Don't panic when a same-mint melt settles without a preimage

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!

[hash-money]

Fixes#91. cdk-mintd settles same-mint payments internally (no Lightning payment → no payment_proof), so the melt-success None-preimage arm is a legitimate, reachable case — but it carries a debug_assert!(false) that panics every debug build the moment two wallets sharing a mint pay each other (full repro in #91; hits us on every trusted→trusted transfer between staging test wallets).

The success path already handles None (event hash falls back to the invoice payment_hash, which is always present for these melts), so the fix is just: stop asserting, log the expected case at info. cargo test -p orange-sdk --lib 17/17.

[benthecarman]

lgtm but is built on #89, can you make this on top of master

[hash-money]

Rebased onto master — this branch now carries only the same-mint melt fix; the #89list_transactions commit is dropped and stays as its own PR. Applies cleanly on current master (5a540f6), no changes beyond the rebase. Thanks for the review, and sorry for the lag on this!