Updated the codecs used with latest changes [CTT-713]

[ihsandemir]

Updated the codecs used with latest changes. Also, updated the code to use these codecs properly with the new introduced parameters.

Associated protocol PR is hazelcast/hazelcast-client-protocol#557

[JackPGreen]

Updated the codecs used with latest changes. Also, updated the code to use these codecs properly with the new introduced parameters.

Could you try to keep the PRs separate, if possible, please?

I.E. are these changes only the autogenerated output of hazelcast/hazelcast-client-protocol#557? or are there additional changes as well?

[ihsandemir]

Updated the codecs used with latest changes. Also, updated the code to use these codecs properly with the new introduced parameters.

Could you try to keep the PRs separate, if possible, please?

I.E. are these changes only the autogenerated output of hazelcast/hazelcast-client-protocol#557? or are there additional changes as well?

No, they are related and needed. I can not separate them.

[JackPGreen]

Updated the codecs used with latest changes. Also, updated the code to use these codecs properly with the new introduced parameters.

Could you try to keep the PRs separate, if possible, please?
I.E. are these changes only the autogenerated output of hazelcast/hazelcast-client-protocol#557? or are there additional changes as well?

No, they are related and needed. I can not separate them.

Ok, I've tried splitting it (locally to understand it) but I think there's a problem - the generated client-protocol from hazelcast/hazelcast-client-protocol#557 doesn't match the generated-sources in this PR.

You can see my working in this branch.

[ihsandemir]

Updated the codecs used with latest changes. Also, updated the code to use these codecs properly with the new introduced parameters.

Could you try to keep the PRs separate, if possible, please?
I.E. are these changes only the autogenerated output of hazelcast/hazelcast-client-protocol#557? or are there additional changes as well?

No, they are related and needed. I can not separate them.

Ok, I've tried splitting it (locally to understand it) but I think there's a problem - the generated client-protocol from hazelcast/hazelcast-client-protocol#557 doesn't match the generated-sources in this PR.

You can see my working in this branch.

I generate and format the code. Is the problem about formatting(clang format) difference or do you see other differences?

[JackPGreen]

I generate and format the code. Is the problem about formatting(clang format) {...}?

I think this makes it very challenging to review - if it's generated code, as long as I'm happy with the generator configuration (i.e. hazelcast/hazelcast-client-protocol#557) and I can validate the code is generated as it should be, I can simply rubber-stamp it.

In this scenario, I can't validate the code is generated as it should be, so instead I need to review the generator configuration and review the generated code as well.

I tried to update the client-protocol template formatting in https://github.com/JackPGreen/hazelcast-client-protocol/tree/clang-formatting to see if I could produce the output heree "out-of-the-box" but it's not (easily) possible as the formatter is line breaking in different positions due to the line length (i.e. a parameter may be declared on the same line as a function or a newline depending on the name of the method - which isn't easy to implement in a template).

I think it would be better simply to leave the formatting of the generated code as-is.

do you see other differences?

I don't think so.

[ihsandemir]

I generate and format the code. Is the problem about formatting(clang format) {...}?

I think this makes it very challenging to review - if it's generated code, as long as I'm happy with the generator configuration (i.e. hazelcast/hazelcast-client-protocol#557) and I can validate the code is generated as it should be, I can simply rubber-stamp it.

In this scenario, I can't validate the code is generated as it should be, so instead I need to review the generator configuration and review the generated code as well.

I tried to update the client-protocol template formatting in https://github.com/JackPGreen/hazelcast-client-protocol/tree/clang-formatting to see if I could produce the output heree "out-of-the-box" but it's not (easily) possible as the formatter is line breaking in different positions due to the line length (i.e. a parameter may be declared on the same line as a function or a newline depending on the name of the method - which isn't easy to implement in a template).

I think it would be better simply to leave the formatting of the generated code as-is.

do you see other differences?

I don't think so.

We can integrate clang-format into the generator. the commands is similar to clang-format -i --verbose hazelcast/generated-sources/src/hazelcast/client/protocol/codec/codecs.* but then we need to find how to pass the .clamg-format file in cpp repo to this command at protocol repo, Well we can just copy for the start.

[JackPGreen]

We can integrate clang-format into the generator. the commands is similar to clang-format -i --verbose hazelcast/generated-sources/src/hazelcast/client/protocol/codec/codecs.*

🤦

Good point - it's much simpler to actually use the formatter than to try to approximate it.

[ihsandemir]

We can integrate clang-format into the generator. the commands is similar to clang-format -i --verbose hazelcast/generated-sources/src/hazelcast/client/protocol/codec/codecs.*

🤦

Good point - it's much simpler to actually use the formatter than to try to approximate it.

Sent the PR for this hazelcast/hazelcast-client-protocol#558