Fixed problem in CollectionDeserializer where _nullProvider was not used if it was not JsonToken.VALUE_NULL
#5140
Conversation
src/main/java/com/fasterxml/jackson/databind/deser/std/CollectionDeserializer.java
Outdated
Show resolved
Hide resolved
src/main/java/com/fasterxml/jackson/databind/deser/std/CollectionDeserializer.java
Show resolved
Hide resolved
There was a problem hiding this comment.
Is the name, placement and content of the test appropriate?
The following test was referenced for name and placement.
JooHyukKim@51a92b9
There was a problem hiding this comment.
Should go next to CollectionDeserTest in com.fasterxml.jackson.databind.deser.jdk. Name is fine.
| @Test | ||
| public void nullsFailTest() { | ||
| ObjectMapper mapper = new ObjectMapper(); | ||
| mapper.configOverride(List.class).setSetterInfo(JsonSetter.Value.forContentNulls(Nulls.FAIL)); |
There was a problem hiding this comment.
Let's use JsonMapper.builder() approach so it merges more easily into 3.x (3.x has immutable ObjectMapper).
Something like
mapper = JsonMapper.builder()
.defaultSetterInfo(JsonSetter.Value.forContentNulls(Nulls.FAIL))
.build();
There was a problem hiding this comment.
Oh, I am sorry.
There was a mistake and I have made an additional correction.
88bc7e4
src/test/java/com/fasterxml/jackson/databind/deser/std/CollectionDeserializer5139Test.java
Outdated
Show resolved
Hide resolved
src/main/java/com/fasterxml/jackson/databind/deser/std/CollectionDeserializer.java
Outdated
Show resolved
Hide resolved
| } | ||
|
|
||
| // _skipNullValues is checked by _tryToAddNull. | ||
| if (value == null) { |
There was a problem hiding this comment.
This part can be moved within previous block, after value = _nullProvider.getNullValue(ctxt);
src/main/java/com/fasterxml/jackson/databind/deser/std/CollectionDeserializer.java
Show resolved
Hide resolved
| * This method only performs deserialization and does not consider _skipNullValues, _nullProvider, etc. | ||
| * @since 2.19.1 | ||
| */ | ||
| protected Object _deserializeRawContentValue( |
There was a problem hiding this comment.
Method makes sense but I think I'll rename it and can then probably merge
|
Excellent work as usual. Thank you @k163377 -- merged in for 2.19(.1), will merge to 3.x Bit too risky to backport in 2.18, unfortunately, I think. |
|
Thank you very much. |
|
Is there an estimate of when 2.19.1 might be released? We are eagerly awaiting for this fix to be released. Thanks! |
|
@DavidRigglemanININ Unfortunately due to effort involved (3-4 hours for full patch release) it'll probably be couple of weeks yet. I need to find time to do that (not being part of my day job). |
|
Yes, with lots going on at the moment, you might want to use and overwrite with latest jar? (not an offical release, but with fix you need. Using Gradle or Maven I guess. --In case you re urgent cc @DavidRigglemanININ . |
|
I can wait a few weeks. It's not that urgent as we've been living for the pain for quite a while now (as we couldn't use the strictNullChecks on the kotlin module side due to performance issues). It's just been a pain point for us over the years that I'm really excited to finally get behind us as you can probably tell :). Thanks for all your hard work on this project! |
|
@DavidRigglemanININ Thank you! With Sonatype's aggressive push to shut down OSSRH I may end up actually making release earlier, although first needing to spend a bit of time on migrating publishing to use Sonatype Central Portal. We'll see. |
Are we any closer to a release in the next few weeks? |
|
@DavidRigglemanININ I think so -- I do want to get 2.19.1 out before I go on my vacation in about 2 weeks. |
4 participants
SSIA
fix for #5139
It was a bit complicated, so I apologize if there are any problems with the modifications.