Files
insertr/internal/engine/utils.go
Joakim 27179dc943 refactor: remove legacy parser system and migrate to unified engine
- Remove internal/parser package and all legacy ID generation logic
- Update enhancer and auto_enhancer to use unified engine functions
- Migrate utility functions (FindViableChildren, HasEditableContent) to engine
- Create stub enhancer implementation that uses unified engine architecture
- Ensure all enhancement workflows now go through single unified system
- Remove parser dependencies and consolidate content processing logic

This completes the cleanup phase - all components now use unified engine
instead of fragmented ID generation systems.
2025-09-16 15:18:40 +02:00

316 lines
7.5 KiB
Go

package engine
import (
"strings"
"golang.org/x/net/html"
)
// GetClasses extracts CSS classes from an HTML node
func GetClasses(node *html.Node) []string {
classAttr := getAttribute(node, "class")
if classAttr == "" {
return []string{}
}
classes := strings.Fields(classAttr)
return classes
}
// ContainsClass checks if a class list contains a specific class
func ContainsClass(classes []string, target string) bool {
for _, class := range classes {
if class == target {
return true
}
}
return false
}
// getAttribute gets an attribute value from an HTML node
func getAttribute(node *html.Node, key string) string {
for _, attr := range node.Attr {
if attr.Key == key {
return attr.Val
}
}
return ""
}
// extractTextContent gets the text content from an HTML node
func extractTextContent(node *html.Node) string {
var text strings.Builder
extractTextRecursive(node, &text)
return strings.TrimSpace(text.String())
}
// extractTextRecursive recursively extracts text from node and children
func extractTextRecursive(node *html.Node, text *strings.Builder) {
if node.Type == html.TextNode {
text.WriteString(node.Data)
}
for child := node.FirstChild; child != nil; child = child.NextSibling {
// Skip script and style elements
if child.Type == html.ElementNode &&
(child.Data == "script" || child.Data == "style") {
continue
}
extractTextRecursive(child, text)
}
}
// hasOnlyTextContent checks if a node contains only text content (no nested HTML elements)
// DEPRECATED: Use hasEditableContent for more sophisticated detection
func hasOnlyTextContent(node *html.Node) bool {
if node.Type != html.ElementNode {
return false
}
for child := node.FirstChild; child != nil; child = child.NextSibling {
switch child.Type {
case html.ElementNode:
// Found a nested HTML element - not text-only
return false
case html.TextNode:
// Text nodes are fine, continue checking
continue
default:
// Comments, etc. - continue checking
continue
}
}
return true
}
// Inline formatting elements that are safe for editing
var inlineFormattingTags = map[string]bool{
"strong": true,
"b": true,
"em": true,
"i": true,
"span": true,
"code": true,
"small": true,
"sub": true,
"sup": true,
"a": true, // Links within content are fine
}
// Elements that should NOT be nested within editable content
var blockingElements = map[string]bool{
"button": true, // Buttons shouldn't be nested in paragraphs
"input": true,
"select": true,
"textarea": true,
"img": true,
"video": true,
"audio": true,
"canvas": true,
"svg": true,
"iframe": true,
"object": true,
"embed": true,
"div": true, // Nested divs usually indicate complex structure
"section": true, // Block-level semantic elements
"article": true,
"header": true,
"footer": true,
"nav": true,
"aside": true,
"main": true,
"form": true,
"table": true,
"ul": true,
"ol": true,
"dl": true,
}
// hasEditableContent checks if a node contains content that can be safely edited
// This includes text and safe inline formatting elements
func hasEditableContent(node *html.Node) bool {
if node.Type != html.ElementNode {
return false
}
return hasOnlyTextAndSafeFormatting(node)
}
// hasOnlyTextAndSafeFormatting recursively checks if content is safe for editing
func hasOnlyTextAndSafeFormatting(node *html.Node) bool {
for child := node.FirstChild; child != nil; child = child.NextSibling {
switch child.Type {
case html.TextNode:
continue // Text is always safe
case html.ElementNode:
// Check if it's a blocking element
if blockingElements[child.Data] {
return false
}
// Allow safe inline formatting
if inlineFormattingTags[child.Data] {
// Recursively validate the formatting element
if !hasOnlyTextAndSafeFormatting(child) {
return false
}
continue
}
// Unknown/unsafe element
return false
default:
continue // Comments, whitespace, etc.
}
}
return true
}
// isContainer checks if a tag is typically used as a container element
func isContainer(node *html.Node) bool {
if node.Type != html.ElementNode {
return false
}
containerTags := map[string]bool{
"div": true,
"section": true,
"article": true,
"header": true,
"footer": true,
"main": true,
"aside": true,
"nav": true,
}
return containerTags[node.Data]
}
// findViableChildren finds all child elements that are viable for editing
func findViableChildren(node *html.Node) []*html.Node {
var viable []*html.Node
for child := node.FirstChild; child != nil; child = child.NextSibling {
// Skip whitespace-only text nodes
if child.Type == html.TextNode {
if strings.TrimSpace(child.Data) == "" {
continue
}
}
// Only consider element nodes
if child.Type != html.ElementNode {
continue
}
// Skip self-closing elements for now
if isSelfClosing(child) {
continue
}
// Check if element has editable content (improved logic)
if hasEditableContent(child) {
viable = append(viable, child)
}
}
return viable
}
// findViableChildrenLegacy uses the old text-only logic for backwards compatibility
func findViableChildrenLegacy(node *html.Node) []*html.Node {
var viable []*html.Node
for child := node.FirstChild; child != nil; child = child.NextSibling {
if child.Type == html.TextNode {
if strings.TrimSpace(child.Data) == "" {
continue
}
}
if child.Type != html.ElementNode {
continue
}
if isSelfClosing(child) {
continue
}
if hasOnlyTextContent(child) {
viable = append(viable, child)
}
}
return viable
}
// isSelfClosing checks if an element is typically self-closing
func isSelfClosing(node *html.Node) bool {
if node.Type != html.ElementNode {
return false
}
selfClosingTags := map[string]bool{
"img": true,
"input": true,
"br": true,
"hr": true,
"meta": true,
"link": true,
"area": true,
"base": true,
"col": true,
"embed": true,
"source": true,
"track": true,
"wbr": true,
}
return selfClosingTags[node.Data]
}
// FindElementInDocument finds an element in HTML document tree using content matching
func FindElementInDocument(doc *html.Node, tag, content string) *html.Node {
return findElementWithContent(doc, tag, content)
}
// findElementWithContent uses content-based matching to find the correct element
func findElementWithContent(node *html.Node, targetTag, targetContent string) *html.Node {
normalizedTarget := strings.TrimSpace(targetContent)
if node.Type == html.ElementNode && node.Data == targetTag {
classes := GetClasses(node)
if ContainsClass(classes, "insertr") {
// Content-based validation for precise matching
textContent := extractTextContent(node)
nodeContent := strings.TrimSpace(textContent)
if nodeContent == normalizedTarget {
return node
}
}
}
// Recursively search children
for child := node.FirstChild; child != nil; child = child.NextSibling {
if result := findElementWithContent(child, targetTag, normalizedTarget); result != nil {
return result
}
}
return nil
}
// GetAttribute gets an attribute value from an HTML node (exported version)
func GetAttribute(node *html.Node, key string) string {
return getAttribute(node, key)
}
// HasEditableContent checks if a node has editable content (exported version)
func HasEditableContent(node *html.Node) bool {
return hasEditableContent(node)
}
// FindViableChildren finds viable children for editing (exported version)
func FindViableChildren(node *html.Node) []*html.Node {
return findViableChildren(node)
}