Fix extract_first_email_addr() to really return the first email

The use of find_all() in extract_email_addrs_vec() extracted occurrences
to an intermediate set and thus lost any sense of ordering.

This changes extract_email_addrs_vec() to use find_all_ordered() and
return all occurrences of email addresses found in the argument,
included duplicates, with their order of occurrence preserved.

NEWS

@@ -27,6 +27,11 @@ Changed Functionality
   - The use as enum of type ``NetControl::RuleType`` is unchanged and still
     named ``NetControl::DROP``
 
+- The extract_email_addrs_vec() BIF now returns all occurrences of emails,
+  including duplicates, with preserved order of occurrence.  This seems like
+  the original/documented intent of the function, but the previous
+  implementation did not preserve ordering or duplicates.
+
 Removed Functionality
 ---------------------
 

scripts/base/utils/email.zeek

@@ -8,9 +8,9 @@ function extract_email_addrs_vec(str: string): string_vec
 	{
 	local addrs: vector of string = vector();
 
-	local raw_addrs = find_all(str, /(^|[<,:[:blank:]])[^<,:[:blank:]@]+"@"[^>,;[:blank:]]+([>,;[:blank:]]|$)/);
-	for ( raw_addr in raw_addrs )
-		addrs += gsub(raw_addr, /[<>,:;[:blank:]]/, "");
+	local raw_addrs = find_all_ordered(str, /(^|[<,:[:blank:]])[^<,:[:blank:]@]+"@"[^>,;[:blank:]]+([>,;[:blank:]]|$)/);
+	for ( i in raw_addrs )
+		addrs += gsub(raw_addrs[i], /[<>,:;[:blank:]]/, "");
 
 	return addrs;
 	}

testing/btest/Baseline/scripts.base.utils.email/output

@@ -0,0 +1,21 @@
+one@example.com
+[one@example.com, two@example.com, three@example.com, one@example.com]
+{
+three@example.com,
+two@example.com,
+one@example.com
+}
+one@example.com
+[one@example.com, two@example.com, three@example.com, one@example.com]
+{
+three@example.com,
+two@example.com,
+one@example.com
+}
+one@example.com
+[one@example.com, two@example.com, three@example.com, one@example.com]
+{
+three@example.com,
+one@example.com,
+two@example.com
+}

testing/btest/scripts/base/utils/email.zeek

@@ -0,0 +1,17 @@
+# @TEST-EXEC: zeek -b %INPUT >output
+# @TEST-EXEC: btest-diff output
+
+@load base/utils/email
+
+local s = "one@example.com two@example.com three@example.com one@example.com";
+print extract_first_email_addr(s);
+print extract_email_addrs_vec(s);
+print extract_email_addrs_set(s);
+s = "one@example.com,two@example.com,three@example.com,one@example.com";
+print extract_first_email_addr(s);
+print extract_email_addrs_vec(s);
+print extract_email_addrs_set(s);
+s = "ieje one@example.com, eifj two@example.com, asdf three@example.com, one@example.com";
+print extract_first_email_addr(s);
+print extract_email_addrs_vec(s);
+print extract_email_addrs_set(s);

testing/external/commit-hash.zeek-testing-private

@@ -1 +1 @@
-8cee05d3405d85a4ae8163d1999b78d182b25213
+c6e8c83cf4d74cde39b57f70a3259cac5a833c1a