[pkg/ottl] Add ParseKeyValue function

.chloggen/pkg-ottl-add-parse-key-value-function.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: pkg/ottl
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add `ParseKeyValue` function for parsing key value pairs from a target string
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [30998]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: []

pkg/ottl/e2e/e2e_test.go

@@ -437,6 +437,30 @@ func Test_e2e_converters(t *testing.T) {
 				m.PutDouble("id", 1)
 			},
 		},
+		{
+			statement: `set(attributes["test"], ParseKeyValue("k1=v1 k2=v2"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				m := tCtx.GetLogRecord().Attributes().PutEmptyMap("test")
+				m.PutStr("k1", "v1")
+				m.PutStr("k2", "v2")
+			},
+		},
+		{
+			statement: `set(attributes["test"], ParseKeyValue("k1!v1_k2!v2", "!", "_"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				m := tCtx.GetLogRecord().Attributes().PutEmptyMap("test")
+				m.PutStr("k1", "v1")
+				m.PutStr("k2", "v2")
+			},
+		},
+		{
+			statement: `set(attributes["test"], ParseKeyValue("k1!v1_k2!\"v2__!__v2\"", "!", "_"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				m := tCtx.GetLogRecord().Attributes().PutEmptyMap("test")
+				m.PutStr("k1", "v1")
+				m.PutStr("k2", "v2__!__v2")
+			},
+		},
 		{
 			statement: `set(attributes["test"], Seconds(Duration("1m")))`,
 			want: func(tCtx ottllog.TransformContext) {

pkg/ottl/ottlfuncs/README.md

@@ -410,6 +410,7 @@ Available Converters:
 - [Nanoseconds](#nanoseconds)
 - [Now](#now)
 - [ParseJSON](#parsejson)
+- [ParseKeyValue](#parsekeyvalue)
 - [Seconds](#seconds)
 - [SHA1](#sha1)
 - [SHA256](#sha256)
@@ -840,6 +841,26 @@ Examples:
 
 - `ParseJSON(body)`
 
+### ParseKeyValue
+
+`ParseKeyValue(target, Optional[delimiter], Optional[pair_delimiter])`
+
+The `ParseKeyValue` Converter returns a `pcommon.Map` that is a result of parsing the target string for key value pairs.
+
+`target` is a Getter that returns a string. `delimiter` is an optional string that is used to split the key and value in a pair, the default is `=`. `pair_delimiter` is an optional string that is used to split key value pairs, the default is white space.
+
+For example, the following target `"k1=v1 k2=v2 k3=v3"` will use default delimiters and be parsed into the following map:
+```
+{ "k1": "v1", "k2": "v2", "k3": "v3" }
+```
+
+Examples:
+
+- `ParseKeyValue("k1=v1 k2=v2 k3=v3")`
+- `ParseKeyValue("k1!v1_k2!v2_k3!v3", "!", "_")`
+- `ParseKeyValue(attributes["pairs"])`
+
+
 ### Seconds
 
 `Seconds(value)`

pkg/ottl/ottlfuncs/func_parse_key_value.go

@@ -0,0 +1,120 @@
+package ottlfuncs
+
+import (
+	"context"
+	"fmt"
+	"strings"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl"
+	"go.opentelemetry.io/collector/pdata/pcommon"
+)
+
+type ParseKeyValueArguments[K any] struct {
+	Target        ottl.StringGetter[K]
+	Delimiter     ottl.Optional[string]
+	PairDelimiter ottl.Optional[string]
+}
+
+func NewParseKeyValueFactory[K any]() ottl.Factory[K] {
+	return ottl.NewFactory("ParseKeyValue", &ParseKeyValueArguments[K]{}, createParseKeyValueFunction[K])
+}
+
+func createParseKeyValueFunction[K any](_ ottl.FunctionContext, oArgs ottl.Arguments) (ottl.ExprFunc[K], error) {
+	args, ok := oArgs.(*ParseKeyValueArguments[K])
+
+	if !ok {
+		return nil, fmt.Errorf("ParseKeyValueFactory args must be of type *ParseKeyValueArguments[K]")
+	}
+
+	return parseKeyValue[K](args.Target, args.Delimiter, args.PairDelimiter)
+}
+
+func parseKeyValue[K any](target ottl.StringGetter[K], d ottl.Optional[string], p ottl.Optional[string]) (ottl.ExprFunc[K], error) {
+	delimiter := "="
+	if !d.IsEmpty() {
+		delimiter = d.Get()
+	}
+
+	pairDelimiter := " "
+	if !p.IsEmpty() {
+		pairDelimiter = p.Get()
+	}
+
+	if pairDelimiter == delimiter {
+		return nil, fmt.Errorf("pair delimiter \"%s\" cannot be equal to delimiter \"%s\"", pairDelimiter, delimiter)
+	}
+
+	return func(ctx context.Context, tCtx K) (any, error) {
+		source, err := target.Get(ctx, tCtx)
+		if err != nil {
+			return nil, err
+		}
+
+		if source == "" {
+			return nil, fmt.Errorf("cannot parse from empty target")
+		}
+
+		pairs, err := splitPairs(source, pairDelimiter)
+		if err != nil {
+			return nil, fmt.Errorf("splitting pairs failed: %w", err)
+		}
+
+		parsed := make(map[string]any)
+		for _, p := range pairs {
+			pair := strings.SplitN(p, delimiter, 2)
+			if len(pair) != 2 {
+				return nil, fmt.Errorf("cannot split '%s' into 2 items, got %d", p, len(pair))
+			}
+			key := strings.TrimSpace(strings.Trim(pair[0], "\"'"))
+			value := strings.TrimSpace(strings.Trim(pair[1], "\"'"))
+			parsed[key] = value
+		}
+
+		result := pcommon.NewMap()
+		err = result.FromRaw(parsed)
+		return result, err
+	}, nil
+}
+
+// splitPairs will split the input on the pairDelimiter and return the resulting slice.
+// `strings.Split` is not used because it does not respect quotes and will split if the delimiter appears in a quoted value
+func splitPairs(input, pairDelimiter string) ([]string, error) {
+	var result []string
+	currentPair := ""
+	delimiterLength := len(pairDelimiter)
+	quoteChar := "" // "" means we are not in quotes
+
+	i := 0
+	for i < len(input) {
+		if quoteChar == "" && i+delimiterLength <= len(input) && input[i:i+delimiterLength] == pairDelimiter {
+			if currentPair == "" {
+				i++
+				continue
+			}
+			result = append(result, currentPair)
+			currentPair = ""
+			i += delimiterLength
+			continue
+		} else if input[i] == '"' || input[i] == '\'' {
+			if quoteChar != "" {
+				if quoteChar == string(input[i]) {
+					quoteChar = ""
+				}
+			} else {
+				quoteChar = string(input[i])
+			}
+		}
+		currentPair += string(input[i])
+		i++
+	}
+
+	if quoteChar != "" {
+		return nil, fmt.Errorf("never reached end of a quoted value")
+	}
+
+	if currentPair != "" {
+		result = append(result, currentPair)
+	}
+
+	return result, nil
+}