be more permissive about parsing types than a predefined list from 1988.