Open skeggse opened 3 months ago
Ok I should have looked closer at the input - I guess the AWS ECR API escapes the forward-slashes?
Maybe a realistic issue here is to make the error message a little clearer. I don't suppose there's a way to have it show the source bytes rather than the string after it's been had its escape sequences resolved?
I'm trying to deserialize an OCI manifest into some structs that have
&'a str
fields, andserde_json
is indicating that it can't do that for some specific values. I can fix the issue by usingCow<'a, str>
instead, but I'm kinda curious to understand why this is happening. I can understand why this would happen for escape sequences, but I don't seem to have any escape sequences in my JSON! Are there other reasons this might be happening?References: