[thebrandonlucas]

Stack 3 of 6. Base: feat/cli-next-step-after-accept (#62). Next: fix/scripts-build-then-run.

Closes #50.

Problem

The wallet picker showed each wallet as name UUID. A USD wallet and a BTC wallet looked the same, so a user who picked a USD wallet for a BTC receive without --quote got quote_required. The guide said to pick a BTC wallet but gave no way to see which one that was.

Behavior

Wallet's balances	Picker label
All report usd	Payroll USD UUID
Empty, missing, not an array, a balance with no currency, or currencies that disagree	Payroll UUID (as before)

• The label is the picker's filter text, so typing btc matches wallets that show BTC (and any wallet with btc in its name). A wallet with no balance yet shows no currency; the guide and CHANGELOG say so.
• The currency is scrubbed like the name.
• Organization and environment choices are unchanged.
• The picker still offers every wallet. It does not hide wallets that cannot take the payment, which is out of scope in Show the wallet currency in the wallet picker #50.

login-and-profiles.md and the changelog are updated.

Checks

On head 1f4770e:

• cargo fmt --check, cargo clippy --locked --all-targets -- -D warnings, cargo machete --with-metadata, cargo test --locked (109 unit, 93 integration), and python3 scripts/check-coverage.py pass. All ran in nix develop.
• New paths have tests. A unit test covers agreeing balances, a scrubbed currency, and each case with no label. The pseudo-terminal wallet-pick test shows ops USD UUID and picks the BTC wallet by typing btc.
• CRAP: 0 of 523 measured functions at or above 20. The highest is wait_for_payment at 19.60, unchanged from master.
• I reviewed the diff with voltage-pr-review and the Rust style checklist.
• Not run: a bless run, because no global flag changed.

Current-head reviewer validation

Reviewed 386c61c against feat/cli-next-step-after-accept. The head is unchanged from the completed local review.

• Verification gate passed on this head: cargo fmt --check, cargo clippy --locked --all-targets -- -D warnings, cargo machete --with-metadata, cargo test --locked, and python3 scripts/check-coverage.py. These reviewer checks used Rust 1.98.1 on Linux.
• scripts/crap-report.sh --output-dir target/crap-pr-64 passed. Scope: src/, build.rs, and tests/; all targets and features. Highest score: wait_for_payment at 19.60. No functions scored 20 or higher. Measured: 523; unmeasured: 139 test functions.

[thebrandonlucas]

Review at d8fdb1a (Aurelius). No findings.

• currency() returns a currency only when every balance names the same one. The unit test covers empty, mixed and missing currencies, and a non-array value.
• The label is uppercased, then passed through scrub_line. The test checks that an ESC sequence is stripped.
• The pseudo-terminal test shows ops USD … and picks the BTC wallet by typing btc. A UUID is hex, so btc and usd can't match one by accident.

Checks:

• Gate on 1f4770e: fmt, clippy, machete, tests (109 unit, 93 integration) and coverage pass.
• CRAP on 1f4770e: 0 functions at or above 20. The highest is wait_for_payment at 19.60.
• d8fdb1a changes only CHANGELOG.md and troubleshooting.md from 1f4770e. Signed G blu@blu.cx.

[thebrandonlucas]

Review of d8fdb1a, which has the same src/ and tests/ as 1f4770e. This is not an approval. I checked the backend at origin/master 1630b77c.

1. Medium: a BTC wallet that has never been funded shows no currency, and typing btc hides it. currency() (src/pick.rs:207) returns None for empty balances. In the backend, Wallet.balances is a Vec<Balance> (crates/voltage_eventing/src/wallet/entity.rs:53). The backend's own wallet classifier says that a wallet with empty balances is a never-funded BTC wallet, because USD and asset wallets always get balance entries when they are set up (crates/cli/scripts/fetch_wallet_config.sh:117-125). So a new BTC wallet, which is the likely target of the payments receive example, is listed as name UUID. login-and-profiles.md:55 then says that typing btc "lists only those", which leaves that wallet out. The CHANGELOG line at CHANGELOG.md:19 says "shows each wallet's currency", which also does not hold for those wallets.
Fix: choose one of these.

• (a) Make the docs and CHANGELOG say what the code does: a wallet with no balance yet shows no currency, and filtering on btc leaves it out.
• (b) Show empty balances as BTC, as the backend classifier does. Cite that rule in a comment and add a unit test.

I lean towards (a) unless the backend guarantees the rule. Also, a BTC wallet does not need a line of credit (line_of_credit_id: Option<Uuid>, entity.rs:50). So "a BTC wallet, which is on a BTC line of credit" is stronger than what the currency shows.

2. Nit: the filter matches anywhere in the label. inquire 0.9.4 is built without the fuzzy feature, so its default scorer is a case-insensitive substring match on the full label (select/mod.rs:176). A USD wallet named btc-sweep also matches btc. Fix: change "lists only those" to "lists those first", or to "matches their currency".

No other problems:

• Organization and environment listings have no balances, so they keep name UUID.
• The currency goes through scrub_line, and a unit test covers the escape sequence.
• The unit test covers each None branch: empty, mixed, a missing currency, and a non-array value. The pseudo-terminal test checks the USD label and the btc filter.

Checks on d8fdb1a:

• Gate, run in nix develop: fmt, clippy, machete, tests (109 unit, 93 integration) and coverage all pass, and the tree is clean afterwards.
• CRAP: 0 of 523 measured functions are at or above 20. The highest is wait_for_payment at 19.60, the same as on master. nix develop has no llvm-tools-preview, so I ran CRAP with the rustup toolchain.
• Not run: a staging check of what wallets list returns for an unfunded BTC wallet. Finding 1 comes from reading the backend source.

[thebrandonlucas]

Re-review at 386c61c (Aurelius). src/ is unchanged since my review of 1f4770e. The new login-and-profiles.md:55 text is accurate: no currency for a wallet with no balance, and the filter matches the whole label. No findings. Signed G blu@blu.cx.

[thebrandonlucas]

Re-review at 386c61c. This is not an approval. My medium finding and my nit are fixed, and src/ is unchanged.

• login-and-profiles.md:55 and the CHANGELOG now say that a wallet with no balance yet shows no currency, and that btc also matches a wallet's name.
• The line-of-credit claim is dropped.

The gate passes on 8dd04b0, which contains this commit: 109 unit and 93 integration tests, and the tree is clean.

[tee8z]

Reviewed at 386c61c. CRAP on this head: 0 of 523 functions at or above 20, highest wait_for_payment at 19.60. No blocking issues.

• The picker reads the currency from balances[].currency, while api::wallet_currency derives it from the line-of-credit summary, and the two can disagree (an unfunded wallet, for example). Skipping a request per wallet is reasonable. A one-line comment on currency() saying it is a display hint would prevent someone from relying on it later.
• If an asset wallet's balance currency has the asset:<hex> form, like the line-of-credit currency, the label becomes ASSET:<64 HEX>, and uppercasing changes the asset ID. Worth checking against the backend.
• The PR description ends with a "Generated with Claude Code" line; AGENTS.md says to leave tool attribution out of PR descriptions.

The inline suggestions remove a Some(match …) by letting with_currency take the Option. I applied the suggestions on this head and checked them: cargo clippy --all-targets -- -D warnings is clean and the unit and integration tests pass.

[tee8z]

With with_currency taking an Option (see terminal.rs), the match goes away:

Suggested change

	let choice = Choice::new(id, name);
	Some(match currency(item) {
	Some(currency) => choice.with_currency(currency),
	None => choice,
	})
	Some(Choice::new(id, name).with_currency(currency(item)))

[tee8z]

Taking the Option here lets choices chain it:

Suggested change

	pub fn with_currency(mut self, currency: &str) -> Self {
	self.currency = Some(scrub_line(&currency.to_uppercase()));
	pub fn with_currency(mut self, currency: Option<&str>) -> Self {
	self.currency = currency.map(|currency| scrub_line(&currency.to_uppercase()));

[tee8z]

Approved 386c61c against feat/cli-next-step-after-accept. The picker preserves unknown or mixed currency behavior and scrubs displayed text. Remaining simplification suggestions are nonblocking. This approves the layer against #62; that base still has requested changes.

Validation on this unchanged head: cargo fmt --check, cargo clippy --locked --all-targets -- -D warnings, cargo machete --with-metadata, cargo test --locked, and python3 scripts/check-coverage.py passed. CRAP: 523 measured functions, none at or above 20; highest wait_for_payment 19.60. 139 test functions were unmeasured.