Hello, At my current company we're heavy protobuf users, but we started facing a big limitation: in the current Java implementation, since there's no lazy parsing, or ability to do direct sub-message parsing, we pay a huge penalty of time and, especially, memory (we have an articulated schema and the memory footprint is often 5X+ the serialized size) even when we just want to read a small subset of fields of a big protobuf (~10-100 KB).
I understand that, according to [1] and [2], such feature has not yet been released and it will come at some point, but nonetheless I'd like to ping again to see if there's a more defined ETA (of course I realize this is an open source project so support and roadmap are shared at a best-effort). I am aware that there are projects (like flatbuffers) that might better suited for our use case, but we would really like to stick to protobuf because we have a huge code base already working on that, and we consider it the most stable and robust solution available. Thanks. [1] https://groups.google.com/forum/#!searchin/protobuf/lazy|sort:date/protobuf/2P4ge8MfYKw/dcsmXDPaE84J [2] https://groups.google.com/forum/#!searchin/protobuf/lazy|sort:date/protobuf/G9Jhv_xrxqg/tYcXzyy6zV8J -- You received this message because you are subscribed to the Google Groups "Protocol Buffers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/protobuf. For more options, visit https://groups.google.com/d/optout.
