优化邮件转发正文链接处理

Author: trah01

config.example.json

@@ -97,6 +97,7 @@
         "wh_17700000000"
       ],
       "filter_rule_ids": [],
+      "include_links": false,
       "enabled": true
     }
   ],

globals.go

@@ -26,5 +26,7 @@ var (
 )
 
 const (
-	ConfigFile = "data/config.json"
+	ConfigFile              = "data/config.json"
+	maxForwardURLLength     = 600
+	maxForwardLinkTextRunes = 80
 )

mail.go

@@ -5,6 +5,7 @@ import (
 	"io"
 	"net"
 	"net/url"
+	"regexp"
 	"strings"
 	"time"
 
@@ -16,6 +17,8 @@ import (
 
 // ===================== 邮件处理 =====================
 
+var markdownLinkRegex = regexp.MustCompile(`\[([^\]\n]{0,240})\]\((https?://[^)\s]+|长链接由于超长已被过滤)\)`)
+
 func cleanHTML(htmlStr string) string {
 	doc, err := goquery.NewDocumentFromReader(strings.NewReader(htmlStr))
 	if err != nil {
@@ -35,14 +38,14 @@ func cleanHTML(htmlStr string) string {
 		href = sanitizeURL(href)
 		if text == "" {
 			text = "链接"
-		} else if len(text) > 80 {
-			text = text[:77] + "..."
+		} else {
+			text = truncateRunes(text, maxForwardLinkTextRunes)
 		}
 		text = escapeMarkdownText(text)
 		if href != "" {
 			href = escapeMarkdownLinkURL(href)
-			if len(href) > 600 {
-				s.SetText(fmt.Sprintf("[%s](长链接由于超长已被过滤)", text))
+			if len(href) > maxForwardURLLength {
+				s.SetText(text)
 			} else {
 				s.SetText(fmt.Sprintf("[%s](%s)", text, href))
 			}
@@ -62,8 +65,8 @@ func cleanHTML(htmlStr string) string {
 
 	text := normalizeFormattedText(doc.Text())
 	text = urlRegex.ReplaceAllStringFunc(text, func(u string) string {
-		if len(u) > 600 {
-			return u[:80] + "...(该段长链接由于超长已被过滤)"
+		if len(u) > maxForwardURLLength {
+			return ""
 		}
 		return u
 	})
@@ -146,19 +149,112 @@ func normalizeFormattedText(text string) string {
 
 func formatPlainTextBody(body string) string {
 	body = urlRegex.ReplaceAllStringFunc(body, func(u string) string {
-		disp := u
-		if len(disp) > 80 {
-			disp = disp[:77] + "..."
+		if len(u) > maxForwardURLLength {
+			return ""
 		}
+		disp := truncateRunes(u, maxForwardLinkTextRunes)
 		u = escapeMarkdownLinkURL(u)
-		if len(u) > 600 {
-			return fmt.Sprintf("[%s](长链接由于超长已被过滤)", disp)
-		}
 		return fmt.Sprintf("[%s](%s)", disp, u)
 	})
 	return normalizeFormattedText(body)
 }
 
+func formatForwardBody(body string, includeLinks bool) string {
+	body = normalizeFormattedText(body)
+	if includeLinks {
+		return normalizeFormattedText(formatURLsAsMarkdown(body))
+	}
+	return normalizeFormattedText(removeLinksFromText(body))
+}
+
+func formatURLsAsMarkdown(text string) string {
+	var b strings.Builder
+	last := 0
+	matches := urlRegex.FindAllStringIndex(text, -1)
+	for _, loc := range matches {
+		start, end := loc[0], loc[1]
+		if isMarkdownURLDestination(text, start, end) {
+			continue
+		}
+		u := text[start:end]
+		b.WriteString(text[last:start])
+		if len(u) <= maxForwardURLLength {
+			disp := truncateRunes(u, maxForwardLinkTextRunes)
+			b.WriteString(fmt.Sprintf("[%s](%s)", disp, escapeMarkdownLinkURL(u)))
+		}
+		last = end
+	}
+	if last == 0 {
+		return text
+	}
+	b.WriteString(text[last:])
+	return b.String()
+}
+
+func removeLinksFromText(text string) string {
+	text = markdownLinkRegex.ReplaceAllStringFunc(text, func(match string) string {
+		parts := markdownLinkRegex.FindStringSubmatch(match)
+		if len(parts) < 2 {
+			return ""
+		}
+		return unescapeMarkdownText(parts[1])
+	})
+	text = urlRegex.ReplaceAllString(text, "")
+
+	lines := strings.Split(text, "\n")
+	kept := make([]string, 0, len(lines))
+	for _, line := range lines {
+		if isDanglingLinkLabel(line) {
+			continue
+		}
+		kept = append(kept, line)
+	}
+	return strings.Join(kept, "\n")
+}
+
+func isMarkdownURLDestination(text string, start int, end int) bool {
+	if start <= 0 || text[start-1] != '(' {
+		return false
+	}
+	return (end < len(text) && text[end] == ')') || (end > start && text[end-1] == ')')
+}
+
+func isDanglingLinkLabel(line string) bool {
+	line = strings.TrimSpace(line)
+	line = strings.TrimRight(line, "：:")
+	switch strings.ToLower(line) {
+	case "链接", "url", "link":
+		return true
+	default:
+		return false
+	}
+}
+
+func truncateRunes(text string, max int) string {
+	runes := []rune(text)
+	if len(runes) <= max {
+		return text
+	}
+	if max <= 3 {
+		return string(runes[:max])
+	}
+	return string(runes[:max-3]) + "..."
+}
+
+func unescapeMarkdownText(text string) string {
+	replacer := strings.NewReplacer(
+		`\\[`, `[`,
+		`\\]`, `]`,
+		`\\(`, `(`,
+		`\\)`, `)`,
+		"\\`", "`",
+		`\\*`, `*`,
+		`\\_`, `_`,
+		`\\\\`, `\`,
+	)
+	return replacer.Replace(text)
+}
+
 func normalizeFolderKey(folder string) string {
 	folder = strings.TrimSpace(folder)
 	if folder == "" {

main_test.go

@@ -25,16 +25,16 @@ func TestCleanHTML_PreservesReadableStructure(t *testing.T) {
 	}
 }
 
-func TestFormatPlainTextBody_ConvertsURLToMarkdownAndTruncatesLongLinks(t *testing.T) {
+func TestFormatPlainTextBody_ConvertsURLToMarkdownAndRemovesLongLinks(t *testing.T) {
 	body := "请查看 https://example.com/docs?id=1 和这个超长链接 https://example.com/" + strings.Repeat("a", 650)
 
 	got := formatPlainTextBody(body)
 
 	if !strings.Contains(got, "[https://example.com/docs?id=1](https://example.com/docs?id=1)") {
 		t.Fatalf("expected normal URL converted to markdown link, got: %q", got)
 	}
-	if !strings.Contains(got, "长链接由于超长已被过滤") {
-		t.Fatalf("expected overlong URL filtered message, got: %q", got)
+	if strings.Contains(got, strings.Repeat("a", 80)) || strings.Contains(got, "长链接由于超长已被过滤") {
+		t.Fatalf("expected overlong URL removed without placeholder, got: %q", got)
 	}
 }
 
@@ -74,6 +74,32 @@ func TestFormatPlainTextBody_FirstLineUsesTwoSpaceIndent(t *testing.T) {
 	}
 }
 
+func TestFormatForwardBody_RemovesLinksByDefault(t *testing.T) {
+	body := "查看 [详情](https://example.com/docs?id=1)\n链接：https://example.com/raw\n第二段"
+
+	got := formatForwardBody(body, false)
+
+	if strings.Contains(got, "https://") || strings.Contains(got, "[详情]") {
+		t.Fatalf("expected links removed by default, got: %q", got)
+	}
+	if !strings.Contains(got, "查看 详情") || !strings.Contains(got, "第二段") {
+		t.Fatalf("expected readable text preserved, got: %q", got)
+	}
+}
+
+func TestFormatForwardBody_KeepsShortLinksAndRemovesLongLinksWhenEnabled(t *testing.T) {
+	body := "查看 https://example.com/docs?id=1\n长链接：https://example.com/" + strings.Repeat("b", 650)
+
+	got := formatForwardBody(body, true)
+
+	if !strings.Contains(got, "[https://example.com/docs?id=1](https://example.com/docs?id=1)") {
+		t.Fatalf("expected short URL converted to markdown link, got: %q", got)
+	}
+	if strings.Contains(got, strings.Repeat("b", 80)) {
+		t.Fatalf("expected overlong URL removed, got: %q", got)
+	}
+}
+
 func TestApplyFilterRules_BlacklistBlocksSender(t *testing.T) {
 	msg := &Message{From: "newsletter@example.com", Subject: "周报", Body: "正文"}
 	ctx := filterContext{DisplaySender: msg.From}

models.go

@@ -54,6 +54,7 @@ type ForwardRule struct {
 	TargetWebhook  string   `json:"target_webhook"`  // 兼容旧配置的单 Webhook 目标ID
 	TargetWebhooks []string `json:"target_webhooks"` // Webhook 目标ID列表
 	FilterRuleIDs  []string `json:"filter_rule_ids"` // 独立过滤规则ID
+	IncludeLinks   bool     `json:"include_links"`   // 是否在转发正文中保留邮件链接，默认不保留
 	Enabled        bool     `json:"enabled"`
 }
 

processor.go

@@ -84,7 +84,7 @@ func processPendingMessages() {
 			senderForSend := filterCtx.DisplaySender
 
 			// 智能提取验证码并高亮前置
-			displayBody := msg.Body
+			displayBody := formatForwardBody(msg.Body, rule.IncludeLinks)
 			var verificationCode string
 			if matches := codeRegex.FindStringSubmatch(msg.Subject); len(matches) > 1 {
 				verificationCode = matches[1]
@@ -93,7 +93,7 @@ func processPendingMessages() {
 			}
 
 			if verificationCode != "" {
-				displayBody = fmt.Sprintf("**[智能提取验证码] %s**\n\n%s", verificationCode, msg.Body)
+				displayBody = fmt.Sprintf("**[智能提取验证码] %s**\n\n%s", verificationCode, displayBody)
 			}
 
 			sentTargets := make([]string, 0, len(targetIDs))

static/rules.js

@@ -63,6 +63,7 @@ function renderRules() {
                 <td>${targetNames.length ? targetNames.map(name => `<span class="tag tag-info">${escapeHtml(name)}</span>`).join(' ') : '未知'}</td>
                 <td>
                     ${selectedFilters.length ? selectedFilters.map(f => `<span class="tag tag-info">${escapeHtml(f.name)}</span>`).join(' ') : '无'}
+                    ${rule.include_links ? '<span class="tag tag-neutral">保留链接</span>' : ''}
                 </td>
                 <td>
                     <span class="tag ${rule.enabled ? 'tag-success' : 'tag-neutral'}">
@@ -140,6 +141,7 @@ async function openRuleModal(data = null) {
     renderRuleSourceDropdown(normalizeRuleSources(data || { source_account: 'all' }));
     renderRuleTargetDropdown(normalizeRuleTargets(data || {}));
     renderRuleFilterDropdown(data?.filter_rule_ids || []);
+    document.getElementById('rule-include-links').checked = data?.include_links === true;
     document.getElementById('rule-enabled').checked = data?.enabled !== false;
     document.getElementById('rule-modal').classList.add('active');
 }
@@ -341,6 +343,7 @@ async function saveRule() {
         target_webhook: targetIDs[0] || '',
         target_webhooks: targetIDs,
         filter_rule_ids: getSelectedRuleFilterIDs(),
+        include_links: document.getElementById('rule-include-links').checked,
         enabled: document.getElementById('rule-enabled').checked
     };
 

templates/index.html

@@ -805,6 +805,13 @@ <h3 class="modal-title" id="rule-modal-title">添加转发规则</h3>
                             <div class="filter-dropdown-menu" id="rule-filter-menu"></div>
                         </div>
                     </div>
+                    <div class="form-group">
+                        <label class="form-checkbox">
+                            <input type="checkbox" id="rule-include-links">
+                            <span>保留邮件正文中的链接</span>
+                        </label>
+                        <div class="form-hint">默认关闭。关闭时转发正文会移除链接；开启时短链接会转换为 Markdown 超链接，超长链接会删除。</div>
+                    </div>
                     <div class="form-group">
                         <label class="form-checkbox">
                             <input type="checkbox" id="rule-enabled" checked>