re-worked implementation of lowercase into a more generic strings processor (influxdata#3890)

Author: Berin Smaldon

plugins/processors/all/all.go

@@ -7,4 +7,5 @@ import (
 	_ "github.com/influxdata/telegraf/plugins/processors/printer"
 	_ "github.com/influxdata/telegraf/plugins/processors/regex"
 	_ "github.com/influxdata/telegraf/plugins/processors/topk"
+	_ "github.com/influxdata/telegraf/plugins/processors/strings"
 )

plugins/processors/strings/README.md

@@ -0,0 +1,48 @@
+# Strings Processor Plugin
+
+The `strings` plugin maps certain go string functions onto tag and field values. If `result_key` parameter is present, it can produce new tags and fields from existing ones.
+
+Implemented functions are: Lowercase, Uppercase, Trim, TrimPrefix, TrimSuffix, TrimRight, TrimLeft
+
+Please note that in this implementation these are processed in the order that they appear above.
+
+Specify the `tag` or `field` that you want processed in each section and optionally a `result_key` if you want the result stored in a new tag or field. You can specify lots of transformations on data with a single strings processor. Certain functions require an `argument` field to specify how they should process their throughput.
+
+Functions that require an `argument` are: Trim, TrimPrefix, TrimSuffix, TrimRight, TrimLeft
+
+### Configuration:
+
+```toml
+[[processors.strings]]
+  namepass = ["uri_stem"]
+
+  # Tag and field conversions defined in a separate sub-tables
+  [[processors.strings.lowercase]]
+    ## Tag to change
+    tag = "uri_stem"
+
+  [[processors.strings.lowercase]]
+    ## Multiple tags or fields may be defined
+    tag = "method"
+
+  [[processors.strings.uppercase]]
+    key = "cs-host"
+    result_key = "cs-host_normalised"
+
+  [[processors.strings.trimprefix]]
+    tag = "uri_stem"
+    argument = "/api/"
+```
+
+### Tags:
+
+No tags are applied by this processor.
+
+### Example Input:
+```
+iis_log,method=get,uri_stem=/API/HealthCheck cs-host="MIXEDCASE_host",referrer="-",ident="-",http_version=1.1,agent="UserAgent",resp_bytes=270i 1519652321000000000
+```
+### Example Output:
+```
+iis_log,method=get,uri_stem=healthcheck cs-host="MIXEDCASE_host",cs-host_normalised="MIXEDCASE_HOST",referrer="-",ident="-",http_version=1.1,agent="UserAgent",resp_bytes=270i 1519652321000000000
+```

plugins/processors/strings/strings.go

@@ -0,0 +1,114 @@
+package strings
+
+import (
+	"strings"
+
+	"github.com/influxdata/telegraf"
+	"github.com/influxdata/telegraf/plugins/processors"
+)
+
+type Strings struct {
+    Lowercase  []converter
+    Uppercase  []converter
+    Trim       []converter
+    TrimLeft   []converter
+    TrimRight  []converter
+    TrimPrefix []converter
+    TrimSuffix []converter
+}
+
+type converter struct {
+	Tag         string
+    Field       string
+	ResultKey   string
+    Argument    string
+}
+
+const sampleConfig = `
+  ## Tag and field conversions defined in a separate sub-tables
+
+  # [[processors.strings.uppercase]]
+  #   tag = "method"
+
+  # [[processors.strings.lowercase]]
+  #   field = "uri_stem"
+  #   result_key = "uri_stem_normalised"
+`
+
+func (r *Strings) SampleConfig() string {
+	return sampleConfig
+}
+
+func (r *Strings) Description() string {
+	return "Transforms tag and field values to lower case"
+}
+
+func ApplyFunction(
+        metric telegraf.Metric,
+        c converter,
+        fn func(string) string) {
+
+    if value, ok := metric.Tags()[c.Tag]; ok {
+        metric.AddTag(
+            getKey(c),
+            fn(value),
+        )
+    } else if value, ok := metric.Fields()[c.Field]; ok {
+        switch value := value.(type) {
+        case string:
+            metric.AddField(
+                getKey(c),
+                fn(value),
+            )
+        }
+    }
+}
+
+func (r *Strings) Apply(in ...telegraf.Metric) []telegraf.Metric {
+	for _, metric := range in {
+		for _, converter := range r.Lowercase {
+            ApplyFunction(metric, converter, strings.ToLower)
+		}
+		for _, converter := range r.Uppercase {
+            ApplyFunction(metric, converter, strings.ToUpper)
+		}
+        for _, converter := range r.Trim {
+            ApplyFunction(metric, converter,
+                func(s string) string { return strings.Trim(s, converter.Argument) })
+        }
+        for _, converter := range r.TrimPrefix {
+            ApplyFunction(metric, converter,
+                func(s string) string { return strings.TrimPrefix(s, converter.Argument) })
+        }
+        for _, converter := range r.TrimSuffix{
+            ApplyFunction(metric, converter,
+                func(s string) string { return strings.TrimSuffix(s, converter.Argument) })
+        }
+        for _, converter := range r.TrimRight {
+            ApplyFunction(metric, converter,
+                func(s string) string { return strings.TrimRight(s, converter.Argument) })
+        }
+        for _, converter := range r.TrimLeft {
+            ApplyFunction(metric, converter,
+                func(s string) string { return strings.TrimLeft(s, converter.Argument) })
+        }
+	}
+
+	return in
+}
+
+func getKey(c converter) string {
+	if c.ResultKey != "" {
+		return c.ResultKey
+	} else if c.Field != "" {
+        return c.Field
+    } else {
+	    return c.Tag
+    }
+}
+
+func init() {
+	processors.Add("strings", func() telegraf.Processor {
+		return &Strings{}
+	})
+}