thrift-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Neumark <neumark.pe...@gmail.com>
Subject Re: Why does Binary/Compact protocol write field types to the bytestream?
Date Mon, 14 Oct 2013 13:13:17 GMT
Thanks for the quick reply!


On Mon, Oct 14, 2013 at 2:19 PM, Bill Dortch <bill.dortch@gmail.com> wrote:

> That's so receivers that don't know about a field at least know how to
> skip over it. One of the key benefits of Thrift is providing
> interoperability over different schema versions, so that it's not necessary
> to update all clients immediately when new fields are added to structs, and
> so on.
>
> -Bill
>
>
> On Mon, Oct 14, 2013 at 3:34 AM, Peter Neumark <neumark.peter@gmail.com>wrote:
>
>> Does anybody know why protocol.writeFieldBegin() always writes the
>> following value's type to the bytestream? I know this is also done by
>> protobuf, but I think writing only the field's ID should be enough (as
>> long
>> as both sender and receiver have the same idea of the type of the field).
>>
>> Any explanation would be appreciated!
>>
>> Thanks,
>> Peter
>>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message