Support for MapStruct SPI

[coney]

My DTOs are auto generated. each field is generated with an setter(setXXX) and chained accessor(withXXX). I wrote a customized AccessorNamingStrategy SPI to exclude those methods start with with and added the SPI to my project's preprocessor.

My SPI has successfully enlimited warnings of withXXX accessors. but in my IDEA, it still promotes the hits for adding withXXX mappings:

[91wzhang]

Having same issue when trying to use the protobuf SPI

[filiphr]

@coney This ia a tricky one to be honest. There is no way for us to invoke the SPI from within the plugin. What we could offer perhaps is a way for people to somehow configure how certain methods need to be ignored.

@91wzhang supporting the protobuf SPI is perhaps that we need to look into independently. I would suggest creating a separate issue for that

[coney]

@coney This ia a tricky one to be honest. There is no way for us to invoke the SPI from within the plugin. What we could offer perhaps is a way for people to somehow configure how certain methods need to be ignored.

@91wzhang supporting the protobuf SPI is perhaps that we need to look into independently. I would suggest creating a separate issue for that

Great, maybe a simple configurable regex pattern is enough for my scenario

[filiphr]

Great, maybe a simple configurable regex pattern is enough for my scenario

That is an interesting idea @coney. We can look into that one.

[filiphr]

@coney would you be interested in providing this functionality to the plugin?

[jomu78]

@filiphr : I gave it a first try - see #150

[filiphr]

Thanks @jomu78. I've added a response on the PR in #150 (comment)

[kruijs-bbraun]

We are considering to use https://github.com/entur/mapstruct-spi-protobuf like @91wzhang and would require a solution for this issue #106

the mapstruct protobuf processor works fine in maven builds, but emits errors in intellij

is anyone working on a solution?

[filiphr]

@kruijs-bbraun I don't think that anyone is working on a solution for this. My old comment (#91 (comment)) still stands.

[kruijs-bbraun]

@kruijs-bbraun I don't think that anyone is working on a solution for this. My old comment (#91 (comment)) still stands.

okay, thanks for the heads up. I think it might as well have been an avoidable misconfiguration on my side.