`break` or `return` inside `for ... in` leaves the enumeration on the stack
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 48/100
Research direction
Start in crates/rascal/src/internal/as2_codegen/statement.rs, especially gen_for_loop and StatementKind::Return, then inspect how loop cleanup and return values are emitted. Reproduce the nested for ... in cases from the issue and check related code generation tests. Done means both break and return leave enumeration values cleaned up, so the outer loop prints x and y once each.
Written by the indexing model from the issue text.
Description
for ... in keeps the enumerated names on the stack, terminated by null, and only consumes all of them when the loop runs to the end. Leaving the loop early with break or return jumps out without removing the rest. Any enclosing for ... in then picks up those leftover values: depending on what's left, it iterates over stale names, or it stops early on the inner loop's null. In the reproducers below, the outer loop gets b from every inner loop, so it never ends.
break:
var o = {a: 1, b: 2};
for (var k in {x: 1, y: 2}) {
trace(k);
for (var j in o) {
break;
}
}
return:
function first(o) {
for (var k in o) {
return k;
}
}
for (var k in {x: 1, y: 2}) {
trace(k);
first({a: 1, b: 2});
}
Expected for both: x and y, once each.
Actual, in Flash Player (Rascal 0.3.6, SWF 8), for both: x, then b repeated without end.
So the stack isn't reset when a function returns, which is why return is affected too.
gen_for_loop sets the break label to end_label, which is after the loop's cleanup.
https://github.com/Dinnerbone/Rascal/blob/3f1a80c85dcc657cf4a67b72754b9c7869a6d9c5/crates/rascal/src/internal/as2_codegen/statement.rs#L292-L302
StatementKind::Return emits Return directly. Before leaving a for ... in early, the remaining values need popping up to and including the null. Perhaps by emitting a small loop for that, similar to the loop header: Push null, Equals2, Not, If back to the start.
For a return from nested for ... in loops, each enclosing enumeration needs this, and the return value needs to be kept, for example in a register.
Real code hitting this: https://github.com/ruffle-rs/ruffle/blob/c7c2cc871c07d94f246d496e0b41479ba429085a/tests/tests/swfs/avm1/global_instance_decls/test.as#L46-L60
- Dominant language
- Rust
- Stars
- 22
- Forks
- 2
- PR merge metrics
- No merged PRs in 30d
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from Dinnerbone/Rascal
-
Difficulty 2/5 1-3 hours Newbie friendliness 79/100
Dinnerbone/Rascal#92 · 1 comment ·
-
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
Dinnerbone/Rascal#91 · 2 reactions ·
-
Difficulty 5/5 Over a week Newbie friendliness 30/100
Dinnerbone/Rascal#88 ·
-
Difficulty 3/5 1-2 days Newbie friendliness 55/100
Dinnerbone/Rascal#87 · 1 comment ·
-
Difficulty 3/5 1-2 days Newbie friendliness 58/100
Dinnerbone/Rascal#76 ·
All issues in Dinnerbone/Rascal
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
CakeRepository/agentshangout#995 ·
-
area:frontend kind:bug priority:p3 ready
Difficulty 2/5 1-3 hours Newbie friendliness 88/100
Maintainers usually reply within 1 day