feat(parsers.xpath): Add option to store fields as base64 (#13690)

plugins/parsers/xpath/README.md

@@ -162,6 +162,11 @@ XPath expressions.
     ## By default, all byte-array-fields are converted to string.
     # fields_bytes_as_hex = []
 
+    ## Optional: List of fields to convert to base64-strings if they
+    ## contain byte-arrays. Resulting string will generally be shorter
+    ## than using hex encoding. Base64 encoding is RFC4648 compliant.
+    # fields_bytes_as_base64 = []
+
     ## Tag definitions using the given XPath queries.
     [inputs.file.xpath.tags]
       name   = "substring-after(Sensor/@name, ' ')"

plugins/parsers/xpath/parser.go

@@ -1,6 +1,7 @@
 package xpath
 
 import (
+	"encoding/base64"
 	"encoding/hex"
 	"errors"
 	"fmt"
@@ -66,6 +67,7 @@ type Config struct {
 	Fields       map[string]string `toml:"fields"`
 	FieldsInt    map[string]string `toml:"fields_int"`
 	FieldsHex    []string          `toml:"fields_bytes_as_hex"`
+	FieldsBase64 []string          `toml:"fields_bytes_as_base64"`
 
 	FieldSelection  string `toml:"field_selection"`
 	FieldNameQuery  string `toml:"field_name"`
@@ -77,8 +79,9 @@ type Config struct {
 	TagValueQuery string `toml:"tag_value"`
 	TagNameExpand bool   `toml:"tag_name_expansion"`
 
-	FieldsHexFilter filter.Filter
-	Location        *time.Location
+	FieldsHexFilter    filter.Filter
+	FieldsBase64Filter filter.Filter
+	Location           *time.Location
 }
 
 func (p *Parser) Init() error {
@@ -175,6 +178,12 @@ func (p *Parser) Init() error {
 		}
 		config.FieldsHexFilter = f
 
+		bf, err := filter.Compile(config.FieldsBase64)
+		if err != nil {
+			return fmt.Errorf("creating base64-fields filter failed: %w", err)
+		}
+		config.FieldsBase64Filter = bf
+
 		p.Configs[i] = config
 	}
 
@@ -420,6 +429,9 @@ func (p *Parser) parseQuery(starttime time.Time, doc, selected dataNode, config
 							if config.FieldsHexFilter != nil && config.FieldsHexFilter.Match(name) {
 								v = hex.EncodeToString(b)
 							}
+							if config.FieldsBase64Filter != nil && config.FieldsBase64Filter.Match(name) {
+								v = base64.StdEncoding.EncodeToString(b)
+							}
 						} else {
 							v = fmt.Sprintf("%v", v)
 						}
@@ -477,6 +489,9 @@ func (p *Parser) parseQuery(starttime time.Time, doc, selected dataNode, config
 					if config.FieldsHexFilter != nil && config.FieldsHexFilter.Match(name) {
 						v = hex.EncodeToString(b)
 					}
+					if config.FieldsBase64Filter != nil && config.FieldsBase64Filter.Match(name) {
+						v = base64.StdEncoding.EncodeToString(b)
+					}
 				} else {
 					v = fmt.Sprintf("%v", v)
 				}

plugins/parsers/xpath/testcases/cbor_base64_encoding/expected.out

@@ -0,0 +1 @@
+data str_a="this is a test",str_b="foobar",bytes_a="0001020304050607",fielda="AAECAwQFBgc=",bytes_b="Zm9vYmFy",timestamp=1687852514u 1687852514000000000

plugins/parsers/xpath/testcases/cbor_base64_encoding/telegraf.conf

@@ -0,0 +1,28 @@
+# Example data:
+# [
+#   {
+#     "str_a": bytearray("this is a test"),
+#     "str_b": bytearray("foobar"),
+#     "bytes_a": bytearray([0, 1, 2, 3, 4, 5, 6, 7]),
+#     "bytes_b": bytearray("foobar"),
+#     "timestamp": 1687852514
+#   }
+# ]
+
+[[inputs.file]]
+  files = ["./testcases/cbor_base64_encoding/data.bin"]
+  data_format = "xpath_cbor"
+
+  xpath_native_types = true
+
+  [[inputs.file.xpath]]
+    metric_name = "'data'"
+    metric_selection = "child::*"
+    timestamp = "timestamp"
+    timestamp_format = "unix"
+    field_selection = "child::*"
+    fields_bytes_as_base64 = ["bytes_b", "field*"]
+    fields_bytes_as_hex = ["bytes_a"]
+
+    [inputs.file.xpath.fields]
+      fielda = "bytes_a"