fix unit tests for grok parser

docs/DATA_FORMATS_INPUT.md

@@ -9,6 +9,7 @@ Telegraf is able to parse the following input data formats into metrics:
 1. [Nagios](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#nagios) (exec input only)
 1. [Collectd](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#collectd)
 1. [Dropwizard](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#dropwizard)
+1. [Grok](https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md#grok)
 
 Telegraf metrics, like InfluxDB
 [points](https://docs.influxdata.com/influxdb/v0.10/write_protocols/line/),
@@ -651,5 +652,106 @@ For more information about the dropwizard json format see
   # [inputs.exec.dropwizard_tag_paths]
   #   tag1 = "tags.tag1"
   #   tag2 = "tags.tag2"
-
+```
+
+#### Grok
+Parse logstash-style "grok" patterns. Patterns can be added to patterns, or custom patterns read from custom_pattern_files.
+
+# View logstash grok pattern docs here:
+#   https://www.elastic.co/guide/en/logstash/current/plugins-filters-grok.html
+# All default logstash patterns are supported, these can be viewed here:
+#   https://github.com/logstash-plugins/logstash-patterns-core/blob/master/patterns/grok-patterns
+
+# Available modifiers:
+#   string   (default if nothing is specified)
+#   int
+#   float
+#   duration (ie, 5.23ms gets converted to int nanoseconds)
+#   tag      (converts the field into a tag)
+#   drop     (drops the field completely)
+# Timestamp modifiers:
+#   ts-ansic         ("Mon Jan _2 15:04:05 2006")
+#   ts-unix          ("Mon Jan _2 15:04:05 MST 2006")
+#   ts-ruby          ("Mon Jan 02 15:04:05 -0700 2006")
+#   ts-rfc822        ("02 Jan 06 15:04 MST")
+#   ts-rfc822z       ("02 Jan 06 15:04 -0700")
+#   ts-rfc850        ("Monday, 02-Jan-06 15:04:05 MST")
+#   ts-rfc1123       ("Mon, 02 Jan 2006 15:04:05 MST")
+#   ts-rfc1123z      ("Mon, 02 Jan 2006 15:04:05 -0700")
+#   ts-rfc3339       ("2006-01-02T15:04:05Z07:00")
+#   ts-rfc3339nano   ("2006-01-02T15:04:05.999999999Z07:00")
+#   ts-httpd         ("02/Jan/2006:15:04:05 -0700")
+#   ts-epoch         (seconds since unix epoch)
+#   ts-epochnano     (nanoseconds since unix epoch)
+#   ts-"CUSTOM"
+# CUSTOM time layouts must be within quotes and be the representation of the
+# "reference time", which is Mon Jan 2 15:04:05 -0700 MST 2006
+# See https://golang.org/pkg/time/#Parse for more details.
+
+# Example log file pattern, example log looks like this:
+#   [04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs
+# Breakdown of the DURATION pattern below:
+#   NUMBER  is a builtin logstash grok pattern matching float & int numbers.
+#   [nuµm]? is a regex specifying 0 or 1 of the characters within brackets.
+#   s       is also regex, this pattern must end in "s".
+# so DURATION will match something like '5.324ms' or '6.1µs' or '10s'
+DURATION %{NUMBER}[nuµm]?s
+RESPONSE_CODE %{NUMBER:response_code:tag}
+RESPONSE_TIME %{DURATION:response_time_ns:duration}
+EXAMPLE_LOG \[%{HTTPDATE:ts:ts-httpd}\] %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
+
+# Wider-ranging username matching vs. logstash built-in %{USER}
+NGUSERNAME [a-zA-Z0-9\.\@\-\+_%]+
+NGUSER %{NGUSERNAME}
+# Wider-ranging client IP matching
+CLIENT (?:%{IPORHOST}|%{HOSTPORT}|::1)
+
+##
+## COMMON LOG PATTERNS
+##
+
+# apache & nginx logs, this is also known as the "common log format"
+#   see https://en.wikipedia.org/wiki/Common_Log_Format
+COMMON_LOG_FORMAT %{CLIENT:client_ip} %{NOTSPACE:ident} %{NOTSPACE:auth} \[%{HTTPDATE:ts:ts-httpd}\] "(?:%{WORD:verb:tag} %{NOTSPACE:request}(?: HTTP/%{NUMBER:http_version:float})?|%{DATA})" %{NUMBER:resp_code:tag} (?:%{NUMBER:resp_bytes:int}|-)
+
+# Combined log format is the same as the common log format but with the addition
+# of two quoted strings at the end for "referrer" and "agent"
+#   See Examples at http://httpd.apache.org/docs/current/mod/mod_log_config.html
+COMBINED_LOG_FORMAT %{COMMON_LOG_FORMAT} %{QS:referrer} %{QS:agent}
+
+# HTTPD log formats
+HTTPD20_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{LOGLEVEL:loglevel:tag}\] (?:\[client %{IPORHOST:clientip}\] ){0,1}%{GREEDYDATA:errormsg}
+HTTPD24_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{WORD:module}:%{LOGLEVEL:loglevel:tag}\] \[pid %{POSINT:pid:int}:tid %{NUMBER:tid:int}\]( \(%{POSINT:proxy_errorcode:int}\)%{DATA:proxy_errormessage}:)?( \[client %{IPORHOST:client}:%{POSINT:clientport}\])? %{DATA:errorcode}: %{GREEDYDATA:message}
+HTTPD_ERRORLOG %{HTTPD20_ERRORLOG}|%{HTTPD24_ERRORLOG}
+
+#### Grok Configuration:
+```toml
+[[inputs.reader]]
+  ## This is a list of patterns to check the given log file(s) for.
+  ## Note that adding patterns here increases processing time. The most
+  ## efficient configuration is to have one pattern per logparser.
+  ## Other common built-in patterns are:
+  ##   %{COMMON_LOG_FORMAT}   (plain apache & nginx access logs)
+  ##   %{COMBINED_LOG_FORMAT} (access logs + referrer & agent)
+  grok_patterns = ["%{COMBINED_LOG_FORMAT}"]
+
+  ## Name of the outputted measurement name.
+  grok_name_override = "apache_access_log"
+
+  ## Full path(s) to custom pattern files.
+  grok_custom_pattern_files = []
+
+  ## Custom patterns can also be defined here. Put one pattern per line.
+  grok_custom_patterns = '''
+
+  ## Timezone allows you to provide an override for timestamps that
+  ## don't already include an offset
+  ## e.g. 04/06/2016 12:41:45 data one two 5.43µs
+  ##
+  ## Default: "" which renders UTC
+  ## Options are as follows:
+  ##   1. Local             -- interpret based on machine localtime
+  ##   2. "Canada/Eastern"  -- Unix TZ values like those found in https://en.wikipedia.org/wiki/List_of_tz_database_time_zones
+  ##   3. UTC               -- or blank/unspecified, will return timestamp in UTC
+  grok_timezone = "Canada/Eastern"
 ```

internal/config/config.go

@@ -1338,6 +1338,59 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
 		}
 	}
 
+	//for grok data_format
+	if node, ok := tbl.Fields["grok_named_patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.NamedPatterns = append(c.NamedPatterns, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["grok_patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.Patterns = append(c.Patterns, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["grok_custom_patterns"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if str, ok := kv.Value.(*ast.String); ok {
+				c.CustomPatterns = str.Value
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["grok_custom_pattern_files"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if ary, ok := kv.Value.(*ast.Array); ok {
+				for _, elem := range ary.Value {
+					if str, ok := elem.(*ast.String); ok {
+						c.CustomPatternFiles = append(c.CustomPatternFiles, str.Value)
+					}
+				}
+			}
+		}
+	}
+
+	if node, ok := tbl.Fields["grok_timezone"]; ok {
+		if kv, ok := node.(*ast.KeyValue); ok {
+			if str, ok := kv.Value.(*ast.String); ok {
+				c.TimeZone = str.Value
+			}
+		}
+	}
+
 	c.MetricName = name
 
 	delete(tbl.Fields, "data_format")
@@ -1353,6 +1406,11 @@ func buildParser(name string, tbl *ast.Table) (parsers.Parser, error) {
 	delete(tbl.Fields, "dropwizard_time_format")
 	delete(tbl.Fields, "dropwizard_tags_path")
 	delete(tbl.Fields, "dropwizard_tag_paths")
+	delete(tbl.Fields, "grok_named_patterns")
+	delete(tbl.Fields, "grok_patterns")
+	delete(tbl.Fields, "grok_custom_patterns")
+	delete(tbl.Fields, "grok_custom_pattern_files")
+	delete(tbl.Fields, "grok_timezone")
 
 	return parsers.NewParser(c)
 }

plugins/inputs/all/all.go

@@ -85,6 +85,7 @@ import (
 	_ "github.com/influxdata/telegraf/plugins/inputs/puppetagent"
 	_ "github.com/influxdata/telegraf/plugins/inputs/rabbitmq"
 	_ "github.com/influxdata/telegraf/plugins/inputs/raindrops"
+	_ "github.com/influxdata/telegraf/plugins/inputs/reader"
 	_ "github.com/influxdata/telegraf/plugins/inputs/redis"
 	_ "github.com/influxdata/telegraf/plugins/inputs/rethinkdb"
 	_ "github.com/influxdata/telegraf/plugins/inputs/riak"

plugins/inputs/logparser/logparser.go

@@ -14,9 +14,8 @@ import (
 	"github.com/influxdata/telegraf"
 	"github.com/influxdata/telegraf/internal/globpath"
 	"github.com/influxdata/telegraf/plugins/inputs"
-
+	"github.com/influxdata/telegraf/plugins/parsers"
 	// Parsers
-	"github.com/influxdata/telegraf/plugins/inputs/logparser/grok"
 )
 
 const (
@@ -39,6 +38,7 @@ type LogParserPlugin struct {
 	Files           []string
 	FromBeginning   bool
 	WatchMethod     string
+	MeasurementName string `toml:"measurement"`
 
 	tailers map[string]*tail.Tail
 	lines   chan logEntry
@@ -49,7 +49,13 @@ type LogParserPlugin struct {
 
 	sync.Mutex
 
-	GrokParser *grok.Parser `toml:"grok"`
+	GrokParser parsers.Parser `toml:"grok"`
+
+	Patterns           []string
+	NamedPatterns      []string
+	CustomPatterns     string
+	CustomPatternFiles []string
+	TimeZone           string
 }
 
 const sampleConfig = `
@@ -132,6 +138,21 @@ func (l *LogParserPlugin) Start(acc telegraf.Accumulator) error {
 
 	// Looks for fields which implement LogParser interface
 	l.parsers = []LogParser{}
+	config := &parsers.Config{
+		Patterns:           l.Patterns,
+		NamedPatterns:      l.NamedPatterns,
+		CustomPatterns:     l.CustomPatterns,
+		CustomPatternFiles: l.CustomPatternFiles,
+		TimeZone:           l.TimeZone,
+		DataFormat:         "grok",
+	}
+
+	var err error
+	l.GrokParser, err = parsers.NewParser(config)
+	if err != nil {
+		return err
+	}
+
 	s := reflect.ValueOf(l).Elem()
 	for i := 0; i < s.NumField(); i++ {
 		f := s.Field(i)
@@ -152,13 +173,6 @@ func (l *LogParserPlugin) Start(acc telegraf.Accumulator) error {
 		return fmt.Errorf("logparser input plugin: no parser defined")
 	}
 
-	// compile log parser patterns:
-	for _, parser := range l.parsers {
-		if err := parser.Compile(); err != nil {
-			return err
-		}
-	}
-
 	l.wg.Add(1)
 	go l.parser()
 
@@ -247,8 +261,8 @@ func (l *LogParserPlugin) receiver(tailer *tail.Tail) {
 	}
 }
 
-// parser is launched as a goroutine to watch the l.lines channel.
-// when a line is available, parser parses it and adds the metric(s) to the
+// parse is launched as a goroutine to watch the l.lines channel.
+// when a line is available, parse parses it and adds the metric(s) to the
 // accumulator.
 func (l *LogParserPlugin) parser() {
 	defer l.wg.Done()
@@ -265,18 +279,17 @@ func (l *LogParserPlugin) parser() {
 				continue
 			}
 		}
-		for _, parser := range l.parsers {
-			m, err = parser.ParseLine(entry.line)
+		m, err = l.GrokParser.ParseLine(entry.line)
 		if err == nil {
 			if m != nil {
 				tags := m.Tags()
 				tags["path"] = entry.path
-					l.acc.AddFields(m.Name(), m.Fields(), tags, m.Time())
+				l.acc.AddFields(l.MeasurementName, m.Fields(), tags, m.Time())
 			}
 		} else {
 			log.Println("E! Error parsing log line: " + err.Error())
 		}
-		}
+
 	}
 }
 

plugins/inputs/logparser/logparser_test.go

@@ -2,6 +2,7 @@ package logparser
 
 import (
 	"io/ioutil"
+	"log"
 	"os"
 	"runtime"
 	"strings"
@@ -9,8 +10,6 @@ import (
 
 	"github.com/influxdata/telegraf/testutil"
 
-	"github.com/influxdata/telegraf/plugins/inputs/logparser/grok"
-
 	"github.com/stretchr/testify/assert"
 )
 
@@ -26,15 +25,12 @@ func TestStartNoParsers(t *testing.T) {
 
 func TestGrokParseLogFilesNonExistPattern(t *testing.T) {
 	thisdir := getCurrentDir()
-	p := &grok.Parser{
-		Patterns:           []string{"%{FOOBAR}"},
-		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
-	}
 
 	logparser := &LogParserPlugin{
 		FromBeginning:      true,
 		Files:              []string{thisdir + "grok/testdata/*.log"},
-		GrokParser:    p,
+		Patterns:           []string{"%{FOOBAR}"},
+		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
 	}
 
 	acc := testutil.Accumulator{}
@@ -44,15 +40,13 @@ func TestGrokParseLogFilesNonExistPattern(t *testing.T) {
 
 func TestGrokParseLogFiles(t *testing.T) {
 	thisdir := getCurrentDir()
-	p := &grok.Parser{
-		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
-		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
-	}
 
 	logparser := &LogParserPlugin{
+		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
+		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
 		FromBeginning:      true,
 		Files:              []string{thisdir + "grok/testdata/*.log"},
-		GrokParser:    p,
+		MeasurementName:    "logparser_grok",
 	}
 
 	acc := testutil.Accumulator{}
@@ -62,6 +56,7 @@ func TestGrokParseLogFiles(t *testing.T) {
 
 	logparser.Stop()
 
+	log.Printf("metric[0] %v, tags: %v, fields: %v", acc.Metrics[0].Measurement, acc.Metrics[0].Tags, acc.Metrics[0].Fields)
 	acc.AssertContainsTaggedFields(t, "logparser_grok",
 		map[string]interface{}{
 			"clientip":      "192.168.1.1",
@@ -91,15 +86,13 @@ func TestGrokParseLogFilesAppearLater(t *testing.T) {
 	assert.NoError(t, err)
 
 	thisdir := getCurrentDir()
-	p := &grok.Parser{
-		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
-		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
-	}
 
 	logparser := &LogParserPlugin{
 		FromBeginning:      true,
 		Files:              []string{emptydir + "/*.log"},
-		GrokParser:    p,
+		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
+		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
+		MeasurementName:    "logparser_grok",
 	}
 
 	acc := testutil.Accumulator{}
@@ -130,16 +123,13 @@ func TestGrokParseLogFilesAppearLater(t *testing.T) {
 // pattern available for test_b.log
 func TestGrokParseLogFilesOneBad(t *testing.T) {
 	thisdir := getCurrentDir()
-	p := &grok.Parser{
-		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_BAD}"},
-		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
-	}
-	assert.NoError(t, p.Compile())
 
 	logparser := &LogParserPlugin{
 		FromBeginning:      true,
 		Files:              []string{thisdir + "grok/testdata/test_a.log"},
-		GrokParser:    p,
+		Patterns:           []string{"%{TEST_LOG_A}", "%{TEST_LOG_BAD}"},
+		CustomPatternFiles: []string{thisdir + "grok/testdata/test-patterns"},
+		MeasurementName:    "logparser_grok",
 	}
 
 	acc := testutil.Accumulator{}

plugins/inputs/reader/README.md

@@ -0,0 +1,23 @@
+# Reader Input Plugin
+
+The Reader Plugin updates a list of files every interval and parses the data inside.
+Files will always be read from the beginning.  
+This plugin can parse any "data_format" formats.
+
+### Configuration:
+```toml
+[[inputs.reader]]
+## Files to parse each interval.
+## These accept standard unix glob matching rules, but with the addition of
+## ** as a "super asterisk". ie:
+##   /var/log/**.log     -> recursively find all .log files in /var/log
+##   /var/log/*/*.log    -> find all .log files with a parent dir in /var/log
+##   /var/log/apache.log -> only tail the apache log file
+files = ["/var/log/apache/access.log"]
+
+## The dataformat to be read from files
+## Each data format has its own unique set of configuration options, read
+## more about them here:
+## https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md
+data_format = ""
+```

plugins/inputs/reader/dev/docker-compose.yml

@@ -0,0 +1,13 @@
+version: '3'
+
+services:
+  telegraf:
+      image: glinton/scratch
+      volumes:
+        - ./telegraf.conf:/telegraf.conf
+        - ../../../../telegraf:/telegraf
+        - ./json_a.log:/var/log/test.log
+      entrypoint:
+        - /telegraf
+        - --config
+        - /telegraf.conf

plugins/inputs/reader/dev/json_a.log

@@ -0,0 +1,14 @@
+{
+	"parent": {
+		"child": 3.0,
+		"ignored_child": "hi"
+	},
+	"ignored_null": null,
+	"integer": 4,
+	"list": [3, 4],
+	"ignored_parent": {
+		"another_ignored_null": null,
+		"ignored_string": "hello, world!"
+	},
+	"another_list": [4]
+}

plugins/inputs/reader/dev/telegraf.conf

@@ -0,0 +1,7 @@
+[[inputs.reader]]
+  files = ["/var/log/test.log"]
+  data_format = "json"
+  name_override = "json_reader"
+
+[[outputs.file]]
+  files = ["stdout"]

plugins/inputs/reader/reader.go

@@ -0,0 +1,95 @@
+package reader
+
+import (
+	"fmt"
+	"io/ioutil"
+
+	"github.com/influxdata/telegraf"
+	"github.com/influxdata/telegraf/internal/globpath"
+	"github.com/influxdata/telegraf/plugins/inputs"
+	"github.com/influxdata/telegraf/plugins/parsers"
+)
+
+type Reader struct {
+	Filepaths     []string `toml:"files"`
+	FromBeginning bool
+	parser        parsers.Parser
+
+	Filenames []string
+}
+
+const sampleConfig = `## Files to parse each interval.
+## These accept standard unix glob matching rules, but with the addition of
+## ** as a "super asterisk". ie:
+##   /var/log/**.log     -> recursively find all .log files in /var/log
+##   /var/log/*/*.log    -> find all .log files with a parent dir in /var/log
+##   /var/log/apache.log -> only tail the apache log file
+files = ["/var/log/apache/access.log"]
+
+## The dataformat to be read from files
+## Each data format has its own unique set of configuration options, read
+## more about them here:
+## https://github.com/influxdata/telegraf/blob/master/docs/DATA_FORMATS_INPUT.md
+data_format = ""
+`
+
+// SampleConfig returns the default configuration of the Input
+func (r *Reader) SampleConfig() string {
+	return sampleConfig
+}
+
+func (r *Reader) Description() string {
+	return "reload and gather from file[s] on telegraf's interval"
+}
+
+func (r *Reader) Gather(acc telegraf.Accumulator) error {
+	r.refreshFilePaths()
+	for _, k := range r.Filenames {
+		metrics, err := r.readMetric(k)
+		if err != nil {
+			return err
+		}
+
+		for _, m := range metrics {
+			acc.AddFields(m.Name(), m.Fields(), m.Tags(), m.Time())
+		}
+	}
+	return nil
+}
+
+func (r *Reader) SetParser(p parsers.Parser) {
+	r.parser = p
+}
+
+func (r *Reader) refreshFilePaths() error {
+	var allFiles []string
+	for _, filepath := range r.Filepaths {
+		g, err := globpath.Compile(filepath)
+		if err != nil {
+			return fmt.Errorf("E! Error Glob: %v could not be compiled, %s", filepath, err)
+		}
+		files := g.Match()
+
+		for k := range files {
+			allFiles = append(allFiles, k)
+		}
+	}
+
+	r.Filenames = allFiles
+	return nil
+}
+
+func (r *Reader) readMetric(filename string) ([]telegraf.Metric, error) {
+	fileContents, err := ioutil.ReadFile(filename)
+	if err != nil {
+		return nil, fmt.Errorf("E! Error file: %v could not be read, %s", filename, err)
+	}
+	return r.parser.Parse(fileContents)
+
+}
+
+func init() {
+	inputs.Add("reader", func() telegraf.Input {
+		return &Reader{}
+	})
+}

plugins/inputs/reader/reader_test.go

@@ -0,0 +1,64 @@
+package reader
+
+import (
+	"runtime"
+	"strings"
+	"testing"
+
+	"github.com/influxdata/telegraf/plugins/parsers"
+	"github.com/influxdata/telegraf/testutil"
+	"github.com/stretchr/testify/assert"
+)
+
+func TestRefreshFilePaths(t *testing.T) {
+	testDir := getPluginDir()
+	r := Reader{
+		Filepaths: []string{testDir + "/logparser/grok/testdata/**.log"},
+	}
+
+	r.refreshFilePaths()
+	assert.Equal(t, len(r.Filenames), 2)
+}
+func TestJSONParserCompile(t *testing.T) {
+	testDir := getPluginDir()
+	var acc testutil.Accumulator
+	r := Reader{
+		Filepaths: []string{testDir + "/reader/testfiles/json_a.log"},
+	}
+	parserConfig := parsers.Config{
+		DataFormat: "json",
+		TagKeys:    []string{"parent_ignored_child"},
+	}
+	nParser, err := parsers.NewParser(&parserConfig)
+	r.parser = nParser
+	assert.NoError(t, err)
+
+	r.Gather(&acc)
+	assert.Equal(t, map[string]string{"parent_ignored_child": "hi"}, acc.Metrics[0].Tags)
+	assert.Equal(t, 5, len(acc.Metrics[0].Fields))
+}
+
+func TestGrokParser(t *testing.T) {
+	testDir := getPluginDir()
+	var acc testutil.Accumulator
+	r := Reader{
+		Filepaths: []string{testDir + "/reader/testfiles/grok_a.log"},
+	}
+
+	parserConfig := parsers.Config{
+		DataFormat: "grok",
+		Patterns:   []string{"%{COMMON_LOG_FORMAT}"},
+	}
+
+	nParser, err := parsers.NewParser(&parserConfig)
+	r.parser = nParser
+	assert.NoError(t, err)
+
+	err = r.Gather(&acc)
+	assert.Equal(t, 2, len(acc.Metrics))
+}
+
+func getPluginDir() string {
+	_, filename, _, _ := runtime.Caller(1)
+	return strings.Replace(filename, "/reader/reader_test.go", "", 1)
+}

plugins/inputs/reader/testfiles/grok_a.log

@@ -0,0 +1,2 @@
+127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326
+128.0.0.1 user-identifier tony [10/Oct/2000:13:55:36 -0800] "GET /apache_pb.gif HTTP/1.0" 300 45

plugins/inputs/reader/testfiles/json_a.log

@@ -0,0 +1,14 @@
+{
+		"parent": {
+			"child": 3.0,
+			"ignored_child": "hi"
+		},
+		"ignored_null": null,
+		"integer": 4,
+		"list": [3, 4],
+		"ignored_parent": {
+			"another_ignored_null": null,
+			"ignored_string": "hello, world!"
+		},
+		"another_list": [4]
+	}

plugins/parsers/grok/influx-patterns

@@ -0,0 +1,73 @@
+# Captures are a slightly modified version of logstash "grok" patterns, with
+#  the format %{<capture syntax>[:<semantic name>][:<modifier>]}
+# By default all named captures are converted into string fields.
+# Modifiers can be used to convert captures to other types or tags.
+# Timestamp modifiers can be used to convert captures to the timestamp of the
+#  parsed metric.
+
+# View logstash grok pattern docs here:
+#   https://www.elastic.co/guide/en/logstash/current/plugins-filters-grok.html
+# All default logstash patterns are supported, these can be viewed here:
+#   https://github.com/logstash-plugins/logstash-patterns-core/blob/master/patterns/grok-patterns
+
+# Available modifiers:
+#   string   (default if nothing is specified)
+#   int
+#   float
+#   duration (ie, 5.23ms gets converted to int nanoseconds)
+#   tag      (converts the field into a tag)
+#   drop     (drops the field completely)
+# Timestamp modifiers:
+#   ts-ansic         ("Mon Jan _2 15:04:05 2006")
+#   ts-unix          ("Mon Jan _2 15:04:05 MST 2006")
+#   ts-ruby          ("Mon Jan 02 15:04:05 -0700 2006")
+#   ts-rfc822        ("02 Jan 06 15:04 MST")
+#   ts-rfc822z       ("02 Jan 06 15:04 -0700")
+#   ts-rfc850        ("Monday, 02-Jan-06 15:04:05 MST")
+#   ts-rfc1123       ("Mon, 02 Jan 2006 15:04:05 MST")
+#   ts-rfc1123z      ("Mon, 02 Jan 2006 15:04:05 -0700")
+#   ts-rfc3339       ("2006-01-02T15:04:05Z07:00")
+#   ts-rfc3339nano   ("2006-01-02T15:04:05.999999999Z07:00")
+#   ts-httpd         ("02/Jan/2006:15:04:05 -0700")
+#   ts-epoch         (seconds since unix epoch)
+#   ts-epochnano     (nanoseconds since unix epoch)
+#   ts-"CUSTOM"
+# CUSTOM time layouts must be within quotes and be the representation of the
+# "reference time", which is Mon Jan 2 15:04:05 -0700 MST 2006
+# See https://golang.org/pkg/time/#Parse for more details.
+
+# Example log file pattern, example log looks like this:
+#   [04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs
+# Breakdown of the DURATION pattern below:
+#   NUMBER  is a builtin logstash grok pattern matching float & int numbers.
+#   [nuµm]? is a regex specifying 0 or 1 of the characters within brackets.
+#   s       is also regex, this pattern must end in "s".
+# so DURATION will match something like '5.324ms' or '6.1µs' or '10s'
+DURATION %{NUMBER}[nuµm]?s
+RESPONSE_CODE %{NUMBER:response_code:tag}
+RESPONSE_TIME %{DURATION:response_time_ns:duration}
+EXAMPLE_LOG \[%{HTTPDATE:ts:ts-httpd}\] %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
+
+# Wider-ranging username matching vs. logstash built-in %{USER}
+NGUSERNAME [a-zA-Z0-9\.\@\-\+_%]+
+NGUSER %{NGUSERNAME}
+# Wider-ranging client IP matching
+CLIENT (?:%{IPORHOST}|%{HOSTPORT}|::1)
+
+##
+## COMMON LOG PATTERNS
+##
+
+# apache & nginx logs, this is also known as the "common log format"
+#   see https://en.wikipedia.org/wiki/Common_Log_Format
+COMMON_LOG_FORMAT %{CLIENT:client_ip} %{NOTSPACE:ident} %{NOTSPACE:auth} \[%{HTTPDATE:ts:ts-httpd}\] "(?:%{WORD:verb:tag} %{NOTSPACE:request}(?: HTTP/%{NUMBER:http_version:float})?|%{DATA})" %{NUMBER:resp_code:tag} (?:%{NUMBER:resp_bytes:int}|-)
+
+# Combined log format is the same as the common log format but with the addition
+# of two quoted strings at the end for "referrer" and "agent"
+#   See Examples at http://httpd.apache.org/docs/current/mod/mod_log_config.html
+COMBINED_LOG_FORMAT %{COMMON_LOG_FORMAT} %{QS:referrer} %{QS:agent}
+
+# HTTPD log formats
+HTTPD20_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{LOGLEVEL:loglevel:tag}\] (?:\[client %{IPORHOST:clientip}\] ){0,1}%{GREEDYDATA:errormsg}
+HTTPD24_ERRORLOG \[%{HTTPDERROR_DATE:timestamp}\] \[%{WORD:module}:%{LOGLEVEL:loglevel:tag}\] \[pid %{POSINT:pid:int}:tid %{NUMBER:tid:int}\]( \(%{POSINT:proxy_errorcode:int}\)%{DATA:proxy_errormessage}:)?( \[client %{IPORHOST:client}:%{POSINT:clientport}\])? %{DATA:errorcode}: %{GREEDYDATA:message}
+HTTPD_ERRORLOG %{HTTPD20_ERRORLOG}|%{HTTPD24_ERRORLOG}

plugins/parsers/grok/parser.go

@@ -68,10 +68,10 @@ type Parser struct {
 	// specified by the user in Patterns.
 	// They will look like:
 	//   GROK_INTERNAL_PATTERN_0, GROK_INTERNAL_PATTERN_1, etc.
-	namedPatterns      []string
+	NamedPatterns      []string
 	CustomPatterns     string
 	CustomPatternFiles []string
-	Measurement        string
+	MetricName         string
 
 	// Timezone is an optional component to help render log dates to
 	// your chosen zone.
@@ -133,7 +133,7 @@ func (p *Parser) Compile() error {
 
 	// Give Patterns fake names so that they can be treated as named
 	// "custom patterns"
-	p.namedPatterns = make([]string, 0, len(p.Patterns))
+	p.NamedPatterns = make([]string, 0, len(p.Patterns))
 	for i, pattern := range p.Patterns {
 		pattern = strings.TrimSpace(pattern)
 		if pattern == "" {
@@ -141,10 +141,10 @@ func (p *Parser) Compile() error {
 		}
 		name := fmt.Sprintf("GROK_INTERNAL_PATTERN_%d", i)
 		p.CustomPatterns += "\n" + name + " " + pattern + "\n"
-		p.namedPatterns = append(p.namedPatterns, "%{"+name+"}")
+		p.NamedPatterns = append(p.NamedPatterns, "%{"+name+"}")
 	}
 
-	if len(p.namedPatterns) == 0 {
+	if len(p.NamedPatterns) == 0 {
 		return fmt.Errorf("pattern required")
 	}
 
@@ -167,10 +167,6 @@ func (p *Parser) Compile() error {
 		p.addCustomPatterns(scanner)
 	}
 
-	if p.Measurement == "" {
-		p.Measurement = "logparser_grok"
-	}
-
 	p.loc, err = time.LoadLocation(p.Timezone)
 	if err != nil {
 		log.Printf("W! improper timezone supplied (%s), setting loc to UTC", p.Timezone)
@@ -191,7 +187,7 @@ func (p *Parser) ParseLine(line string) (telegraf.Metric, error) {
 	var values map[string]string
 	// the matching pattern string
 	var patternName string
-	for _, pattern := range p.namedPatterns {
+	for _, pattern := range p.NamedPatterns {
 		if values, err = p.g.Parse(pattern, line); err != nil {
 			return nil, err
 		}
@@ -335,9 +331,6 @@ func (p *Parser) ParseLine(line string) (telegraf.Metric, error) {
 		case DROP:
 		// goodbye!
 		default:
-			// Replace commas with dot character
-			v = strings.Replace(v, ",", ".", -1)
-
 			ts, err := time.ParseInLocation(t, v, p.loc)
 			if err == nil {
 				timestamp = ts
@@ -351,7 +344,26 @@ func (p *Parser) ParseLine(line string) (telegraf.Metric, error) {
 		return nil, fmt.Errorf("logparser_grok: must have one or more fields")
 	}
 
-	return metric.New(p.Measurement, tags, fields, p.tsModder.tsMod(timestamp))
+	return metric.New(p.MetricName, tags, fields, p.tsModder.tsMod(timestamp))
+}
+
+func (p *Parser) Parse(buf []byte) ([]telegraf.Metric, error) {
+	lines := strings.Split(string(buf), "\n")
+	var metrics []telegraf.Metric
+
+	for _, line := range lines {
+		m, err := p.ParseLine(line)
+		if err != nil {
+			return nil, err
+		}
+		metrics = append(metrics, m)
+	}
+
+	return metrics, nil
+}
+
+func (p *Parser) SetDefaultTags(tags map[string]string) {
+	//needs implementation
 }
 
 func (p *Parser) addCustomPatterns(scanner *bufio.Scanner) {

plugins/parsers/grok/parser_test.go

@@ -4,79 +4,18 @@ import (
 	"testing"
 	"time"
 
-	"github.com/influxdata/telegraf"
-
 	"github.com/stretchr/testify/assert"
 	"github.com/stretchr/testify/require"
 )
 
-var benchM telegraf.Metric
-
-func Benchmark_ParseLine_CommonLogFormat(b *testing.B) {
-	p := &Parser{
+func TestGrokParse(t *testing.T) {
+	parser := Parser{
+		MetricName: "t_met",
 		Patterns:   []string{"%{COMMON_LOG_FORMAT}"},
 	}
-	_ = p.Compile()
-
-	var m telegraf.Metric
-	for n := 0; n < b.N; n++ {
-		m, _ = p.ParseLine(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
-	}
-	benchM = m
-}
-
-func Benchmark_ParseLine_CombinedLogFormat(b *testing.B) {
-	p := &Parser{
-		Patterns: []string{"%{COMBINED_LOG_FORMAT}"},
-	}
-	_ = p.Compile()
-
-	var m telegraf.Metric
-	for n := 0; n < b.N; n++ {
-		m, _ = p.ParseLine(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326 "-" "Mozilla"`)
-	}
-	benchM = m
-}
-
-func Benchmark_ParseLine_CustomPattern(b *testing.B) {
-	p := &Parser{
-		Patterns: []string{"%{TEST_LOG_A}", "%{TEST_LOG_B}"},
-		CustomPatterns: `
-			DURATION %{NUMBER}[nuµm]?s
-			RESPONSE_CODE %{NUMBER:response_code:tag}
-			RESPONSE_TIME %{DURATION:response_time:duration}
-			TEST_LOG_A %{NUMBER:myfloat:float} %{RESPONSE_CODE} %{IPORHOST:clientip} %{RESPONSE_TIME}
-		`,
-	}
-	_ = p.Compile()
-
-	var m telegraf.Metric
-	for n := 0; n < b.N; n++ {
-		m, _ = p.ParseLine(`[04/Jun/2016:12:41:45 +0100] 1.25 200 192.168.1.1 5.432µs 101`)
-	}
-	benchM = m
-}
-
-// Test a very simple parse pattern.
-func TestSimpleParse(t *testing.T) {
-	p := &Parser{
-		Patterns: []string{"%{TESTLOG}"},
-		CustomPatterns: `
-			TESTLOG %{NUMBER:num:int} %{WORD:client}
-		`,
-	}
-	assert.NoError(t, p.Compile())
-
-	m, err := p.ParseLine(`142 bot`)
+	parser.Compile()
+	_, err := parser.Parse([]byte(`127.0.0.1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`))
 	assert.NoError(t, err)
-	require.NotNil(t, m)
-
-	assert.Equal(t,
-		map[string]interface{}{
-			"num":    int64(142),
-			"client": "bot",
-		},
-		m.Fields())
 }
 
 // Verify that patterns with a regex lookahead fail at compile time.
@@ -96,7 +35,6 @@ func TestParsePatternsWithLookahead(t *testing.T) {
 
 func TestMeasurementName(t *testing.T) {
 	p := &Parser{
-		Measurement: "my_web_log",
 		Patterns: []string{"%{COMMON_LOG_FORMAT}"},
 	}
 	assert.NoError(t, p.Compile())
@@ -116,12 +54,10 @@ func TestMeasurementName(t *testing.T) {
 		},
 		m.Fields())
 	assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
-	assert.Equal(t, "my_web_log", m.Name())
 }
 
 func TestCLF_IPv6(t *testing.T) {
 	p := &Parser{
-		Measurement: "my_web_log",
 		Patterns: []string{"%{COMMON_LOG_FORMAT}"},
 	}
 	assert.NoError(t, p.Compile())
@@ -140,7 +76,6 @@ func TestCLF_IPv6(t *testing.T) {
 		},
 		m.Fields())
 	assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
-	assert.Equal(t, "my_web_log", m.Name())
 
 	m, err = p.ParseLine(`::1 user-identifier frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326`)
 	require.NotNil(t, m)
@@ -156,7 +91,6 @@ func TestCLF_IPv6(t *testing.T) {
 		},
 		m.Fields())
 	assert.Equal(t, map[string]string{"verb": "GET", "resp_code": "200"}, m.Tags())
-	assert.Equal(t, "my_web_log", m.Name())
 }
 
 func TestCustomInfluxdbHttpd(t *testing.T) {
@@ -970,33 +904,3 @@ func TestNewlineInPatterns(t *testing.T) {
 	require.NoError(t, err)
 	require.NotNil(t, m)
 }
-
-func TestSyslogTimestampParser(t *testing.T) {
-	p := &Parser{
-		Patterns: []string{`%{SYSLOGTIMESTAMP:timestamp:ts-syslog} value=%{NUMBER:value:int}`},
-		timeFunc: func() time.Time { return time.Date(2018, time.April, 1, 0, 0, 0, 0, nil) },
-	}
-	require.NoError(t, p.Compile())
-	m, err := p.ParseLine("Sep 25 09:01:55 value=42")
-	require.NoError(t, err)
-	require.NotNil(t, m)
-	require.Equal(t, 2018, m.Time().Year())
-}
-
-func TestReplaceTimestampComma(t *testing.T) {
-
-	p := &Parser{
-		Patterns: []string{`%{TIMESTAMP_ISO8601:timestamp:ts-"2006-01-02 15:04:05.000"} successfulMatches=%{NUMBER:value:int}`},
-	}
-
-	require.NoError(t, p.Compile())
-	m, err := p.ParseLine("2018-02-21 13:10:34,555 successfulMatches=1")
-	require.NoError(t, err)
-	require.NotNil(t, m)
-
-	require.Equal(t, 2018, m.Time().Year())
-	require.Equal(t, 13, m.Time().Hour())
-	require.Equal(t, 34, m.Time().Second())
-	//Convert Nanosecond to milisecond for compare
-	require.Equal(t, 555, m.Time().Nanosecond()/1000000)
-}

plugins/parsers/registry.go

@@ -8,6 +8,7 @@ import (
 	"github.com/influxdata/telegraf/plugins/parsers/collectd"
 	"github.com/influxdata/telegraf/plugins/parsers/dropwizard"
 	"github.com/influxdata/telegraf/plugins/parsers/graphite"
+	"github.com/influxdata/telegraf/plugins/parsers/grok"
 	"github.com/influxdata/telegraf/plugins/parsers/influx"
 	"github.com/influxdata/telegraf/plugins/parsers/json"
 	"github.com/influxdata/telegraf/plugins/parsers/nagios"
@@ -87,6 +88,13 @@ type Config struct {
 	// an optional map containing tag names as keys and json paths to retrieve the tag values from as values
 	// used if TagsPath is empty or doesn't return any tags
 	DropwizardTagPathsMap map[string]string
+
+	//grok patterns
+	Patterns           []string
+	NamedPatterns      []string
+	CustomPatterns     string
+	CustomPatternFiles []string
+	TimeZone           string
 }
 
 // NewParser returns a Parser interface based on the given config.
@@ -120,12 +128,38 @@ func NewParser(config *Config) (Parser, error) {
 			config.DefaultTags,
 			config.Separator,
 			config.Templates)
+	case "grok":
+		parser, err = newGrokParser(
+			config.MetricName,
+			config.Patterns,
+			config.NamedPatterns,
+			config.CustomPatterns,
+			config.CustomPatternFiles,
+			config.TimeZone)
 	default:
 		err = fmt.Errorf("Invalid data format: %s", config.DataFormat)
 	}
 	return parser, err
 }
 
+func newGrokParser(metricName string,
+	patterns []string,
+	nPatterns []string,
+	cPatterns string,
+	cPatternFiles []string, tZone string) (Parser, error) {
+	parser := grok.Parser{
+		MetricName:         metricName,
+		Patterns:           patterns,
+		NamedPatterns:      nPatterns,
+		CustomPatterns:     cPatterns,
+		CustomPatternFiles: cPatternFiles,
+		Timezone:           tZone,
+	}
+
+	err := parser.Compile()
+	return &parser, err
+}
+
 func NewJSONParser(
 	metricName string,
 	tagKeys []string,