fix leading zero bug #255
fix leading zero bug #255
Conversation
src/services/verifier.rs
Outdated
| let encoded_info = &attr_info.encoded; | ||
|
|
||
| let reveal_attr_encoded = if leading_zero_regex.captures(encoded_info).is_some() { | ||
| encoded_info.trim_start_matches('0') |
There was a problem hiding this comment.
I could not use the captured group sadly as it captures the entire number, seems like an insignificant performance decrease to me.
There was a problem hiding this comment.
You could create a capture group for this I think?
Not sure if it's better for performance, or if it's better at all. But this would work I think:
^(0+)[1-9]([0-9]+)?$
There was a problem hiding this comment.
Actually the regex probably should use $ to avoid stripping leading zeros on "01.0" for instance
There was a problem hiding this comment.
Moved from the regex to String::parse<i32> as it changes -0100 to -100.
There was a problem hiding this comment.
I'll need to add a similar fix to credx. TBH I wasn't sure why it was trimming zeroes there when I came across it.
Related: hyperledger/indy-shared-rs#43
berendsliedrecht
force-pushed
the
support-0-value-in-credential
branch
from
11a3f0c
to
556fa9f
Compare
| assert_eq!(normalize_encoded_attr("01.0"), "01.0"); | ||
| assert_eq!(normalize_encoded_attr("0abc"), "0abc"); | ||
| assert_eq!(normalize_encoded_attr("-100"), "-100"); | ||
| assert_eq!(normalize_encoded_attr("-0100"), "-100"); |
There was a problem hiding this comment.
Would you expect this to have the leading 0 trimmed?
We need to make sure the spec aligns with this behavior
There was a problem hiding this comment.
The spec mentions convert any string integer (e.g. "1234") to be a 32-bit integer (e.g. 1234). Since integers can be negative, I thought this was the correct way. Maybe @swcurran can offer some clarification on this?
Signed-off-by: Berend Sliedrecht <blu3beri@proton.me>
berendsliedrecht
force-pushed
the
support-0-value-in-credential
branch
from
556fa9f
to
0f84ead
Compare
closes #252
Minor bug where a value contains a "0" without having any digits behind it.
Updated the regex to check for leading zeroes which must be followed by another digit. If this is the case, we trim the zeroes and return the leftover. If it is something like
0011212ababit does not strip the zeroes.https://github.com/hyperledger/aries-rfcs/tree/main/features/0592-indy-attachments#notes-on-encoding-claims