I can't really give a definitive answer without seeing how the serializer/deserializer is implemented, but my initial suspicion is related to field name transformations.
GWT performs quite a bit of renaming and optimization during compilation, and domino-jackson works by mapping the original property string to the renamed field. Since the compiler renames the field consistently everywhere it's used, the mapping still works.
TeaVM, however, might be doing something different—possibly renaming fields inconsistently between the POJO and the classes that reference it. I'm not familiar enough with TeaVM's internals to say for sure.
The fact that explicitly specifying the property name in the annotation solves the issue makes the behavior even more puzzling without seeing the relevant code.
Works great for me in GWT so far, trying it in TeaVM now, so not sure if this issue is in scope ?TeaVM at the minimum does a bit of name changing and domino-jackson deserializer fails with my DTO. One suggestion is add basic Jackson annotations to every field in the DTO I think it's @JsonProperty, then domino-jackson will adapt the deserializer. Anyone know if that's correct ? Thanks
You received this message because you are subscribed to the Google Groups "GWT Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to google-web-toolkit+unsubscribe@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/google-web-toolkit/51a1ad21-3ed6-44a7-8cd2-c9040af91266n%40googlegroups.com.
No comments:
Post a Comment