From: | Bear Giles <bgiles(at)coyotesong(dot)com> |
---|---|
To: | Vladimir Sitnikov <sitnikov(dot)vladimir(at)gmail(dot)com> |
Cc: | Udit Juneja <uditjuneja1(at)gmail(dot)com>, pgsql-hackers(at)lists(dot)postgresql(dot)org |
Subject: | Re: [GSoC Idea Discussion] "Thrift datatype support" Project |
Date: | 2018-02-13 18:28:49 |
Message-ID: | CALBNtw50yWnE2rdZWvfrU3oK-pLbYic3aNZoEkRVRAGNQByCgQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Feb 13, 2018 at 10:24 AM, Vladimir Sitnikov <
sitnikov(dot)vladimir(at)gmail(dot)com> wrote:
> Technically speaking, Thrift is "language-independent"
> serialization-deserialization format with clean approach to backward
> compatibility.
>
> I think Thrift (or something like that) can be useful, as it can generate
> serializers/deserializers for lots of languages.
> PostgreSQL's "binary" format is tied to the PostgreSQL and it has to be
> reimplemented for each and every client language.
>
> "text" format is non-trivial as well (e.g. it is hard to get quoting right
> for structs, and text is server-locale-dependent)
>
> Vladimir
>
Thanks. I've been helping a coworker with an older parquet writer and
the URL for the hive metastore uses a thrift:// scheme. That is why I
thought it is a communications protocol.
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2018-02-13 18:43:40 | Re: JIT compiling with LLVM v9.0 |
Previous Message | Tom Lane | 2018-02-13 18:13:02 | Re: Converting plpgsql to use DTYPE_REC for named composite types |