Hi Arash,

On Wed, Apr  1, 2026 at 16:25 (+0200), Arash Esbati wrote:

> Jim <[email protected]> writes:

>> The way I read that, in light of the knowledge about the standalone
>> package, is that the person who wrote the manual either didn't know
>> about that when they wrote it, or they didn't consider it.

> Or standalone wasn't there at the time as that portion of the manual was
> written (Remember that this is Emacs and we're old ;-)  But that doesn't
> make a difference IMO, packages like standalone or subfiles[1] offer
> some non-standard, special purpose functionalities, and we try to
> support them within AUCTeX and/or RefTeX to some extent, but people
> shouldn't expect full support.

>> That is nothing *I* would disagree with, but I wonder what the people
>> who use standalone think.

> Maybe some standalone users chime in.

So far, they have been silent.  But I doubt the number of "users" who read
this list is large compared to the total number of users.

>> I note that in "info auctex", in the Multifile "Chapter" (5), it says
>> You should always set this variable to the name of the top level
>> document.
>> This may have been written without consideration of the standalone package.

> Again, I'd say that bit was written before the advent of standalone.

Quite likely.

>> Maybe it is.  I'm not trying to be argumentative (all appearances to the
>> contrary ;-), but I can imagine someone creating a TikZ diagram in a
>> standalone file (requiring multiple re-compiles), and when they are happy
>> with the diagram, wanting to compile the whole document.

> The person has to switch buffer and run C-c C-c there; that's the price
> to pay, I think.

OK, that is fine with me.

>> If you are saying that
>>         "*standalone* sub-file which uses TeX-master = the master file"
>> is an   undefined / forbidden / uninteresting   in the LaTeX world,

> I think this is what I mean, yes.

>> then that is great, because it means that I/we can define what that
>> configuration should mean in the ConTeXt world without worry about
>> breaking established, intentional behaviour in the LaTeX world.

>> Equivalently (I think), the above says
>>    "a preamble existing in this LaTeX file logically implies TeX-master = t"

>> Also equivalently (I think), the above says
>>    "TeX-master = t in a LaTeX file means we don't look for a preamble here".


>> Q: Are all those things equivalent *and* what you are saying?

> Sorry, I don't follow here.  Can you please rephrase the above?  TIA.

I meant that are all of these equivalent in your opinion (just in case it
makes fixing things up for ConTeXt easier):

(a) "*standalone* sub-file which uses TeX-master = the master file"
    is   undefined / forbidden / uninteresting   in the LaTeX world
    << I should have said "unsupported configuration" when I wrote that,
       but the expression escaped me at the time. >>

(b) "a preamble existing in this LaTeX file logically implies TeX-master = t"
    << "this LaTeX file" meaning the file you are editing when you type
       C-c C-{c,r,b} >>

Replace the third one with what I should have typed:

(c) "TeX-master = <some master file>  in a LaTeX file logically implies
    there is no preamble here"
    << i.e., AUCTeX doesn't need to look for a preamble in the LaTeX file
       we are editing when we type C-c C-{c,r,b} if TeX-master is set to a
       master file >>

[[ Aside:
With (b) and (c) I attempt to represent the two ways of taking TeX-master
and the existence of a preamble into account:
(b)
-> check the current file being edited has a preamble, and
   preamble here -> use it
   no preamble here
       -> only then see if TeX-master = <master file> and if so look there.
(c)
-> check TeX-master's value:
   <master file> -> look there, **don't even bother looking here**
   t -> look for a preamble here
]]

I'm not sure my rephrasing is helpful.  It seems rather long.

                                Jim

Reply via email to