From 65e8686b0e9e909b6ea5629f95a9b14e81927872 Mon Sep 17 00:00:00 2001
From: Aleksandr Nogikh <nogikh@google.com>
Date: Fri, 25 Oct 2024 14:29:36 +0200
Subject: dashboard: adapt to the new Googe Groups footer

It changed recently and it broke our email discussion link extraction.

Closes #5431.
---
 pkg/email/parser.go | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

(limited to 'pkg/email/parser.go')

diff --git a/pkg/email/parser.go b/pkg/email/parser.go
index e8aa1a8c0..b748d02c3 100644
--- a/pkg/email/parser.go
+++ b/pkg/email/parser.go
@@ -11,6 +11,7 @@ import (
 	"mime/multipart"
 	"mime/quotedprintable"
 	"net/mail"
+	"net/url"
 	"regexp"
 	"sort"
 	"strings"
@@ -59,8 +60,8 @@ const (
 	cmdTest5
 )
 
-var groupsLinkRe = regexp.MustCompile("\nTo view this discussion on the web visit" +
-	" (https://groups\\.google\\.com/.*?)\\.(?:\r)?\n")
+var groupsLinkRe = regexp.MustCompile(`(?m)\nTo view this discussion (?:on the web )?visit` +
+	` (https://groups\.google\.com/.*?)\.(:?$|\n|\r)`)
 
 func prepareEmails(list []string) map[string]bool {
 	ret := make(map[string]bool)
@@ -153,6 +154,9 @@ func Parse(r io.Reader, ownEmails, goodLists, domains []string) (*Email, error)
 	link := ""
 	if match := groupsLinkRe.FindStringSubmatchIndex(bodyStr); match != nil {
 		link = bodyStr[match[2]:match[3]]
+		if unescaped, err := url.QueryUnescape(link); err == nil {
+			link = unescaped
+		}
 	}
 
 	author := CanonicalEmail(from[0].Address)
-- 
cgit mrf-deployment