Daniel G. Taylor

Results 335 comments of Daniel G. Taylor
trafficstars

@superstas I'm not sure how to solve this honestly. Maybe if there is additional validation on the struct field via tags then you just cannot use a `$ref` and must...

@bbedward thanks for the feedback! I'll see what can be done, but keep in mind the auto-generation is best-effort and won't work for all use-cases. You can always use `huma.Register`...

@pgr0ss thanks for opening an issue for this! It isn't actually a Huma thing, but has to do with how Golang serializes `time.Time` structs, while the `format` tag is just...

@stargeneration-winston take a look at https://huma.rocks/features/response-outputs/#body. You can use a type of `[]byte` for the body if the file fits in memory, otherwise you can use https://huma.rocks/features/response-streaming/ to get the...

Yes I think this would be nice, however we can't make major changes like this without breaking everyone. Perhaps it's time to start a list of things that would be...

@dennisgsmith thanks! This looks a lot like an OpenAPI 3 discriminated union: https://swagger.io/docs/specification/v3_0/data-models/inheritance-and-polymorphism/#discriminator. I'm not opposed to adding such a feature (we already support manually creating such a schema today,...

@coury-clark thanks for opening this issue. The `format` tag is intended to document and limit the input string value, while the `timeFormat` is intended to tell Go how to marshal/unmarshal...

@sjayach unfortunately no, not without building a custom JSON (un)marshaler (and supporting any other formats you need like CBOR if desired). You are better off accepting a string with a...

@benkawecki no, this isn't currently possible as the parsing happens after the middlewares have run and Huma doesn't publicly expose the param parsing logic. What specifically are you trying to...

@spa5k I'm not familiar with scs, but assuming you mean https://github.com/alexedwards/scs > the problem I'm having is that it's not considering the cookies/anything that is set after the next(ctx) is...