Re: [Sbcl-commits] master: Add remark about a COPY-LIST added in b1b08c0929

Stas Boukarev wrote:
> Is that really not evident? The standard says specifically that the
> result might be shared or identical.

You're right, the specification does say that, but with some slightly
confusing wording in one part.

CL specification regarding the REMOVE functions:
> If any elements need to be removed, the result will be a copy. The
> result of remove may share with sequence [...]

So the result "will be a copy", but it "may share with sequence"... One
should of course assume the result will indeed share with the sequence,
but I wouldn't say it's completely obvious that one should assume
that. I got bit by this recently in fact and spent quite some time
staring at this sentence before accepting reality.

The result of removing nothing from a sequence being able to be
identical to the original sequence is however quite obvious from the
text. (Neither case is really intuitive in my opinion, but the
specification is as it is.)

I have no real opinion on the commit in question, just felt like
mentioning that while you are right, the spec isn't super clear on the
details of this point.

-duuq-