PHP: rfc:match_blocks

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== PHP RFC: Match blocks ======

  * Date: 2023-08-24
  * Author: Ilija Tovilo, tovilo.ilija@gmail.com
  * Status: Draft
  * Target Version: PHP 8.x
  * Implementation: https://github.com/php/php-src/pull/11933

===== Proposal =====

The [[https://wiki.php.net/rfc/match_expression_v2|match expression]] was added to PHP 8.0 with the goal of being a safer and more useful alternative to the switch statement. In its current form, each match arm is limited to a single expression. Any arm that does not nicely fold into a singular expression prevents match from being used. This RFC proposes to lift this restriction and allow the placement of blocks at the match arm site.

<code php>
// Before
switch (true) {
    // InputParameter
    case $entityExpr instanceof AST\InputParameter:
        $dqlParamKey = $entityExpr->name;
        $entitySql   = '?';
        break;

    // snip
};

// After
$entitySql = match (true) {
    // InputParameter
    $entityExpr instanceof AST\InputParameter => {
        $dqlParamKey = $entityExpr->name;
        <- '?';
    },

    // snip
};
</code>

Source: [[https://github.com/doctrine/orm/blob/2f9e98754bb451107dcffbf0a9844b90dcdee992/lib/Doctrine/ORM/Query/SqlWalker.php#L2169-L2183|doctrine/orm SqlWalker.php]]

===== Semantics =====

==== Return value ====

Match blocks may have return values. Match will propagate the return value of the executed arm. The return value of the block is the last expression after an optional list of statements, preceded by a ''<-'' symbol to denote the value flowing out of the block. If the match return value is used, each block is expected to return a value, unless it terminates early. If the match return value is not used, the returned value is simply discarded.

<code php>
$result = match($foo) {
    'bar' => {
        $l = 1;
        $r = 2;
        <- $l + $r;
    },
    'baz' => {
        throw new Exception();
    },
    'qux' => {
        // Forgot to return something
        // This will throw a MatchBlockNoValueError if executed
    },
};
</code>

Note that the return keyword is not reused in place of ''<-'' because it would be ambiguous whether the user meant to return from the match block, or return from the function. Similarly, yield can already refer to pausing a generator in this context.

==== Control statements ====

return, break, continue and goto statements out of the block are allowed //only// if the return value of match itself is not used.

<code php>
match ($foo) {
    'bar' => {
        // Returning 'baz' from the *function* (not match)
        // This is ok, because match is a standalone expression
        return 'baz';
    },
};

var_dump(match ($foo) {
    'bar' => {
        // Breaking out of match (like breaking out of switch, foreach, etc)
        // This is **not** ok, because match needs to return something to var_dump
        break;
    },
});

var_dump(match ($foo) {
    'bar' => {
        for ($i = 0; $i < 10; $i++) {
            // Continuing the for loop inside the block
            // This is ok, because the block itself is not escaped
            continue;
        }
        <- 42;
    },
});
</code>

The rationale for this decision is twofold:

  * It attempts to avoid potentially unsound control flow (second example above).
  * There are technical challenges to correctly implementing control flow that escapes mid-expression. For the interested, this is explained in more detail under "Technical implications of control statements" below. Disallowing escaping of the match block completely dodges this problem.

==== Scoping ====

Match blocks behave just like any other statement list in PHP in terms of scoping. That is, no new scope is created. All variables assigned inside the block are visible outside the block, within the same function.

<code php>
match ($foo) {
    'bar' => {
        $bar = 'I can see this';
    },
};
echo $bar; // I can see this
</code>

===== Motivation =====

The match expression has been introduced to address some shortcomings of switch statements, but currently fails to address approximately half of its use cases. Switch cases commonly contain more than one statement. [[https://github.com/nikic/popular-package-analysis|popular-package-analysis]] revealed that 3 507 of 6 012 switch statements contained at least one case with more than one statement (excluding breaks). Moreover, 29 690 of 67 563 cases were multi-statement. Since match is limited to one expressions per arm, a single arm that does not nicely fold into a singular expression prevents match from being used entirely.

It has previously been argued that limiting match arms to single expressions is beneficial for enforcing clean code. While keeping functions and consequently match arms short certainly has its merits, I personally find excessively small functions disorienting and hard to name well. Moreover, some statements (e.g. control statements) cannot be moved into separate functions.

Additionally, the [[https://wiki.php.net/rfc/pattern-matching|pattern matching RFC]] plans to further enhance match. Specifically, each match arm will be able to specify a pattern to match the value against, e.g. type checks. The Doctrine example from the introduction could become the following:

<code php>
$entitySql = match ($entityExpr) {
    // InputParameter
    is AST\InputParameter => {
        ...
    },

    // snip
};
</code>

===== Why not language-level blocks? =====

Instead of just implementing blocks for match, it has been suggested to implement blocks as a language-level concept. There are three evident use cases for block expressions.

  * Match blocks
  * Arrow function blocks
  * Right-hand side of short-circuiting operators (''??='', ''??'', ''?:'', ''? :'')

We'll discuss why it might be hard for a general block expression to solve all these use cases well.

==== Return value semantics ====

The optimal return value semantics for these three use cases are all slightly different.

  * For match, whether the block should require returning a value depends on whether the match itself returns a value.
  * Arrow function blocks should never require a return value, as they already provide ''return''. No return value should mean ''null'', to stay consistent with other functions.
  * For the remaining cases, a value should always be returned, unless the block terminates.

We could settle for a solution that works for all cases, returning ''null'' by default. Whether this solution is acceptable is likely a matter of taste.

<code php>
var_dump(match ('foo') {
    'foo' => {
        echo "foo branch reached\n";
    },
});
// foo branch reached
// NULL

var_dump(fn () => {
    <- 'foo';
}());
// string(3) "foo"

$foo ??= {
    bar();
};
var_dump($foo);
// NULL
</code>

==== Limited use cases ====

It's worth noting that the usefulness of blocks is limited due to PHPs scoping rules. In other languages, blocks can be used to prevent pollution of the current scope.

<code rust>
let foo = {
    let tmp = tmp();
    // ...
    Foo { tmp }
};
</code>

In this case, ''tmp'' resides in the isolated scope and is inaccessible outside of the block. However, given that PHP only has a single scope per function, there is no point in lexically nesting the temporary variables, other than potential visual benefits. The benefits are mainly limited to some of the short-circuiting operators (''??='', ''??'', ''?:'', ''? :''), as they may skip the execution of the block under certain conditions.

<code php>
$foo ??= {
    // This is only executed if $foo was null/undefined.
    $tmp = tmp();
    // ...
    <- new Foo($tmp);
};
</code>

==== Arrow function capturing ====

Another issue is that blocks for arrow functions have previously been rejected in two separate RFCs.

  * https://wiki.php.net/rfc/short-functions
  * https://wiki.php.net/rfc/auto-capture-closure

It seems that main concerns for both of these RFCs were related to auto- and explicit capturing, which language-level blocks cannot properly address.

==== Grammar ambiguity ====

Yet another issue is that ''{}'' is ambiguous (without arbitrary lookahead) in the general expression context, as it clashes with statement lists (i.e. the blocks you put after if statements, while loops, etc). An alternative syntax could use parentheses, although this introduces some inconsistency in the grammar.

<code php>
var_dump(match ($value) {
    'foo' => (
        echo "foo branch reached\n";
        'foo'
    ),
});
// foo branch reached
// string(3) "foo"
</code>

While this works fine, it looks somewhat odd, given that we use ''{}'' in all other cases when wrapping statements into blocks.

===== Technical implications of control statements =====

PHPs VM is in three-address form. As opposed to most machines, PHP opcodes are destructive in that they consume their operands. A consumed operand may not be consumed again. Moreover, an unconsumed operand may result in a memory leak. Control statements in match expression blocks pose a problem when they skip over the consuming opcodes of temporary VARs.

<code php>
new Foo() + match (1) {
    1 => { return; },
};
</code>

<code>
0000 V0 = NEW 0 string("Foo")
0001 DO_FCALL
0002 T2 = IS_IDENTICAL int(1) int(1)
0003 JMPNZ T2 0006
0004 JMP 0005
0005 MATCH_ERROR int(1)
0006 RETURN null
0007 MATCH_BLOCK_NO_VALUE_ERROR
0008 T3 = QM_ASSIGN null
0009 JMP 0010
0010 T4 = ADD V0 T3
0011 FREE T4
0012 RETURN int(1)
</code>

The opcode 0006 (RETURN) is always executed, skipping the 0010 (ADD) instruction, not consuming V0 and thus leaking the Foo object. This problem may be avoided by emitting a FREE opcode before RETURN. The same issue can occur when breaking out of switch statements, continuing in loops, using goto, etc. This approach is implemented in [[https://github.com/php/php-src/compare/master...iluuu1994:php-src:match-blocks-var-tracking|this PR]]. However, it has proven to be quite complex for questionable benefit.

Similarly, we run into an issue in this code.

<code php>
foo()->bar(match (1) {
    1 => { return; },
});
</code>

<code>
0000 INIT_FCALL_BY_NAME 0 string("foo")
0001 V0 = DO_FCALL_BY_NAME
0002 INIT_METHOD_CALL 1 V0 string("bar")
0003 T1 = IS_IDENTICAL int(1) int(1)
0004 JMPNZ T1 0007
0005 JMP 0006
0006 MATCH_ERROR int(1)
0007 RETURN null
0008 MATCH_BLOCK_NO_VALUE_ERROR
0009 T2 = QM_ASSIGN null
0010 JMP 0011
0011 SEND_VAL_EX T2 1
0012 DO_FCALL
0013 RETURN int(1)
</code>

The 0007 (RETURN) instruction skips over 0012 (DO_FCALL). However, the 0002 (INIT_METHOD_CALL) instruction has already received V0 (foo()) and increased its refcount to make sure the value is not released before the method bar() is called on it. Given that 0012 (DO_FCALL) is never executed and thus foo() is never released, it leaks.

Both of these issues arise because there are unfreed VARs at the time the escaping control statements in the match blocks are executed, skipping over their consuming opcodes. Disallowing the escaping of the match blocks when there are unconsumed VARs (i.e. when it is used within another expression) prevents skipping over their consuming opcodes, and thus circumvents the issue.


===== Backwards incompatible changes =====

There are no backwards incompatible changes in this RFC.

===== Vote =====

Voting starts ????-??-?? and ends ????-??-??. 

As this is a language change, a 2/3 majority is required.

<doodle title="Add support for blocks at match arms in PHP 8.x?" auth="ilutov" voteType="single" closed="true">
   * Yes
   * No
</doodle>
Page Tools
