it would make the code easier to build: no protoc to run, no pre-compiled library to link
it would make the build system easier to maintain and easier to update deps
it would make the code potentially higher performance
The downside to protozero is that it makes it slightly more work when you want to update the protobuf scheme. However, I think this is a fine tradeoff since the schema here for glyphs has not changed frequently so far.
This in my mind is low urgency, but a nice to have and a good starter task for someone interested in learning and using protozero more.
/cc for visibility @kkaefer @ChrisLoer @mapbox/core-tech
We should port the code to protozero because:
The downside to protozero is that it makes it slightly more work when you want to update the protobuf scheme. However, I think this is a fine tradeoff since the schema here for glyphs has not changed frequently so far.
This in my mind is low urgency, but a nice to have and a good starter task for someone interested in learning and using protozero more.
/cc for visibility @kkaefer @ChrisLoer @mapbox/core-tech