jgriffiths

Hi @alex27riva, thanks for reporting this issue. It appears PyPI has removed the (admittedly ancient) wally version we used. Could you try with the following PR? https://github.com/greenaddress/garecovery/pull/58

FYI the latest Python we currently test with is 3.10, if you have any issue with the upgraded wally PR above please LMK and I'll attempt to setup a 3.12...

OK, thanks for the update. I'll keep this issue open and see if Fedora/Python12 support is possible in the new year.

Oh @alex27riva did you use the above PR? I'd like to merge it if it works for you.

utack 9eea0819bd08e1f98cfcb932728d6bbb5290922d I don't have time at this stage to test this; wally uses the full proof validation code for blind asset/value proofs since we use that code already for...

@apoelstra rangeproof was the painful case for me, so thanks for that, LGTM. wally (following the Elements impl) only creates surjection proofs with a maximum of 3 used inputs, so...

> but not nearly as bad as rangeproof_sign Note for `secp256k1_rangeproof_sign`, the final `memset(prep, 0, 4096);` is redundant, having been zeroed above. If the compiler doesn't eliminate that as a...

> I don't think we have the review bandwidth to do anything with it The eternal problem lol. At this stage things are fine for us; its only if the...

Hi @ChristopherA - yes, I believe we are targeting that format, however this PR and repo aren't the place to discuss it. Please follow https://github.com/Blockstream/Jade where the PR adding support...

@Sjors please see https://github.com/ElementsProject/libwally-core/pull/369 - this allows key iteration and extracting all of their information such as type, derivation path, key origin info etc.