feat: implement minimal server-first markdown processing

Backend implementation:
- Add goldmark dependency for markdown processing
- Create MarkdownProcessor with minimal config (bold, italic, links only)
- Update content injector with HTML injection capabilities
- Add injectHTMLContent() for safe DOM manipulation
- Server now converts **bold**, *italic*, [links](url) to HTML during enhancement

Frontend alignment:
- Restrict marked.js to match server capabilities
- Disable unsupported features (headings, lists, code blocks, tables)
- Update turndown rules to prevent unsupported markdown generation
- Frontend editor preview now matches server output exactly

Server as source of truth:
- Build-time markdown→HTML conversion during enhancement
- Zero runtime overhead for end users
- Consistent formatting between editor preview and final output
- Raw markdown stored in database, HTML served to visitors

Tested features:
- **bold** → <strong>bold</strong> 
- *italic* → <em>italic</em> 
- [text](url) → <a href="url">text</a> 
This commit is contained in:
2025-09-11 16:43:40 +02:00
parent 3db1340cce
commit 350c3f6160
6 changed files with 256 additions and 30 deletions

1
go.mod
View File

@@ -26,6 +26,7 @@ require (
github.com/spf13/cast v1.6.0 // indirect github.com/spf13/cast v1.6.0 // indirect
github.com/spf13/pflag v1.0.5 // indirect github.com/spf13/pflag v1.0.5 // indirect
github.com/subosito/gotenv v1.6.0 // indirect github.com/subosito/gotenv v1.6.0 // indirect
github.com/yuin/goldmark v1.7.8 // indirect
go.uber.org/atomic v1.9.0 // indirect go.uber.org/atomic v1.9.0 // indirect
go.uber.org/multierr v1.9.0 // indirect go.uber.org/multierr v1.9.0 // indirect
golang.org/x/exp v0.0.0-20230905200255-921286631fa9 // indirect golang.org/x/exp v0.0.0-20230905200255-921286631fa9 // indirect

2
go.sum
View File

@@ -63,6 +63,8 @@ github.com/stretchr/testify v1.8.4 h1:CcVxjf3Q8PM0mHUKJCdn+eZZtm5yQwehR5yeSVQQcU
github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo= github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo=
github.com/subosito/gotenv v1.6.0 h1:9NlTDc1FTs4qu0DDq7AEtTPNw6SVm7uBMsUCUjABIf8= github.com/subosito/gotenv v1.6.0 h1:9NlTDc1FTs4qu0DDq7AEtTPNw6SVm7uBMsUCUjABIf8=
github.com/subosito/gotenv v1.6.0/go.mod h1:Dk4QP5c2W3ibzajGcXpNraDfq2IrhjMIvMSWPKKo0FU= github.com/subosito/gotenv v1.6.0/go.mod h1:Dk4QP5c2W3ibzajGcXpNraDfq2IrhjMIvMSWPKKo0FU=
github.com/yuin/goldmark v1.7.8 h1:iERMLn0/QJeHFhxSt3p6PeN9mGnvIKSpG9YYorDMnic=
github.com/yuin/goldmark v1.7.8/go.mod h1:uzxRWxtg69N339t3louHJ7+O03ezfj6PlliRlaOzY1E=
go.uber.org/atomic v1.9.0 h1:ECmE8Bn/WFTYwEW/bpKD3M8VtR/zQVbavAoalC1PYyE= go.uber.org/atomic v1.9.0 h1:ECmE8Bn/WFTYwEW/bpKD3M8VtR/zQVbavAoalC1PYyE=
go.uber.org/atomic v1.9.0/go.mod h1:fEN4uk6kAWBTFdckzkM89CLk9XfWZrxpCo0nPH17wJc= go.uber.org/atomic v1.9.0/go.mod h1:fEN4uk6kAWBTFdckzkM89CLk9XfWZrxpCo0nPH17wJc=
go.uber.org/multierr v1.9.0 h1:7fIwc/ZtS0q++VgcfqFDxSBZVv/Xo49/SYnDFupUwlI= go.uber.org/multierr v1.9.0 h1:7fIwc/ZtS0q++VgcfqFDxSBZVv/Xo49/SYnDFupUwlI=

View File

@@ -108,7 +108,7 @@ func (e *Enhancer) findAndInjectNodes(rootNode *html.Node, elem parser.Element,
} }
// Inject content attributes for the correctly matched node // Inject content attributes for the correctly matched node
e.injector.addContentAttributes(targetNode, elem.ContentID, string(elem.Type)) e.injector.AddContentAttributes(targetNode, elem.ContentID, string(elem.Type))
// Inject content if available // Inject content if available
if contentItem != nil { if contentItem != nil {

View File

@@ -2,6 +2,7 @@ package content
import ( import (
"fmt" "fmt"
"log"
"strings" "strings"
"golang.org/x/net/html" "golang.org/x/net/html"
@@ -9,15 +10,17 @@ import (
// Injector handles content injection into HTML elements // Injector handles content injection into HTML elements
type Injector struct { type Injector struct {
client ContentClient client ContentClient
siteID string siteID string
mdProcessor *MarkdownProcessor
} }
// NewInjector creates a new content injector // NewInjector creates a new content injector
func NewInjector(client ContentClient, siteID string) *Injector { func NewInjector(client ContentClient, siteID string) *Injector {
return &Injector{ return &Injector{
client: client, client: client,
siteID: siteID, siteID: siteID,
mdProcessor: NewMarkdownProcessor(),
} }
} }
@@ -31,7 +34,7 @@ func (i *Injector) InjectContent(element *Element, contentID string) error {
// If no content found, keep original content but add data attributes // If no content found, keep original content but add data attributes
if contentItem == nil { if contentItem == nil {
i.addContentAttributes(element.Node, contentID, element.Type) i.AddContentAttributes(element.Node, contentID, element.Type)
return nil return nil
} }
@@ -48,7 +51,7 @@ func (i *Injector) InjectContent(element *Element, contentID string) error {
} }
// Add data attributes for editor functionality // Add data attributes for editor functionality
i.addContentAttributes(element.Node, contentID, element.Type) i.AddContentAttributes(element.Node, contentID, element.Type)
return nil return nil
} }
@@ -72,7 +75,7 @@ func (i *Injector) InjectBulkContent(elements []ElementWithID) error {
contentItem, exists := contentMap[elem.ContentID] contentItem, exists := contentMap[elem.ContentID]
// Add content attributes regardless // Add content attributes regardless
i.addContentAttributes(elem.Element.Node, elem.ContentID, elem.Element.Type) i.AddContentAttributes(elem.Element.Node, elem.ContentID, elem.Element.Type)
if !exists { if !exists {
// Keep original content if not found in database // Keep original content if not found in database
@@ -112,11 +115,23 @@ func (i *Injector) injectTextContent(node *html.Node, content string) {
node.AppendChild(textNode) node.AppendChild(textNode)
} }
// injectMarkdownContent handles markdown content (for now, just as text) // injectMarkdownContent handles markdown content - converts markdown to HTML
func (i *Injector) injectMarkdownContent(node *html.Node, content string) { func (i *Injector) injectMarkdownContent(node *html.Node, content string) {
// For now, treat markdown as text content if content == "" {
// TODO: Implement markdown to HTML conversion i.injectTextContent(node, "")
i.injectTextContent(node, content) return
}
// Convert markdown to HTML using server processor
htmlContent, err := i.mdProcessor.ToHTML(content)
if err != nil {
log.Printf("⚠️ Markdown conversion failed for content '%s': %v, falling back to text", content, err)
i.injectTextContent(node, content)
return
}
// Inject the HTML content
i.injectHTMLContent(node, htmlContent)
} }
// injectLinkContent handles link/button content with URL extraction // injectLinkContent handles link/button content with URL extraction
@@ -126,8 +141,68 @@ func (i *Injector) injectLinkContent(node *html.Node, content string) {
i.injectTextContent(node, content) i.injectTextContent(node, content)
} }
// addContentAttributes adds necessary data attributes and insertr class for editor functionality // injectHTMLContent safely injects HTML content into a DOM node
func (i *Injector) addContentAttributes(node *html.Node, contentID string, contentType string) { func (i *Injector) injectHTMLContent(node *html.Node, htmlContent string) {
// Clear existing content
i.clearNode(node)
if htmlContent == "" {
return
}
// Wrap content to create valid HTML document for parsing
wrappedHTML := "<div>" + htmlContent + "</div>"
// Parse HTML string
doc, err := html.Parse(strings.NewReader(wrappedHTML))
if err != nil {
log.Printf("Failed to parse HTML content '%s': %v, falling back to text", htmlContent, err)
i.injectTextContent(node, htmlContent)
return
}
// Find the wrapper div and move its children to target node
wrapper := i.findElementByTag(doc, "div")
if wrapper == nil {
log.Printf("Could not find wrapper div in parsed HTML")
return
}
// Move parsed nodes to target element
for child := wrapper.FirstChild; child != nil; {
next := child.NextSibling
wrapper.RemoveChild(child)
node.AppendChild(child)
child = next
}
}
// clearNode removes all child nodes from a given node
func (i *Injector) clearNode(node *html.Node) {
for child := node.FirstChild; child != nil; {
next := child.NextSibling
node.RemoveChild(child)
child = next
}
}
// findElementByTag finds the first element with the specified tag name
func (i *Injector) findElementByTag(node *html.Node, tag string) *html.Node {
if node.Type == html.ElementNode && node.Data == tag {
return node
}
for child := node.FirstChild; child != nil; child = child.NextSibling {
if found := i.findElementByTag(child, tag); found != nil {
return found
}
}
return nil
}
// AddContentAttributes adds necessary data attributes and insertr class for editor functionality
func (i *Injector) AddContentAttributes(node *html.Node, contentID string, contentType string) {
i.setAttribute(node, "data-content-id", contentID) i.setAttribute(node, "data-content-id", contentID)
i.setAttribute(node, "data-content-type", contentType) i.setAttribute(node, "data-content-type", contentType)
i.addClass(node, "insertr") i.addClass(node, "insertr")

View File

@@ -0,0 +1,67 @@
package content
import (
"bytes"
"log"
"github.com/yuin/goldmark"
"github.com/yuin/goldmark/parser"
"github.com/yuin/goldmark/renderer/html"
"github.com/yuin/goldmark/util"
)
// MarkdownProcessor handles minimal markdown processing
// Supports only: **bold**, *italic*, and [link](url)
type MarkdownProcessor struct {
parser goldmark.Markdown
}
// NewMarkdownProcessor creates a new markdown processor with minimal configuration
func NewMarkdownProcessor() *MarkdownProcessor {
// Configure goldmark to only support basic inline formatting
md := goldmark.New(
goldmark.WithParserOptions(
parser.WithInlineParsers(
// Bold (**text**) and italic (*text*) - same parser handles both
util.Prioritized(parser.NewEmphasisParser(), 500),
// Links [text](url)
util.Prioritized(parser.NewLinkParser(), 600),
),
// Disable all block parsers except paragraph (no headings, lists, etc.)
parser.WithBlockParsers(
util.Prioritized(parser.NewParagraphParser(), 200),
),
),
goldmark.WithRendererOptions(
html.WithXHTML(), // <br /> instead of <br>
html.WithHardWraps(), // Line breaks become <br />
html.WithUnsafe(), // Allow existing HTML to pass through
),
)
return &MarkdownProcessor{parser: md}
}
// ToHTML converts markdown string to HTML
func (mp *MarkdownProcessor) ToHTML(markdown string) (string, error) {
if markdown == "" {
return "", nil
}
var buf bytes.Buffer
if err := mp.parser.Convert([]byte(markdown), &buf); err != nil {
log.Printf("Markdown conversion failed: %v", err)
return "", err
}
html := buf.String()
// Clean up goldmark's paragraph wrapping - we want inline content
// Remove <p> and </p> tags if the content is wrapped in a single paragraph
if len(html) > 7 && html[:3] == "<p>" && html[len(html)-4:] == "</p>" {
html = html[3 : len(html)-4]
}
return html, nil
}

View File

@@ -14,32 +14,75 @@ export class MarkdownConverter {
} }
/** /**
* Configure marked for HTML output * Configure marked for HTML output - MINIMAL MODE
* Only supports: **bold**, *italic*, and [links](url)
* Matches server-side goldmark configuration
*/ */
initializeMarked() { initializeMarked() {
marked.setOptions({ marked.setOptions({
gfm: true, // GitHub Flavored Markdown gfm: false, // Disable GFM to match server minimal mode
breaks: true, // Convert \n to <br> breaks: true, // Convert \n to <br> (matches server)
pedantic: false, // Don't be overly strict pedantic: false, // Don't be overly strict
sanitize: false, // Allow HTML (we control the input) sanitize: false, // Allow HTML (we control the input)
smartLists: true, // Smarter list behavior smartLists: false, // Disable lists (not supported on server)
smartypants: false // Don't convert quotes/dashes smartypants: false // Don't convert quotes/dashes
}); });
// Override renderers to restrict to minimal feature set
marked.use({
renderer: {
// Disable headings - treat as plain text
heading(text, level) {
return text;
},
// Disable lists - treat as plain text
list(body, ordered, start) {
return body.replace(/<\/?li>/g, '');
},
listitem(text) {
return text + '\n';
},
// Disable code blocks - treat as plain text
code(code, language) {
return code;
},
blockquote(quote) {
return quote; // Disable blockquotes - treat as plain text
},
// Disable horizontal rules
hr() {
return '';
},
// Disable tables
table(header, body) {
return header + body;
},
tablecell(content, flags) {
return content;
},
tablerow(content) {
return content;
}
}
});
} }
/** /**
* Configure turndown for markdown output * Configure turndown for markdown output - MINIMAL MODE
* Only supports: **bold**, *italic*, and [links](url)
* Matches server-side goldmark configuration
*/ */
initializeTurndown() { initializeTurndown() {
this.turndown = new TurndownService({ this.turndown = new TurndownService({
headingStyle: 'atx', // # headers instead of underlines // Minimal configuration - only basic formatting
hr: '---', // horizontal rule style headingStyle: 'atx', // # headers (but will be disabled)
bulletListMarker: '-', // bullet list marker hr: '---', // horizontal rule (but will be disabled)
codeBlockStyle: 'fenced', // ``` code blocks bulletListMarker: '-', // bullet list (but will be disabled)
fence: '```', // fence marker codeBlockStyle: 'fenced', // code blocks (but will be disabled)
emDelimiter: '*', // emphasis delimiter fence: '```', // fence marker (but will be disabled)
strongDelimiter: '**', // strong delimiter emDelimiter: '*', // *italic* - matches server
linkStyle: 'inlined', // [text](url) instead of reference style strongDelimiter: '**', // **bold** - matches server
linkStyle: 'inlined', // [text](url) - matches server
linkReferenceStyle: 'full' // full reference links linkReferenceStyle: 'full' // full reference links
}); });
@@ -48,7 +91,9 @@ export class MarkdownConverter {
} }
/** /**
* Add custom turndown rules for better HTML → Markdown conversion * Add custom turndown rules - MINIMAL MODE
* Only supports: **bold**, *italic*, and [links](url)
* Disables all other formatting to match server
*/ */
addTurndownRules() { addTurndownRules() {
// Handle paragraph spacing properly - ensure double newlines between paragraphs // Handle paragraph spacing properly - ensure double newlines between paragraphs
@@ -60,7 +105,7 @@ export class MarkdownConverter {
} }
}); });
// Handle bold text in markdown // Handle bold text in markdown - keep this (supported)
this.turndown.addRule('bold', { this.turndown.addRule('bold', {
filter: ['strong', 'b'], filter: ['strong', 'b'],
replacement: function (content) { replacement: function (content) {
@@ -69,7 +114,7 @@ export class MarkdownConverter {
} }
}); });
// Handle italic text in markdown // Handle italic text in markdown - keep this (supported)
this.turndown.addRule('italic', { this.turndown.addRule('italic', {
filter: ['em', 'i'], filter: ['em', 'i'],
replacement: function (content) { replacement: function (content) {
@@ -77,6 +122,42 @@ export class MarkdownConverter {
return '*' + content + '*'; return '*' + content + '*';
} }
}); });
// DISABLE unsupported features - convert to plain text
this.turndown.addRule('disableHeadings', {
filter: ['h1', 'h2', 'h3', 'h4', 'h5', 'h6'],
replacement: function (content) {
return content; // Just return text content, no # markup
}
});
this.turndown.addRule('disableLists', {
filter: ['ul', 'ol', 'li'],
replacement: function (content) {
return content; // Just return text content, no list markup
}
});
this.turndown.addRule('disableCode', {
filter: ['pre', 'code'],
replacement: function (content) {
return content; // Just return text content, no code markup
}
});
this.turndown.addRule('disableBlockquotes', {
filter: 'blockquote',
replacement: function (content) {
return content; // Just return text content, no > markup
}
});
this.turndown.addRule('disableHR', {
filter: 'hr',
replacement: function () {
return ''; // Remove horizontal rules entirely
}
});
} }
/** /**