bitcoin
BIP174: Specify PSBT_IN_PREVIOUS_TXID serialization order
This change updates descriptions for the PSBT_IN_PREVIOUS_TXID field to specify serialization order.
It was not straightforward the order with which to store the txid in a psbt. It's not clearly specified in the spec, and there are not many resources or parsers to aid in debugging PSBTv2. I found the BIP370 test vector "Case: PSBTv0 but with PSBT_IN_PREVIOUS_TXID." seems to imply that the order of this txid value in PSBT_IN_PREVIOUS_TXID should match network serialization order.
I'm torn on if "little endian" is the correct phrase to use here in conjunction with "txid". I usually think of a "txid" as the big endian, human-presented order of a txn hash, but maybe it's okay to use interchangeably.
"Little endian" is ambiguous; I've used "little endian" to describe the byte swapped form that is displayed to users. The byte order is the same as everywhere else in Bitcoin as it is the direct output from hashing.
"Little endian" is ambiguous; I've used "little endian" to describe the byte swapped form that is displayed to users. The byte order is the same as everywhere else in Bitcoin as it is the direct output from hashing.
Agreed. Endianness with regard to hash output can be ambiguous. I've updated the text removing mention of "endian" while still clearing up the order ambiguity.
32 byte txid in standard byte order, not display byte order, of the previous transaction whose output at PSBT_IN_OUTPUT_INDEX is being spent.