Oracle data type generation #100
Conversation
| ); | ||
|
|
||
| // length is 1 if empty and can ignore, scale codec length | ||
| let pre = fs::read(oracle_data.clone()).unwrap(); |
There was a problem hiding this comment.
will name better but checking if it can be none, non using oracle causes a decoding error.....could probably fix that but using none is better
|
|
||
| // length is 1 if empty and can ignore, scale codec length | ||
| let pre = fs::read(oracle_data.clone()).unwrap(); | ||
| let oracle_option = if pre.len() == 1 { |
There was a problem hiding this comment.
And how come we consider this empty if len == 1?
There was a problem hiding this comment.
because it is scale encoded and scale encoding prefixes the length of the item, so encoding an empty vec![] looks like [0]
| @@ -22,13 +23,27 @@ async fn main() -> anyhow::Result<()> { | |||
| get_project_root()?.to_string_lossy() | |||
| ); | |||
|
|
|||
| let oracle_data = format!( | |||
| "{}/{{project-name}}_serialized_oracle_data_type.txt", | |||
| get_project_root()?.to_string_lossy() | |||
There was a problem hiding this comment.
How come a lossy conversion is being used here?
There was a problem hiding this comment.
mmm idk just copying the ones above
Co-authored-by: Hernando Castano <[email protected]>
Co-authored-by: Hernando Castano <[email protected]>
|
|
||
| pub fn generate_types() { | ||
| let schema_config = schema_for!(UserConfig); | ||
| fs::write( | ||
| "./{{project-name}}_serialized_config_type.txt", | ||
| "./tests_serialized_config_type.txt", |
There was a problem hiding this comment.
Im a bit confused about this bit - does it need to be changed back to "./{{project-name}}_serialized_config_type.txt" when this generate template is actually used?
|
|
||
| let oracle_data = ORACLE_DATA.iter().map(|x| x.encode()).collect::<Vec<_>>(); | ||
| fs::write( | ||
| "./tests_serialized_oracle_data_type.txt", |
There was a problem hiding this comment.
If i file has a .txt extension i would expect it to be human readable and valid utf-8 - im not sure if scale encoding a vector adds bytes which are invalid utf-8.
Generates and posts oracle pointer data for programs pairs with entropyxyz/entropy-core#1222, needs to be merged second to update branch in cargo toml back