Financial-Times · lokendersinghft · Sep 29, 2025 · Sep 25, 2025 · todor-videv1 · Sep 25, 2025
@@ -1,3 +1,7 @@
 module github.com/Financial-Times/content-tree
 
-go 1.23.5
+go 1.25
+
+require github.com/beevik/etree v1.6.0
+
+require github.com/nsf/jsondiff v0.0.0-20230430225905-43f6cf3098c1
@@ -0,0 +1,4 @@
+github.com/beevik/etree v1.6.0 h1:u8Kwy8pp9D9XeITj2Z0XtA5qqZEmtJtuXZRQi+j03eE=
+github.com/beevik/etree v1.6.0/go.mod h1:bh4zJxiIr62SOf9pRzN7UUYaEDa9HEKafK25+sLc0Gc=
+github.com/nsf/jsondiff v0.0.0-20230430225905-43f6cf3098c1 h1:dOYG7LS/WK00RWZc8XGgcUTlTxpp3mKhdR2Q9z9HbXM=
+github.com/nsf/jsondiff v0.0.0-20230430225905-43f6cf3098c1/go.mod h1:mpRZBD8SJ55OIICQ3iWH0Yz3cjzA61JdqMLoWXeB2+8=
@@ -0,0 +1,31 @@
+# XML to Content Tree Transformer
+
+## Overview
+The Transformer converts external XHTML-formatted document into content tree.  
+It supports format stored in the **internalComponent** collection as well as the one returned by the **Internal Content API**.
+The latter is produced by the content-public-read service after applying certain transformations to the bodyXML it retrieves from the internalComponents collection.
+These transformations include renaming the content, related, and concept tags to ft-content, ft-related, and ft-concept, respectively, and replacing the id attribute with url, with a few caveats.
+
+## Usage
+
+```go
+package main
+
+import (
+    "fmt"
+    "log"
+
+    tocontenttree "github.com/Financial-Times/content-tree"
+)
+
+func main() {
+    xmlInput := `<body><p>Hello World</p></body>`
+
+    out, err := tocontenttree.Transform(xmlInput)
+    if err != nil {
+        log.Fatalf("Transform (XmlToTree) failed: %v", err)
+    }
+
+    fmt.Printf("Transformed content tree: %+v\n", out)
+}
+```
@@ -0,0 +1,104 @@
+package tocontenttree
+
+import (
+	"strings"
+
+	"github.com/beevik/etree"
+)
+
+type layoutwidth string
+
+func toValidLayoutWidth(w string) layoutwidth {
+	switch w {
+	case "auto", "in-line", "inset-left", "inset-right",
+		"full-bleed", "full-grid", "mid-grid", "full-width":
+		return layoutwidth(w)
+	default:
+		return "full-width"
+	}
+}
+
+func toValidClipLayoutWidth(w string) layoutwidth {
+	switch w {
+	case "in-line", "full-grid", "mid-grid":
+		return layoutwidth(w)
+	default:
+		return "in-line"
+	}
+
+}
+
+func findChild(el *etree.Element, tag string) *etree.Element {
+	for _, ch := range el.ChildElements() {
+		if ch.Tag == tag {
+			return ch
+		}
+		if found := findChild(ch, tag); found != nil {
+			return found
+		}
+	}
+	return nil
+}
+
+func textContent(el *etree.Element) string {
+	var b strings.Builder
+	for _, tok := range el.Child {
+		switch t := tok.(type) {
+		case *etree.CharData:
+			b.WriteString(t.Data)
+		case *etree.Element:
+			b.WriteString(textContent(t))
+		}
+	}
+	return b.String()
+}
+
+func flattenedChildren(el *etree.Element) []etree.Token {
+	out := make([]etree.Token, 0, len(el.Child))
+	for _, tok := range el.Child {
+		if d, ok := tok.(*etree.Element); ok && d.Tag == "div" {
+			out = append(out, d.Child...)
+		} else {
+			out = append(out, tok)
+		}
+	}
+	return out
+}
+
+func valueOr(v, fallback string) string {
+	if v != "" {
+		return v
+	}
+	return fallback
+}
+
+func attr(el *etree.Element, name string) string {
+	return el.SelectAttrValue(name, "")
+}
+
+var contentTypeTemplates = map[string]string{
+	"http://www.ft.com/ontology/content/Article":             "/content/{{id}}",
+	"http://www.ft.com/ontology/content/ImageSet":            "/content/{{id}}",
+	"http://www.ft.com/ontology/content/ClipSet":             "/content/{{id}}",
+	"http://www.ft.com/ontology/content/CustomCodeComponent": "/content/{{id}}",
+	"http://www.ft.com/ontology/content/MediaResource":       "/content/{{id}}",
+	"http://www.ft.com/ontology/content/Video":               "/content/{{id}}",
+	"http://www.ft.com/ontology/company/PublicCompany":       "/organisations/{{id}}",
+	"http://www.ft.com/ontology/content/ContentPackage":      "/content/{{id}}",
+	"http://www.ft.com/ontology/content/Content":             "/content/{{id}}",
+	"http://www.ft.com/ontology/content/Image":               "/content/{{id}}",
+	"http://www.ft.com/ontology/content/DynamicContent":      "/content/{{id}}",
+	"http://www.ft.com/ontology/content/Graphic":             "/content/{{id}}",
+	"http://www.ft.com/ontology/content/Audio":               "/content/{{id}}",
+	"http://www.ft.com/ontology/company/Organisation":        "/organisations/{{id}}",
+}
+
+func generateUrl(t, id string) string {
+	const host = "http://api.ft.com"
+	template, ok := contentTypeTemplates[t]
+	if !ok {
+		return ""
+	}
+	path := strings.Replace(template, "{{id}}", id, 1)
+	return host + path
+}