bug#30626: 26.0.91; Crash when traversing a `stream-of-directory-files'

[Nicolas Petton <nicolas@petton.fr>]

[-- Attachment #1: Type: text/plain, Size: 894 bytes --]

Noam Postavsky <npostavs@gmail.com> writes:

> Now, I can see that letting stream0 automagically get access to the
> unfolded result can be an optimization in some cases, although in this
> case it's a pessimization.

stream.el is at its core just an implementation of lazy-cons cells, so
not letting stream0 get access to the result would mean changing the
core implementation (I'm not necessarily against it).

If we accept that `seq-elt', and other positional functions of seq.el
should not work on streams, then I could rewrite stream.el to make it a
positioned stream where previous elements are discarded after each
element generation.  However the list of supported functions from seq.el
API would be significantly reduced.

> It could also affect the semantics if unfolding the stream has side
> effects, not sure if stream.el makes guarantees about that though.

No, it doesn't.

Nico

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]