refactor: highlight code blocks function

src/routes/question.go

@@ -180,7 +180,7 @@ func extractQuestionData(doc *goquery.Document, domain string) (question types.F
 	if err != nil {
 		return question, err
 	}
-	question.Body = template.HTML(utils.ReplaceImgTags(questionBodyParentHTML))
+	question.Body = template.HTML(processHTMLBody(questionBodyParentHTML))
 
 	// Extract the shortened body description.
 	shortenedBody := strings.TrimSpace(questionBodyParent.Text())
@@ -212,8 +212,6 @@ func extractMetadata(selection *goquery.Selection, question *types.FilteredQuest
 	questionAuthorURL += questionAuthor.AttrOr("href", "")
 	question.AuthorURL = questionAuthorURL
 
-	fmt.Printf("Author name is: %s\n", question.AuthorName)
-
 	// Determine if the question has been edited and update author details accordingly.
 	isQuestionEdited := selection.Find("a.js-gps-track").Text() == "edited"
 	if isQuestionEdited {
@@ -247,9 +245,7 @@ func extractAnswersData(doc *goquery.Document, domain string) ([]types.FilteredA
 		answerBodyHTML, _ := answerBody.Html()
 
 		// Process code blocks within the answer.
-		processedAnswerBody := processAnswerBody(answerBodyHTML, domain)
+		processedAnswerBody := processHTMLBody(answerBodyHTML)
-		processedAnswerBody = utils.ReplaceImgTags(processedAnswerBody)
-		fmt.Println(processedAnswerBody)
 		answer.Body = template.HTML(html.UnescapeString(processedAnswerBody))
 
 		// Extract author information and timestamp.
@@ -261,10 +257,11 @@ func extractAnswersData(doc *goquery.Document, domain string) ([]types.FilteredA
 	return answers, nil
 }
 
-// processAnswerBody highlights syntax and processes code blocks within an answer's body.
+// processHTMLBody highlights syntax and replaces images with proxied versions.
-func processAnswerBody(bodyHTML string, domain string) string {
+func processHTMLBody(bodyHTML string) string {
-	highlightedBody := utils.HighlightSyntaxViaContent(bodyHTML)
+	highlightedBody := utils.HighlightCodeBlocks(bodyHTML)
-	return highlightedBody
+	imageProxiedBody := utils.ReplaceImgTags(highlightedBody)
+	return imageProxiedBody
 }
 
 // extractAnswerAuthorInfo extracts the author name, URL, and timestamp from an answer block.

src/utils/syntax.go

@@ -12,7 +12,9 @@ import (
 	"github.com/alecthomas/chroma/styles"
 )
 
-func HighlightSyntaxViaContent(content string) (htmlOut string) {
+// highlightSyntaxViaContent uses Chroma to lex code content and apply the appropriate tokenizer engine.
+// If it can't find one, it defaults to JavaScript syntax highlighting.
+func highlightSyntaxViaContent(content string) (htmlOut string) {
 	content = html.UnescapeString(content)
 
 	fallbackOut := html.EscapeString(content)
@@ -20,9 +22,7 @@ func HighlightSyntaxViaContent(content string) (htmlOut string) {
 	// identify the language
 	lexer := lexers.Analyse(content)
 	if lexer == nil {
-		// unable to identify, so just return the wrapped content
+		lexer = lexers.Get(".js")
-		htmlOut = fallbackOut
-		return
 	}
 
 	style := styles.Get("xcode")
@@ -54,7 +54,9 @@ func HighlightSyntaxViaContent(content string) (htmlOut string) {
 
 var preClassRegex = regexp.MustCompile(`(?s)<pre class=".+">`)
 
-func StripBlockTags(content string) (result string) {
+// stripBlockTags takes an extracted code block from HTML and strips it of its pre and code tags.
+// What's returned is just the code.
+func stripBlockTags(content string) (result string) {
 	// strip all "<code>" tags
 	content = strings.Replace(content, "<code>", "", -1)
 	content = strings.Replace(content, "</code>", "", -1)
@@ -68,3 +70,26 @@ func StripBlockTags(content string) (result string) {
 
 	return
 }
+
+var codeBlockRegex = regexp.MustCompile(`(?s)<pre><code>(.*?)<\/code><\/pre>`)
+
+// HighlightCodeBlocks uses both highlightSyntaxViaContent stripCodeBlocks and returns the newly highlighted code HTML.
+func HighlightCodeBlocks(html string) string {
+	// Replace each code block with the highlighted version
+	highlightedHTML := codeBlockRegex.ReplaceAllStringFunc(html, func(codeBlock string) string {
+		// Extract the code content from the code block
+		codeContent := codeBlockRegex.FindStringSubmatch(codeBlock)[1]
+
+		codeContent = stripBlockTags(codeContent)
+
+		// Highlight the code content
+		highlightedCode := highlightSyntaxViaContent(codeContent)
+
+		// Replace the original code block with the highlighted version
+		highlightedCodeBlock := "<pre>" + highlightedCode + "</pre>"
+
+		return highlightedCodeBlock
+	})
+
+	return highlightedHTML
+}