From: "Brian Goetz" <brian.go...@oracle.com>
To: "Remi Forax" <fo...@univ-mlv.fr>
Cc: "amber-spec-experts" <amber-spec-experts@openjdk.java.net>
Sent: Saturday, February 26, 2022 5:49:08 PM
Subject: Re: [External] : Re: Primitive type patterns
BQ_BEGIN
BQ_BEGIN
BQ_BEGIN
#### Relationship with assignment context
BQ_END
That's a huge leap, let's take a step back.
I see two questions that should be answered first.
1) do we really want pattern in case of assignment/declaration to support
assignment conversions ?
2) do we want patterns used by the switch or instanceof to follow the exact
same rules as patterns used in assignment/declaration ?
BQ_END
I agree we should take a step back, but let's take a step farther -- because I
want to make an even bigger leap that you think :)
Stepping way far back .... in the beginning ... Java had reference types with
subtyping, and eight primitive types. Which raises an immediate question: what
types can be assigned to what? Java chose a sensible guideline; assignment
should be allowed if the value set on the left is "bigger" than that on the
right. This gives us String => Object, int => long, int => double, etc. (At
this point, note that we've gone beyond strict value set inclusion; an int is
*not* a floating point number, but we chose (reasonably) to do the conversion
because we can *embed* the ints in the value set of double. Java was already
appealing to the notion of embedding-projection pair even then, in assignment
conversions; assignment from A to B is OK if we have an embedding of A into B.)
On the other hand, Java won't let you assign long => int, because it might be a
lossy conversion. To opt into the loss, you have to cast, which acknowledges
that the conversion may be information-losing. Except! If you can prove the
conversion isn't information losing (because the thing on the right is a
compile-time constant), then its OK, because we know its safe. JLS Ch5 had its
share of ad-hoc-seeming complexity, but mostly stayed in its corner until you
called it, and the rules all seemed justifiable.
Then we added autoboxing. And boxing is not problematic; int embeds into
Integer. So the conversion from int => Integer is fine. (It added more
complexity to overload selection, brought in strict and loose conversion
contexts, and we're still paying when methods like remove(int) merge with
remove(T), but OK.) But the other direction is problematic; there is one value
of Integer that doesn't correspond to any value of int, which is our favorite
value, null. The decision made at the time was to allow the conversion from
Integer => int, and throw on null.
This was again a justifiable choice, and comes from the fact that the mapping
from Integer to int is a _projection_, not an embedding. It was decided
(reasonably, but we could have gone the other way too) that null was a "silly"
enough value to justify not requiring a cast, and throwing if the silly value
comes up. We could have required a cast from Integer to int, as we do from long
to int, and I can imagine the discussion about why that was not chosen.
Having set the stage, one can see all the concepts in pattern matching dancing
on it, just with different names.
Whether we can assign T to U with or without a cast, is something we needed a
static rule for. So we took the set of type pairs (T, U) for which the pattern
`T t` is strictly total on U, and said "these are the conversions allowed in
assignment context" (with a special rule for when the target is an integer
constant.)
When we got to autoboxing, we made a subjective call that `int x` should be
"total enough" on `Integer` that we're willing to throw in the one place it's
not. That's exactly the concept of "P is exhaustive, but not total, on T"
(i.e., there is a non-empty remainder.) All of this has happened before. All of
this will happen again.
So the bigger leap I've got in mind is: what would James et al have done, had
they had pattern matching from day one? I believe that:
- T t = u would be allowed if `T t` is exhaustive on the static type of u;
- If there is remainder, assignment can throw (preserving the invariant that if
the assignment completes normally, something was assigned).
So it's not that I want to align assignment with pattern matching because we've
got a syntactic construct on the whiteboard that operates by pattern matching
but happens to looks like assignment; it's because assignment *is* a
constrained case of pattern matching. We've found the missing primitive, and I
want to put it under the edifice. If we define pattern matching correctly, we
could rewrite JLS 5.2 entirely in terms of pattern matching (whether we want to
actually rewrite it or not, that's a separate story.)
The great thing about pattern matching as a generalization of assignment is
that it takes pressure off the one-size-fits-all ruleset.
BQ_END
Pattern matching is not a generalization of assignment, it's a generalization
of the declaration + assignment, not assignment only.
Apart if you are suggesting that we should extend pattern matching to be able
to bound an existing variable ?
BQ_BEGIN
You can write:
int x = anInteger
but it might throw NPE. In many cases, users are fine with that. But by
interpreting it as a pattern, when we get into more flexible constructs, we
don't *have* to throw eagerly. If the user said:
if (anInteger instanceof int x) { ... }
then we match the pattern on everything but null, and don't match on null, and
since instanceof is a conditional construct, no one needs to throw at all. And
if the user said:
switch (anInteger) {
case int x: ...
// I can have more cases
}
the `int x` case is taken for all values other than null, and the user has a
choice to put more patterns that will catch up the remainder and act on them,
or not; and if the user chooses "not", the remainder is implicitly rejected by
switches handling of "exhaustive with remainder", but the timing of such is
moved later, after the user has had as many bites at the apple as desired
before we throw. The rules about assignment are the way they are because
there's no statically trackable side-channel for "was it a good match". Now
there is; let's use it.
So, enough philosophy; on to the specific objections.
BQ_BEGIN
For 1, given that we are using pattern to do destructured assignment, we may
want to simplify the assignment rules to keep things simple avoid users
shooting themselves in the foot with implicit unboxing.
With an example,
record Box<T>(T value) {}
Box<Integer> box = ...
Box<>(int result) = box; // assignment of result may throw a NPE
BQ_END
I assume you mean "let Box... = box". Assuming so, let's analyze the above.
BQ_END
'let' as keyword is not necessary here and cause more harm than good for
everybody that switch regularly between JS and Java, but let us not be
sidetracked by that (pun intended).
BQ_BEGIN
The pattern Box(P) is exhaustive on Box<Integer> if P is exhaustive on Integer.
If we say that `int result` is exhaustive on Integer (which I'm proposing),
then the remainder of `Box(int result)` will be { null, Box(null) }. The
pattern won't match the remainder (but *matching* does not throw), but the
let/bind construct says "OK, if there was remainder, throw" (unless there's an
else clause, yada yada.) So yes, the above would throw (I don't think it should
throw NPE, but we're going to discuss that in a separate thread) not because of
the pattern -- pattern matching *never* throws -- but because the let construct
wants exhaustiveness, and accepts that some patterns have remainder, and makes
up the difference by completing abruptly. It's just like:
Box<Box<String>> bbs = new Box(null);
let Box(Box(String s)) = bbs;
Here, Box(null) is in the remainder of Box(Box(String s)) on Box<Box<String>>,
so the match fails, and the construct throws. Unboxing is not really any
different, and I wouldn't want to treat them differently (and I worry that yet
again, there's some "anti-null bias" going on.) In both cases, we have a nested
pattern that is exhaustive but has non-empty remainder.
BQ_BEGIN
I don't think we have to support that implicit unboxing given that we have a
way to ask for an unboxing explicitly (once java.lang.Integer have a
de-constructor)
Box<>(Integer(int result)) = box;
BQ_END
This is just moving the remainder somewhere else; Box(null) is in the remainder
of Box(Integer(BLAH)), since deconstructing the Integer requires invoking a
deconstructor whose receiver would be null. I think what you're reaching for
here is "the user should have to explicitly indicate that the conversion might
not succeed", which would be analogous to "the user should have to cast Integer
to int". But we've already made that decision; we don't require such a cast.
BQ_BEGIN
I think we should not jump with the shark too soon here and ask ourselves if we
really want assignment conversions in case of destructured assignment.
BQ_END
See above; I think this is the wrong question. It is not a matter of "do we
want assignment conversions in destructuring", it is "do we want to be able to
*derive* the assignment conversions from pattern matching."
BQ_END
Pattern matching is a restrictive form of assignment, given that it always
creates a fresh variable, so your analogy breaks here.
Worst, there are other existing analogies, that are also perfectly valid
- 'var' is also a restrictive form of assignment that always creates a fresh
variable like pattern matching, and 'var' as different rules as assignment.
- there is also another construct that creates fresh variables, lambdas. And
lambdas does not allow assignment conversions.
In both cases, you can run with that analogy as eloquently as you are doing
with the "pattern matching is assignment" analogy.
I'm not against allowing assignment conversions on pattern in the context of an
assignment but in a switch or in an instanceof (when the context does not
require totality) it means introducing a new semantics i really dislike (cf the
sister message).
Rémi
