Fix ValueError when parsing TXT records with escaped semicolons (\;)

zonefile_parser/helper.py

@@ -3,6 +3,8 @@
 def remove_comments(line:str):
     for index,character in enumerate(line):
         if character == ";" and not is_in_quote(line,index):
+            if index > 0 and line[index - 1] == '\\':
+                continue
             line = line[:index]
             break
     return line

zonefile_parser/helper_test.py

@@ -16,6 +16,16 @@ def test_doesnt_change_string_without_comment(self):
         result = helper.remove_comments(input)
         assert result == input
 
+    def test_doesnt_remove_escaped_semicolon(self):
+        input = r"v=DMARC1\;"
+        result = helper.remove_comments(input)
+        assert result == input
+
+    def test_doesnt_treat_text_after_escaped_semicolon_as_comment(self):
+        input = r"v=DMARC1\; comment"
+        result = helper.remove_comments(input)
+        assert result == input
+
 class TestIsInQuote:
     def test_returns_whether_index_in_quote(self):
         input = '"A"'

zonefile_parser/main_test.py

@@ -328,6 +328,37 @@ def test_origin_not_appended_to_txt_rdata_when_name_matches(self):
         assert record.rtype == "TXT"
         assert record.rdata == {"value": "v=spf1 include:mail.otherdomain.com ~all"}
 
+    def test_issue_49_escaped_semicolon_in_txt(self):
+        # TXT record with a trailing escaped semicolon (\;) should not raise ValueError
+        text = """
+$TTL 3600
+$ORIGIN example.com.
+_dmarc 3600 IN TXT v=DMARC1\\;
+"""
+        result = zonefile_parser.main.parse(text)
+
+        record = result[0]
+
+        assert record.name == "_dmarc.example.com."
+        assert record.rtype == "TXT"
+        assert record.rdata == {"value": "v=DMARC1;"}
+
+    def test_issue_49_dmarc_record_with_comment(self):
+        # DMARC TXT record with an escaped semicolon followed by a real comment
+        # the real comment (unescaped ;) should be stripped, \; should survive as ;
+        text = """
+$TTL 3600
+$ORIGIN example.com.
+_dmarc 3600 IN TXT v=DMARC1\\; ; this is a comment
+"""
+        result = zonefile_parser.main.parse(text)
+
+        record = result[0]
+
+        assert record.name == "_dmarc.example.com."
+        assert record.rtype == "TXT"
+        assert record.rdata == {"value": "v=DMARC1;"}
+
     def test_multiple_cnames_with_name_in_target(self):
         # multiple records in the same zone, each with name appearing in their CNAME target
         text = """