use hex-conservative 1.0 everywhere for decoding #822
Conversation
|
Drafting. Might as well just wait for the real 1.0. |
apoelstra
force-pushed
the
push-ppmrkrkvqkzo
branch
3 times, most recently
from
fb0adc3 to
bea8cde
Compare
|
Published :) will update this tonight or tomorrow morning. |
apoelstra
force-pushed
the
push-ppmrkrkvqkzo
branch
from
bea8cde to
78e8f01
Compare
|
PR title and git log are both stale - they mention 'rc'. |
| } | ||
| } | ||
| } | ||
| use miniscript::hex; |
| let depo_tx: Transaction = deserialize(&Vec::<u8>::from_hex(hex_tx).unwrap()).unwrap(); | ||
| let depo_tx: Transaction = deserialize_hex(hex_tx).unwrap(); |
There was a problem hiding this comment.
This is kind of an unrelated change, bad Andrew no biscuit.
There was a problem hiding this comment.
How is it unrelated? We used to have a hex library that has from_hex and now we don't.
There was a problem hiding this comment.
oooph - I'm a spastic. I did not even look inside the function call to deserialize only at the function name changes - ouch.
| Vec::<u8>::from_hex("76a91479091972186c449eb1ded22b78e40d009bdf008988ac").unwrap()[..] | ||
| hex::decode_to_vec("76a91479091972186c449eb1ded22b78e40d009bdf008988ac").unwrap()[..] |
There was a problem hiding this comment.
Out of interest did you find yourself cursing the f***ing hex-conservative devs or were you happy enough during the upgrade?
There was a problem hiding this comment.
I was happy enough, but the lack of encoding support is idiotic.
apoelstra
changed the title
It turns out we only ever use do decoding of fixed test vectors. It would be nice to have a hex! macro for this. Eventually. In some cases we were able to use inherent methods on bitcoin types -- for example ScriptBuf::from_hex or deserialize_hex.
apoelstra
force-pushed
the
push-ppmrkrkvqkzo
branch
from
78e8f01 to
a1fa898
Compare
|
Going to merge this one -- in general, I don't want to move rust-miniscript to 1.0 dependencies on the main branch because I have been hyping up "Miniscript 13" to people out-of-band and want the next release to be compatible with bdk. But in the case of hex-conservative, the only "1.0 types" that are expose are error types, so it's okay. |
…ywhere for decoding
a1fa898f9f164baee2eca787b06fd89ea10f880d use hex-conservative 1.0 everywhere for decoding (Andrew Poelstra)
Pull request description:
It turns out we only ever use do decoding of fixed test vectors. It would be nice to have a hex! macro for this. Eventually.
In some cases we were able to use inherent methods on bitcoin types -- for example ScriptBuf::from_hex or deserialize_hex.
ACKs for top commit:
tcharding:
ACK a1fa898f9f164baee2eca787b06fd89ea10f880d
Tree-SHA512: 234fb4974e86b6b72d3f0aaefbe7f1ea73cfd90c0152c163d3f0b46e94d50f3e784f5499e00b058736b242b1c0dde39b41d5a2583319bf9c580a0db84dc24237
2 participants
It turns out we only ever use do decoding of fixed test vectors. It would be nice to have a hex! macro for this. Eventually.
In some cases we were able to use inherent methods on bitcoin types -- for example ScriptBuf::from_hex or deserialize_hex.