- fixes a bug where no content codes would be considered structured

[baywet]

fixes #4190

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[andrueastman]

Just to confirm @baywet , should we follow up with issues in the http libs to validate as we may have situations where 3xx are considered as not success codes and this change will still throw an error?

• Java - https://github.com/microsoft/kiota-java/blob/59a86b031985c34a8f9bff88c9af2ce8eb55c354/components/http/okHttp/src/main/java/com/microsoft/kiota/http/OkHttpRequestAdapter.java#L611
• Dotnet - https://github.com/microsoft/kiota-http-dotnet/blob/d960c5fa7f0661ea294b6962be105890a27fe999/src/HttpClientRequestAdapter.cs#L385
• Go (Looks good) - https://github.com/microsoft/kiota-http-go/blob/7f1b45b9907b402c56e1b1cacd02d96227e70da4/nethttp_request_adapter.go#L746
• TS - https://github.com/microsoft/kiota-typescript/blob/985036391e8a9823e363f5c2fde486ae9186ea0a/packages/http/fetch/src/fetchRequestAdapter.ts#L387C9-L387C20

[baywet]

@andrueastman yes, thanks for the nudge, my plan was to create follow up issues in the http repos.
Effectively there are two major cases:

• 3XX response with a location header: the redirect middleware will follow the location value, and return the other response. Should work as is.
• 3XX response without a location header: we'll get the immediate response, and depending on what IsSuccessResponse (or equivalent) evaluates, we might throw when we shouldn't today. This should be fixed in the implementations.