Single File Input (new line delimited)

todo: when more than line splitting is needed the file should be read in chunks and passed to a generic splitter buffer with a token/match specification and a find function similar to the Heka stream reader.

1. Sample Configuration

filename = "file.lua"

-- Name of the input file (nil for stdin)
-- Default:
-- input_filename = nil

-- Multi-line log support
-- delimiter = "^# User@Host:" -- optional if anchored at the beginning of the line it is treated as
                               -- a start of record delimiter and the line belongs to the next log
                               -- otherwise it is an end of record delimiter and the line belongs to
                               -- the current log. "^$" is special cased and is treated as an end of
                               -- line delimiter.

-- Heka message table containing the default header values to use, if they are
-- not populated by the decoder. If 'Fields' is specified it should be in the
-- hashed based format see:
-- Default:
-- default_headers = nil

-- printf_messages = -- see:

-- Specifies a module that will decode the raw data and inject the resulting message.
-- Supports the same syntax as an individual sub decoder
-- see:
-- Default:
-- decoder_module = "decoders.payload"

-- Boolean, if true, any decode failure will inject a  message of Type "error",
-- with the Payload containing the error.
-- Default:
-- send_decode_failures = false

source code: file.lua

results matching ""

    No results matching ""