Prost-build generates i32 for enum fields instead of using the enum type directly in message

[immno]

I have a .proto file:

enum AssetAttrAggregationSourceType {
    AttrMeasureDacoo = 0;
    AttrTask = 1;
    AttrMeasurePipeline = 3;
}

message AssetAttrAggregationCreateEvent {
    int64 id = 1;
    int64 sourceId = 2;
    AssetAttrAggregationSourceType sourceType = 3;
    optional int64 associationId = 4;
}

/// build.rs
/// prost = "0.13"
/// prost-types = "0.13"
/// prost-build = "0.13"
fn main() {
    let mut config = prost_build::Config::new();
    config
        .out_dir("src/queue/msg")
        .compile_protos(&["cluster_event.proto"], &["proto/"])
        .unwrap();
}

After using prost-build to generate the code, it looks like this:

#[derive(Clone, Copy, Debug, PartialEq, Eq, Hash, PartialOrd, Ord, ::prost::Enumeration)]
#[repr(i32)]
pub enum AssetAttrAggregationSourceType {
    AttrMeasureDacoo = 0,
    AttrTask = 1,
    AttrMeasurePipeline = 3,
}

#[derive(Clone, PartialEq, ::prost::Message)]
pub struct AssetAttrAggregationCreateEvent {
    #[prost(int64, tag = "1")]
    pub id: i64,
    #[prost(int64, tag = "2")]
    pub source_id: i64,
    #[prost(enumeration = "AssetAttrAggregationSourceType", tag = "3")]
    pub source_type: i32,
    #[prost(int64, optional, tag = "4")]
    pub association_id: ::core::option::Option<i64>
}

Why can't we generate AssetAttrAggregationCreateEvent like the following (Oneof is fine, but it's inconvenient to use now, and we need to convert From every time):

#[derive(Clone, PartialEq, ::prost::Message)]
pub struct AssetAttrAggregationCreateEvent {
    #[prost(int64, tag = "1")]
    pub id: i64,
    #[prost(int64, tag = "2")]
    pub source_id: i64,
    #[prost(enumeration = "AssetAttrAggregationSourceType", tag = "3")]
    pub source_type: AssetAttrAggregationSourceType,
    #[prost(int64, optional, tag = "4")]
    pub association_id: ::core::option::Option<i64>
}

[caspermeijn]

I agree that would be more convenient. However, that is a breaking change and it requires people to work on it.

I am closing as a duplicate of: #276

If you are interested in working on a solution, you can try to contribute to this PR: #1079