bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

[Leo Liu]

smie-indent-forward-token appears unpredictable because sometimes it
errs with (scan-error "Unbalanced parentheses" 32 33) sometimes not.

I dug a bit deeper and found the following behaviour that might be
questionable.

For example, in a ruby-mode buffer with the following content, place
point inside the first "", and M-: (smie-indent-forward-token). Point
moves to the subsequent "" no matter how far away the second "" is.

-------- buffer content --------
"",
beg do
end
abc,
abc,
abc,
""
--------------------------------

Is this a bug?

Leo

bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

[Stefan Monnier]

> For example, in a ruby-mode buffer with the following content, place
> point inside the first "", and M-: (smie-indent-forward-token). Point
> moves to the subsequent "" no matter how far away the second "" is.
>
> -------- buffer content --------
> "",
> beg do
> end
> abc,
> abc,
> abc,
> ""
> --------------------------------
>
> Is this a bug?

It doesn't look like a bug, in the sense that smie-indent-forward-token
should be able to assume it's called from a position outside of
a string/comment.

IOW I'd tend to put the blame on the code that calls this function.


        Stefan

bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

[Leo Liu]

On 2018-06-23 14:15 -0400, Stefan Monnier wrote:
> It doesn't look like a bug, in the sense that smie-indent-forward-token
> should be able to assume it's called from a position outside of
> a string/comment.
>
> IOW I'd tend to put the blame on the code that calls this function.

I remember reading something along these lines but cannot find it. Is
this documented somewhere? If not should it be documented?

Leo

bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

[Stefan Monnier]

>> It doesn't look like a bug, in the sense that smie-indent-forward-token
>> should be able to assume it's called from a position outside of
>> a string/comment.
>> IOW I'd tend to put the blame on the code that calls this function.
> I remember reading something along these lines but cannot find it.

I don't think it's clearly documented.

> If not should it be documented?

You could try and clarify that the starting point is expected to be
*between* tokens (tho that could also be interpreted to include being
in the middle of a comment).


        Stefan

bug#31948: 26.1; smie-indent-forward-token forward-sexp in strings

[Lars Ingebrigtsen]

Stefan Monnier <monnier@IRO.UMontreal.CA> writes:

>> If not should it be documented?
>
> You could try and clarify that the starting point is expected to be
> *between* tokens (tho that could also be interpreted to include being
> in the middle of a comment).

I've now basically copied that sentence to the doc string in Emacs 28,
and I'm therefore closing this bug report.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no