You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Observed: When I open https://github.com/FasterXML/jackson-databind/blob/master/release-notes/VERSION-2.x , I don't see the substring 2348 anywhere.
Expected: The 2.x branch and master branch update the release notes for 2.10(.0). Mention that the helper readValue will start throwing when null is passed instead of a string (it previously returned null).
I notice that readValue will start throwing in 2.10 for code such as the below, if jsonUsers is the value null (not the string "null"). The application in question doesn't expect null as an actual input, but checked for it anyway as a sanity check. I discovered this in an integration test of handling invalid inputs.
I agree with the fix but want it documented.
// The assertion causing the exceptionpublic <T> TreadValue(Stringcontent, TypeReference<T> valueTypeRef)
throwsJsonProcessingException, JsonMappingException
{
_assertNotNull("content", content);
java.lang.IllegalArgumentException: argument "content" is null
at com.fasterxml.jackson.databind.ObjectMapper._assertNotNull(ObjectMapper.java:4413)
at com.fasterxml.jackson.databind.ObjectMapper.readValue(ObjectMapper.java:3187)
I didn't see anything else in github issues mentioning this or linking to that issue.
The text was updated successfully, but these errors were encountered:
603b395 (Fix #2348) is part of the 2.10 release.
Observed: When I open https://github.com/FasterXML/jackson-databind/blob/master/release-notes/VERSION-2.x , I don't see the substring 2348 anywhere.
Expected: The 2.x branch and master branch update the release notes for 2.10(.0). Mention that the helper readValue will start throwing when
null
is passed instead of a string (it previously returnednull
).I notice that readValue will start throwing in 2.10 for code such as the below, if
jsonUsers
is the valuenull
(not the string"null"
). The application in question doesn't expect null as an actual input, but checked for it anyway as a sanity check. I discovered this in an integration test of handling invalid inputs.I didn't see anything else in github issues mentioning this or linking to that issue.
The text was updated successfully, but these errors were encountered: