[pkg/ottl] Add ConvertTextToElementsXML Converter (open-telemetry#35364)

This adds a converter called `ConvertTextToElementsXML `. This serves as
one of the granular transformations described in
open-telemetry#35281
which will allow users to migrate any arbitrary XML document into a
JSON-equivalent state.

Also see open-telemetry#35328

Author: djaglowski

.chloggen/wrap-hanging-values-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 ConvertTextToElements Converter
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [35364]
+
+# (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

@@ -353,6 +353,18 @@ func Test_e2e_converters(t *testing.T) {
 				tCtx.GetLogRecord().Attributes().PutStr("test", `<Log><Message>This is a log message!</Message><id>1</id></Log>`)
 			},
 		},
+		{
+			statement: `set(body, ConvertTextToElementsXML("<a><b/>foo</a>"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				tCtx.GetLogRecord().Body().SetStr("<a><b></b><value>foo</value></a>")
+			},
+		},
+		{
+			statement: `set(body, ConvertTextToElementsXML("<a><b/>foo</a><c><b/>bar</c>", "/a", "custom"))`,
+			want: func(tCtx ottllog.TransformContext) {
+				tCtx.GetLogRecord().Body().SetStr("<a><b></b><custom>foo</custom></a><c><b></b>bar</c>")
+			},
+		},
 		{
 			statement: `set(attributes["test"], Double(1.0))`,
 			want: func(tCtx ottllog.TransformContext) {

pkg/ottl/ottlfuncs/README.md

@@ -414,6 +414,7 @@ Available Converters:
 - [Concat](#concat)
 - [ConvertCase](#convertcase)
 - [ConvertAttributesToElementsXML](#convertattributestoelementsxml)
+- [ConvertTextToElementsXML](#converttexttoelementsxml)
 - [Day](#day)
 - [Double](#double)
 - [Duration](#duration)
@@ -572,6 +573,36 @@ Convert only attributes within "Record" elements
 
 - `ConvertAttributesToElementsXML(body, "/Log/Record")`
 
+### ConvertTextToElementsXML
+
+`ConvertTextToElementsXML(target, Optional[xpath], Optional[elementName])`
+
+The `ConvertTextToElementsXML` Converter returns an edited version of an XML string where all text belongs to a dedicated element.
+
+`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, `ConvertTextToElementsXML` 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. Content will only be converted within the result(s) of the xpath. The default is `/`.
+
+`elementName` (optional) is a string that is used for any element tags that are created to wrap content.
+The default is `"value"`.
+
+For example, `<a><b>foo</b>bar</a>` will be converted to `<a><b>foo</b><value>bar</value></a>`.
+
+Examples:
+
+Ensure all text content in a document is wrapped in a dedicated element
+
+- `ConvertTextToElementsXML(body)`
+
+Use a custom name for any new elements
+
+- `ConvertTextToElementsXML(body, elementName = "custom")`
+
+Convert only part of the document
+
+- `ConvertTextToElementsXML(body, "/some/part/", "value")`
 
 ### Day
 

pkg/ottl/ottlfuncs/func_convert_text_to_elements_xml.go

@@ -0,0 +1,107 @@
+// 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"
+
+	"github.com/antchfx/xmlquery"
+
+	"github.com/open-telemetry/opentelemetry-collector-contrib/pkg/ottl"
+)
+
+type ConvertTextToElementsXMLArguments[K any] struct {
+	Target      ottl.StringGetter[K]
+	XPath       ottl.Optional[string]
+	ElementName ottl.Optional[string]
+}
+
+func NewConvertTextToElementsXMLFactory[K any]() ottl.Factory[K] {
+	return ottl.NewFactory("ConvertTextToElementsXML", &ConvertTextToElementsXMLArguments[K]{}, createConvertTextToElementsXMLFunction[K])
+}
+
+func createConvertTextToElementsXMLFunction[K any](_ ottl.FunctionContext, oArgs ottl.Arguments) (ottl.ExprFunc[K], error) {
+	args, ok := oArgs.(*ConvertTextToElementsXMLArguments[K])
+
+	if !ok {
+		return nil, fmt.Errorf("ConvertTextToElementsXML args must be of type *ConvertTextToElementsXMLAguments[K]")
+	}
+
+	xPath := args.XPath.Get()
+	if xPath == "" {
+		xPath = "/"
+	} else if err := validateXPath(xPath); err != nil {
+		return nil, err
+	}
+
+	elementName := args.ElementName.Get()
+	if elementName == "" {
+		elementName = "value"
+	}
+
+	return convertTextToElementsXML(args.Target, xPath, elementName), nil
+}
+
+// convertTextToElementsXML returns a string that is a result of wrapping any extraneous text nodes with a dedicated element.
+func convertTextToElementsXML[K any](target ottl.StringGetter[K], xPath string, elementName string) ottl.ExprFunc[K] {
+	return func(ctx context.Context, tCtx K) (any, error) {
+		var doc *xmlquery.Node
+		if targetVal, err := target.Get(ctx, tCtx); err != nil {
+			return nil, err
+		} else if doc, err = parseNodesXML(targetVal); err != nil {
+			return nil, err
+		}
+		for _, n := range xmlquery.Find(doc, xPath) {
+			convertTextToElementsForNode(n, elementName)
+		}
+		return doc.OutputXML(false), nil
+	}
+}
+
+func convertTextToElementsForNode(parent *xmlquery.Node, elementName string) {
+	switch parent.Type {
+	case xmlquery.ElementNode: // ok
+	case xmlquery.DocumentNode: // ok
+	default:
+		return
+	}
+
+	if parent.FirstChild == nil {
+		return
+	}
+
+	// Convert any child nodes and count text and element nodes.
+	var valueCount, elementCount int
+	for child := parent.FirstChild; child != nil; child = child.NextSibling {
+		if child.Type == xmlquery.ElementNode {
+			convertTextToElementsForNode(child, elementName)
+			elementCount++
+		} else if child.Type == xmlquery.TextNode {
+			valueCount++
+		}
+	}
+
+	// If there are no values to wrap, or if there is exactly one value OR one element, this node is all set.
+	if valueCount == 0 || elementCount+valueCount <= 1 {
+		return
+	}
+
+	// At this point, we either have multiple values, or a mix of values and elements.
+	// Either way, we need to wrap the values.
+	for child := parent.FirstChild; child != nil; child = child.NextSibling {
+		if child.Type != xmlquery.TextNode {
+			continue
+		}
+		newTextNode := &xmlquery.Node{
+			Type: xmlquery.TextNode,
+			Data: child.Data,
+		}
+		// Change this node into an element
+		child.Type = xmlquery.ElementNode
+		child.Data = elementName
+		child.FirstChild = newTextNode
+		child.LastChild = newTextNode
+	}
+}

pkg/ottl/ottlfuncs/func_convert_text_to_elements_xml_test.go

@@ -0,0 +1,127 @@
+// 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_ConvertTextToElementsXML(t *testing.T) {
+	tests := []struct {
+		name        string
+		document    string
+		xPath       string
+		elementName 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:     "nop attributes",
+			document: `<a foo="bar" hello="world"/>`,
+			want:     `<a foo="bar" hello="world"></a>`,
+		},
+		{
+			name:     "nop wrapped text",
+			document: `<a>hello world</a>`,
+			want:     `<a>hello world</a>`,
+		},
+		{
+			name:     "simple hanging",
+			document: `<a><b/>foo</a>`,
+			want:     `<a><b></b><value>foo</value></a>`,
+		},
+		{
+			name:        "simple hanging with tag name",
+			elementName: "bar",
+			document:    `<a><b/>foo</a>`,
+			want:        `<a><b></b><bar>foo</bar></a>`,
+		},
+		{
+			name:     "multiple hanging same level",
+			document: `<a>foo<b/>bar</a>`,
+			want:     `<a><value>foo</value><b></b><value>bar</value></a>`,
+		},
+		{
+			name:        "multiple hanging multiple levels",
+			document:    `<a>foo<b/>bar<c/>1<d>not</d>2<e><f/><f/></e></a>`,
+			elementName: "v",
+			want:        `<a><v>foo</v><b></b><v>bar</v><c></c><v>1</v><d>not</d><v>2</v><e><f></f><f></f></e></a>`,
+		},
+		{
+			name:     "xpath select some",
+			document: `<a><b><c/>foo</b><d><c/>bar</d><b><c/>baz</b></a>`,
+			xPath:    "/a/b",
+			want:     `<a><b><c></c><value>foo</value></b><d><c></c>bar</d><b><c></c><value>baz</value></b></a>`,
+		},
+		{
+			name:        "xpath with element name",
+			document:    `<a><b><c/>foo</b><d><c/>bar</d><b><c/>baz</b></a>`,
+			xPath:       "/a/b",
+			elementName: "V",
+			want:        `<a><b><c></c><V>foo</V></b><d><c></c>bar</d><b><c></c><V>baz</V></b></a>`,
+		},
+	}
+	factory := NewConvertTextToElementsXMLFactory[any]()
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			args := &ConvertTextToElementsXMLArguments[any]{
+				Target: ottl.StandardStringGetter[any]{
+					Getter: func(_ context.Context, _ any) (any, error) {
+						return tt.document, nil
+					},
+				},
+				XPath:       ottl.NewTestingOptional(tt.xPath),
+				ElementName: ottl.NewTestingOptional(tt.elementName),
+			}
+			exprFunc, err := factory.CreateFunction(ottl.FunctionContext{}, args)
+			assert.NoError(t, err)
+
+			result, err := exprFunc(context.Background(), nil)
+			assert.NoError(t, err)
+			assert.Equal(t, tt.want, result)
+		})
+	}
+}
+
+func TestCreateConvertTextToElementsXMLFunc(t *testing.T) {
+	factory := NewConvertTextToElementsXMLFactory[any]()
+	fCtx := ottl.FunctionContext{}
+
+	// Invalid arg type
+	exprFunc, err := factory.CreateFunction(fCtx, nil)
+	assert.Error(t, err)
+	assert.Nil(t, exprFunc)
+
+	// Invalid XPath should error on function creation
+	exprFunc, err = factory.CreateFunction(
+		fCtx, &ConvertTextToElementsXMLArguments[any]{
+			XPath: ottl.NewTestingOptional("!"),
+		})
+	assert.Error(t, err)
+	assert.Nil(t, exprFunc)
+
+	// Invalid XML should error on function execution
+	exprFunc, err = factory.CreateFunction(
+		fCtx, &ConvertTextToElementsXMLArguments[any]{
+			Target: invalidXMLGetter(),
+		})
+	assert.NoError(t, err)
+	assert.NotNil(t, exprFunc)
+	_, err = exprFunc(context.Background(), nil)
+	assert.Error(t, err)
+}

pkg/ottl/ottlfuncs/functions.go

@@ -41,6 +41,7 @@ func converters[K any]() []ottl.Factory[K] {
 		NewConcatFactory[K](),
 		NewConvertCaseFactory[K](),
 		NewConvertAttributesToElementsXMLFactory[K](),
+		NewConvertTextToElementsXMLFactory[K](),
 		NewDayFactory[K](),
 		NewDoubleFactory[K](),
 		NewDurationFactory[K](),