Class: FluentExt::TextParser

Inherits:
Object
  • Object
show all
Defined in:
lib/fluent/plugin/fixed_parser.rb

Defined Under Namespace

Classes: ApacheParser, CSVParser, GenericParser, JSONParser, LabeledTSVParser, RegexpParser, TSVParser, ValuesParser

Constant Summary collapse

TEMPLATE_FACTORIES =
{
  'apache' => Proc.new { RegexpParser.new(/^(?<host>[^ ]*) [^ ]* (?<user>[^ ]*) \[(?<time>[^\]]*)\] "(?<method>\S+)(?: +(?<path>[^ ]*) +\S*)?" (?<code>[^ ]*) (?<size>[^ ]*)(?: "(?<referer>[^\"]*)" "(?<agent>[^\"]*)")?$/, {'time_format'=>"%d/%b/%Y:%H:%M:%S %z"}) },
  'apache2' => Proc.new { ApacheParser.new },
  'nginx' => Proc.new { RegexpParser.new(/^(?<remote>[^ ]*) (?<host>[^ ]*) (?<user>[^ ]*) \[(?<time>[^\]]*)\] "(?<method>\S+)(?: +(?<path>[^ ]*) +\S*)?" (?<code>[^ ]*) (?<size>[^ ]*)(?: "(?<referer>[^\"]*)" "(?<agent>[^\"]*)")?$/,  {'time_format'=>"%d/%b/%Y:%H:%M:%S %z"}) },
  'syslog' => Proc.new { RegexpParser.new(/^(?<time>[^ ]*\s*[^ ]* [^ ]*) (?<host>[^ ]*) (?<ident>[a-zA-Z0-9_\/\.\-]*)(?:\[(?<pid>[0-9]+)\])?[^\:]*\: *(?<message>.*)$/, {'time_format'=>"%b %d %H:%M:%S"}) },
  'json' => Proc.new { JSONParser.new },
  'csv' => Proc.new { CSVParser.new },
  'tsv' => Proc.new { TSVParser.new },
  'ltsv' => Proc.new { LabeledTSVParser.new },
}

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initializeTextParser

Returns a new instance of TextParser.


212
213
214
# File 'lib/fluent/plugin/fixed_parser.rb', line 212

def initialize
  @parser = nil
end

Class Method Details

.register_template(name, regexp_or_proc, time_format = nil) ⇒ Object


201
202
203
204
205
206
207
208
209
210
# File 'lib/fluent/plugin/fixed_parser.rb', line 201

def self.register_template(name, regexp_or_proc, time_format=nil)

  factory = if regexp_or_proc.is_a?(Regexp)
              regexp = regexp_or_proc
              Proc.new { RegexpParser.new(regexp, {'time_format'=>time_format}) }
            else
              Proc.new { proc }
            end
  TEMPLATE_FACTORIES[name] = factory
end

Instance Method Details

#configure(conf, required = true) ⇒ Object


216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
# File 'lib/fluent/plugin/fixed_parser.rb', line 216

def configure(conf, required=true)
  format = conf['format']

  if format == nil
    if required
      raise Fluent::ConfigError, "'format' parameter is required"
    else
      return nil
    end
  end

  if format[0] == ?/ && format[format.length-1] == ?/
    # regexp
    begin
      regexp = Regexp.new(format[1..-2])
      if regexp.named_captures.empty?
        raise "No named captures"
      end
    rescue
      raise Fluent::ConfigError, "Invalid regexp '#{format[1..-2]}': #{$!}"
    end
    @parser = RegexpParser.new(regexp)

  else
    # built-in template
    factory = TEMPLATE_FACTORIES[format]
    unless factory
      raise Fluent::ConfigError, "Unknown format template '#{format}'"
    end
    @parser = factory.call

  end

  if @parser.respond_to?(:configure)
    @parser.configure(conf)
  end

  return true
end

#parse(text) ⇒ Object


256
257
258
# File 'lib/fluent/plugin/fixed_parser.rb', line 256

def parse(text)
  return @parser.call(text)
end