Fix reference scripts fetched from ogmios

[nielstron]

Same fix as for kupo (#253) for #252

@KtorZ maybe you can chime in on what exactly the format of scripts in utxos is in kupo/ogmios? Its mentioned to be "raw script" from which I would actually understand 0 cbor wrapping but it appears to be exactly once?

[nielstron]

This fixed the issue I had in #252 in the tool I was using

[codecov-commenter]

Codecov Report

Merging #254 (a8d1b0d) into main (0d5b3d7) will decrease coverage by 0.01%.
The diff coverage is 0.00%.

❗ Your organization is not using the GitHub App Integration. As a result you may experience degraded service beginning May 15th. Please install the GitHub App Integration for your organization. Read more.

@@            Coverage Diff             @@
##             main     #254      +/-   ##
==========================================
- Coverage   85.14%   85.14%   -0.01%     
==========================================
  Files          26       26              
  Lines        2996     2995       -1     
  Branches      719      719              
==========================================
- Hits         2551     2550       -1     
  Misses        335      335              
  Partials      110      110              

Files Changed	Coverage Δ
pycardano/backend/ogmios.py	59.69% <0.00%> (-0.16%)	⬇️

[KtorZ]

The scripts are provided in the format that correspond to the preimage of the various hashes calculated from scripts. For some unknown reasons, few interfaces like the cardano-cli have opted for encoding an already cbor-encoded binary object as cbor, resulting in this double-cbor-wrapping (dare I say it?) non-sense (albeit I have no doubt it wasn't intentional).

Using only the "raw" flat-encoded binary form of the script is unsatisfactory IMO because it still requires being cbor-encoded before hashing to obtain a valid preimage. So the most meaningful form is the cbor-flat-encoded script. Now, why do we have 2 different binary encoding format, and one of them sometimes applied twice is a discussion for another day.

[nielstron]

Thanks for this insight. So

• if the binary is always singly-cbor wrapped i would like to ask for this being clarified in the kupo/ogmios docs
• if not, does this mean the preimage for the hash can be either singly or doubly wrapped? as discussed in Fix kupo plutusv2 script from kupo #253 (comment)

[nielstron]

It was suggested by @cffls to use _try_fix_script here. However, ogmios does not return the script hash required to call the fixing method, so we can not know what the desired hash looks like.

[nielstron]

@cffls bump, this needs a decision and IMO should be merged as is

[nielstron]

Maybe we should add a test case for it? Ogmios is run in the test environment afaik

[cffls]

Maybe we should add a test case for it? Ogmios is run in the test environment afaik

Yes, we should add a test case for it before merging. I will try add one and see if it works.