fix: throw error if hexToBytes or hexToString is provided a string that is not in hex
#2657
Conversation
| @@ -15,25 +15,6 @@ | |||
| * Eliminates 4 runtime dependencies: `base-x`, `base64-js`, `buffer`, and `ieee754`. | |||
| * Execute test in a browser in addition to node | |||
|
|
|||
| ## 5.0.0 Beta 1 (2023-11-30) | |||
There was a problem hiding this comment.
This still feels relevant. There were releases. It also shows we released betas.
There was a problem hiding this comment.
It's not useful in the changelog IMO - it's already in the release list.
There was a problem hiding this comment.
I personally find negative value in having beta info in changelogs, even when looking at other packages. I think it just clutters up the file. But if you disagree I'm happy to roll back this part of the change.
| }) | ||
|
|
||
| it('bytesToHex - bad hex', () => { | ||
| expect(() => hexToBytes('hello')).toThrow(new Error('Invalid hex string')) |
There was a problem hiding this comment.
Should this be a special Error that can be more easily detected?
There was a problem hiding this comment.
this is in the unit tests, so I don't expect a special error code to be useful here. On the other hand, I feel the main code base could make use of a special error code / error status message
There was a problem hiding this comment.
Yeah that is what I mean.
There was a problem hiding this comment.
I checked what errors were used in the rest of the package and it was just Error, so I just copied the norm
There was a problem hiding this comment.
hmm. Assigning special error codes (for all errors) might be a significant deviation from the intent of the PR. I'm okay if it's not included as a part of this change list
mvadari
requested review from
ckeshava,
jonathanlei and
pdp2121
and removed request for
JST5000
| @@ -1,5 +1,7 @@ | |||
| import { concatBytes } from '@noble/hashes/utils' | |||
|
|
|||
| export const HEX_REGEX = /^[A-F0-9]*$/iu | |||
There was a problem hiding this comment.
Do we want to change from * to + to prevent empty strings? @mvadari
There was a problem hiding this comment.
I purposefully kept support for empty strings - I don't see any reason why these functions shouldn't support empty strings, which would convert to an empty bytestring or empty ASCII string.
There was a problem hiding this comment.
Yes, but in what cases would an empty string be a legitimate input?
There was a problem hiding this comment.
One example: your app has a field that accepts memos and runs the result through convertHexToString
…g that is not in hex (XRPLF#2657) * better error handling * fix browser tests * add shared variable * re-add test case
…g that is not in hex (XRPLF#2657) * better error handling * fix browser tests * add shared variable * re-add test case
High Level Overview of Change
Title says it all.
Context of Change
The current methods return an empty bytes object/string if they are not provided a string in hex, which caused an issue.
Type of Change
Did you update
HISTORY.md?I also cleaned up the other
HISTORY.mdfiles.Test Plan
Added tests, they pass.