Re: [PATCH] Log CSV by default

Le 04/12/2018 à 03:37, Michael Paquier a écrit :
> On Mon, Dec 03, 2018 at 09:20:01PM +0100, Gilles Darold wrote:
>> pgBadger is able to parse jsonlog
>> (https://github.com/michaelpq/pg_plugins/tree/master/jsonlog) output in
>> multi-process mode because it do not use an external library to parse
>> the json. In this minimalist implementation I assume that each log line
>> is a full json object. Honestly I do not have enough feedback on this
>> format to be sure that my basic jsonlog parser is enough, maybe it is
>> not. What is really annoying for a log parser is the preservation of
>> newline character in queries, if the format outputs each atomic log
>> messages on a single line each, this is ideal.
> jsonlog generates one JSON object for each elog call, and per line.
> Newlines are treated as they would for a JSON object by being replaced
> with \n within the query strings, the same way te backend handles JSON
> strings as text.

This was what I remember and this is why it is easy for pgbadger to
process these log files in multiprocess mode.

--
Gilles Darold