[pkg/ottl] Add ElementizeAttributesXML Converter

.chloggen/elementize-attributes-xml.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 ElementizeAttributesXML Converter
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: []
+
+# (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

@@ -377,6 +377,12 @@ func Test_e2e_converters(t *testing.T) {
 				tCtx.GetLogRecord().Attributes().PutStr("test", "pass")
 			},
 		},
+		{
+			statement: `set(attributes["test"], ElementizeAttributesXML("<Log id=\"1\"><Message>This is a log message!</Message></Log>"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				tCtx.GetLogRecord().Attributes().PutStr("test", `<Log><Message>This is a log message!</Message><id>1</id></Log>`)
+			},
+		},
 		{
 			statement: `set(attributes["test"], ExtractPatterns("aa123bb", "(?P<numbers>\\d+)"))`,
 			want: func(tCtx ottllog.TransformContext) {

pkg/ottl/ottlfuncs/README.md

@@ -597,6 +597,31 @@ Examples:
 - `Duration("333ms")`
 - `Duration("1000000h")`
 
+
+### ElementizeAttributesXML
+
+`ElementizeAttributesXML(target, Optional[xpath])`
+
+The `ElementizeAttributesXML` Converter returns an edited version of an XML string where attributes are converted into child elements.
+
+`target` is a Getter that returns a string. This string should be in XML format.
+If `target` is not a string, nil, or cannot be parsed as XML, `ElementizeAttributesXML` will return an error.
+
+`xpath` (optional) is a string that specifies an [XPath](https://www.w3.org/TR/1999/REC-xpath-19991116/) expression that
+selects one or more elements. Attributes will only be converted within the result(s) of the xpath.
+
+For example, `<a foo="bar"><b>baz</b></a>` will be converted to `<a><b>baz</b><foo>bar</foo></a>`.
+
+Examples:
+
+Convert all attributes in a document
+
+- `ElementizeAttributesXML(body)`
+
+Convert only attributes within "Record" elements
+
+- `ElementizeAttributesXML(body, "/Log/Record")`
+
 ### ExtractPatterns
 
 `ExtractPatterns(target, pattern)`

pkg/ottl/ottlfuncs/func_elementize_attributes_xml.go

@@ -0,0 +1,103 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package ottlfuncs // import "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl/ottlfuncs"
+
+import (
+	"context"
+	"fmt"
+	"strings"
+
+	"github.com/antchfx/xmlquery"
+	"github.com/antchfx/xpath"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl"
+)
+
+type ElementizeAttributesXMLArguments[K any] struct {
+	Target ottl.StringGetter[K]
+	XPath  ottl.Optional[string]
+}
+
+func NewElementizeAttributesXMLFactory[K any]() ottl.Factory[K] {
+	return ottl.NewFactory("ElementizeAttributesXML", &ElementizeAttributesXMLArguments[K]{}, createElementizeAttributesXMLFunction[K])
+}
+
+func createElementizeAttributesXMLFunction[K any](_ ottl.FunctionContext, oArgs ottl.Arguments) (ottl.ExprFunc[K], error) {
+	args, ok := oArgs.(*ElementizeAttributesXMLArguments[K])
+
+	if !ok {
+		return nil, fmt.Errorf("ElementizeAttributesXML args must be of type *ElementizeAttributesXMLAguments[K]")
+	}
+
+	xPath := args.XPath.Get()
+	if xPath == "" {
+		xPath = "//@*" // All attributes in the document
+	}
+
+	return elementizeAttributesXML(args.Target, xPath), nil
+}
+
+// elementizeAttributesXML returns a `pcommon.String` that is a result of converting all attributes of the
+// target XML into child elements. These new elements are added as the last child elements of the parent.
+// e.g. <a foo="bar" hello="world"><b/></a> -> <a><hello>world</hello><foo>bar</foo><b/></a>
+func elementizeAttributesXML[K any](target ottl.StringGetter[K], xPath string) ottl.ExprFunc[K] {
+	return func(ctx context.Context, tCtx K) (any, error) {
+		if err := validateXPath(xPath); err != nil {
+			return nil, err
+		}
+
+		targetVal, err := target.Get(ctx, tCtx)
+		if err != nil {
+			return nil, err
+		}
+
+		top, err := parseNodesXML(targetVal)
+		if err != nil {
+			return nil, err
+		}
+
+		xmlquery.FindEach(top, xPath, func(_ int, n *xmlquery.Node) {
+			if n.Type != xmlquery.AttributeNode {
+				return
+			}
+
+			xmlquery.AddChild(n.Parent, &xmlquery.Node{
+				Type: xmlquery.ElementNode,
+				Data: n.Data,
+				FirstChild: &xmlquery.Node{
+					Type: xmlquery.TextNode,
+					Data: n.InnerText(),
+				},
+			})
+
+			n.Parent.RemoveAttr(n.Data)
+		})
+
+		return top.OutputXML(false), nil
+	}
+}
+
+// TODO move these and use them in other xml functions
+func validateXPath(xPath string) error {
+	_, err := xpath.Compile(xPath)
+	if err != nil {
+		return fmt.Errorf("invalid xpath: %w", err)
+	}
+	return nil
+}
+
+// Aside from parsing the XML document, this function also ensures that
+// the XML declaration is included in the result only if it was present in
+// the original document.
+func parseNodesXML(targetVal string) (*xmlquery.Node, error) {
+	preserveDeclearation := strings.HasPrefix(targetVal, "<?xml")
+	top, err := xmlquery.Parse(strings.NewReader(targetVal))
+	if err != nil {
+		return nil, fmt.Errorf("parse xml: %w", err)
+	}
+	if !preserveDeclearation {
+		xmlquery.RemoveFromTree(top.FirstChild)
+	}
+	return top, nil
+}

pkg/ottl/ottlfuncs/func_elementize_attributes_xml_test.go

@@ -0,0 +1,136 @@
+// Copyright The OpenTelemetry Authors
+// SPDX-License-Identifier: Apache-2.0
+
+package ottlfuncs // import "github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl/ottlfuncs"
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl"
+)
+
+func Test_ElementizeAttributesXML(t *testing.T) {
+	tests := []struct {
+		name     string
+		document string
+		xPath    string
+		want     string
+	}{
+		{
+			name:     "nop",
+			document: `<a><b/></a>`,
+			want:     `<a><b></b></a>`,
+		},
+		{
+			name:     "nop declaration",
+			document: `<?xml version="1.0" encoding="UTF-8"?><a><b/></a>`,
+			want:     `<?xml version="1.0" encoding="UTF-8"?><a><b></b></a>`,
+		},
+		{
+			name:     "single attribute",
+			document: `<a foo="bar"/>`,
+			want:     `<a><foo>bar</foo></a>`,
+		},
+		{
+			name:     "multiple attributes - order 1",
+			document: `<a foo="bar" hello="world"/>`,
+			want:     `<a><foo>bar</foo><hello>world</hello></a>`,
+		},
+		{
+			name:     "multiple attributes - order 2",
+			document: `<a hello="world" foo="bar"/>`,
+			want:     `<a><hello>world</hello><foo>bar</foo></a>`,
+		},
+		{
+			name:     "with child elements",
+			document: `<a hello="world" foo="bar"><b/><c/><b/></a>`,
+			want:     `<a><b></b><c></c><b></b><hello>world</hello><foo>bar</foo></a>`,
+		},
+		{
+			name:     "with child value",
+			document: `<a hello="world" foo="bar">free value</a>`,
+			want:     `<a>free value<hello>world</hello><foo>bar</foo></a>`,
+		},
+		{
+			name:     "with child elements and values",
+			document: `<a hello="world" foo="bar">free value<b/>2<c/></a>`,
+			want:     `<a>free value<b></b>2<c></c><hello>world</hello><foo>bar</foo></a>`,
+		},
+		{
+			name:     "multiple levels",
+			document: `<a hello="world" foo="bar"><b href="www.example.com"></b></a>`,
+			want:     `<a><b><href>www.example.com</href></b><hello>world</hello><foo>bar</foo></a>`,
+		},
+		{
+			name:     "xpath filtered",
+			document: `<a hello="world" foo="bar"><b href="www.example.com"></b></a>`,
+			xPath:    "/a/b/@*", // only convert attributes of b
+			want:     `<a hello="world" foo="bar"><b><href>www.example.com</href></b></a>`,
+		},
+		{
+			name:     "attributes found with non-attributes xpath",
+			document: `<a hello="world" foo="bar"><b href="www.example.com"></b></a>`,
+			xPath:    "/a/b", // convert b (the attributes of b, even though the element b was selected)
+			want:     `<a hello="world" foo="bar"><b href="www.example.com"></b></a>`,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			args := &ElementizeAttributesXMLArguments[any]{
+				Target: ottl.StandardStringGetter[any]{
+					Getter: func(_ context.Context, _ any) (any, error) {
+						return tt.document, nil
+					},
+				},
+				XPath: ottl.NewTestingOptional(tt.xPath),
+			}
+			exprFunc, err := createElementizeAttributesXMLFunction[any](ottl.FunctionContext{}, args)
+			assert.NoError(t, err)
+
+			result, err := exprFunc(context.Background(), nil)
+			assert.NoError(t, err)
+			assert.Equal(t, tt.want, result)
+		})
+	}
+}
+
+func Test_ElementizeAttributesXML_InvalidXML(t *testing.T) {
+	exprFunc := elementizeAttributesXML(invalidXMLGetter(), "/foo")
+	_, err := exprFunc(context.Background(), nil)
+	assert.Error(t, err)
+}
+
+func Test_ElementizeAttributesXML_InvalidXPath(t *testing.T) {
+	exprFunc := elementizeAttributesXML(invalidXPathGetter(), "!")
+	_, err := exprFunc(context.Background(), nil)
+	assert.Error(t, err)
+}
+
+func TestCreateElementizeAttributesXMLFunc(t *testing.T) {
+	exprFunc, err := createElementizeAttributesXMLFunction[any](ottl.FunctionContext{}, &ElementizeAttributesXMLArguments[any]{})
+	assert.NoError(t, err)
+	assert.NotNil(t, exprFunc)
+
+	_, err = createElementizeAttributesXMLFunction[any](ottl.FunctionContext{}, nil)
+	assert.Error(t, err)
+}
+
+// TODO move these and use them in other xml tests
+func invalidXMLGetter() ottl.StandardStringGetter[any] {
+	return ottl.StandardStringGetter[any]{
+		Getter: func(_ context.Context, _ any) (any, error) {
+			return `<a>>>>>>>`, nil
+		},
+	}
+}
+
+func invalidXPathGetter() ottl.StandardStringGetter[any] {
+	return ottl.StandardStringGetter[any]{
+		Getter: func(_ context.Context, _ any) (any, error) {
+			return `<a></a>`, nil
+		},
+	}
+}

pkg/ottl/ottlfuncs/functions.go

@@ -43,6 +43,7 @@ func converters[K any]() []ottl.Factory[K] {
 		NewDayFactory[K](),
 		NewDoubleFactory[K](),
 		NewDurationFactory[K](),
+		NewElementizeAttributesXMLFactory[K](),
 		NewExtractPatternsFactory[K](),
 		NewExtractGrokPatternsFactory[K](),
 		NewFnvFactory[K](),