Support bit registers, arrays and index expressions as for-loop iterables (#393)

[TheGupta2012]

Closes #393.

Problem

Qasm3Visitor._visit_forin_loop branched on the concrete type of set_declaration and fell through for anything that was not a DiscreteSet or a RangeDefinition. The spec also allows a bit[n] value, a one-dimensional array[<scalar>, n], and an index expression arriving at either.

Approach

The branch is factored into _resolve_loop_iterable, which returns the plain list of values plus the loop-variable initializer. The unrolling loop below is unchanged.

For an Identifier or IndexExpression, the expression is evaluated once and the result dispatched on type: a BitValue yields its bits in b[0], b[1], ... order (bit 0 is the leftmost bitstring character); a one-dimensional ndarray yields tolist(). tolist() produces native Python scalars, so the loop variable is bound by copy and a write to it cannot reach the source array. Anything else — a scalar, a BitstringLiteral — still raises the original "Unexpected type ... of set_declaration in loop." error. The existing loop-limit check now covers the new forms too.

Tests

tests/qasm3/test_loop.py: a bit[n] register asserting MSB-first order, a bit loop variable over a register, a one-dimensional array, an index-expression iterable (a[1:2], inclusive), a write to the loop variable asserting the source array is unchanged, a multi-dimensional array raising ValidationError, and a scalar index expression raising the unsupported-type error.

Gap

A loop variable of type bit inherits the pre-existing unroll behavior: each iteration emits its own bit[1] x; declaration, exactly as for bit x in {0, 1} does today. That is unchanged by this PR.

Classical let aliases (case 3 of the issue) are not implemented. let c = b; on a classical register still fails at the alias statement itself, before any loop is reached, because _visit_alias_statement only resolves quantum registers. That work is tracked in #392.

Stacked on #404 (feature-385-391-bit-repr-negative-indices); rebases onto main once #404 merges.

🤖 Generated with Claude Code

[argus-eye]

Argus review

Auto-review is off for this repo. Tick the box below to run a review on this PR.

• Trigger Argus review

Estimated cost

• Files changed: 3
• Diff lines (±): 221
• Historical avg: ~243.6k tokens · ~$0.95 · across last 10 review(s)

Tip: you can also comment @argus-eye review at any time.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 1c168ea6-f8f2-4b7c-bdc0-41cf0812069a

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}