Remove ANSI codes from hostnames in proxy logs

- Added a function to strip ANSI escape codes from hostname fields in parsed proxy logs, ensuring cleaner data.
- Updated tests to verify that ANSI codes are correctly removed from hostnames.

logline/proxy_parser.go

@@ -29,6 +29,9 @@ func (parser ProxyParser) Parse(logLine string) (HttpRequest, error) {
 	}
 	//fmt.Printf("%+v\n", result)
 
+	// Data quality fixes
+	result["hostname"] = removeANSICodes(result["hostname"])
+
 	httpRequest := HttpRequest{}
 	httpRequest.Host = result["hostname"]
 	httpRequest.SourceIp = result["remote_addr"]
@@ -48,3 +51,8 @@ func (parser ProxyParser) Parse(logLine string) (HttpRequest, error) {
 
 	return httpRequest, nil
 }
+
+func removeANSICodes(input string) string {
+	ansiEscapeRe := regexp.MustCompile(`\[[0-9;]*m`)
+	return ansiEscapeRe.ReplaceAllString(input, "")
+}

logline/proxy_parser_test.go

@@ -114,3 +114,13 @@ func TestProxyLogLogParserImplementation(t *testing.T) {
 	}
 	fmt.Printf("%+v\n", result)
 }
+
+func TestProxyLogRemovesANSICodesFromHostname(t *testing.T) {
+	line := `[0mprom-grafana.tyranus.de 127.0.0.1 - - [29/Aug/2023:07:47:30 +0000] "GET / HTTP/1.1" 200 12345 "-" "Mozilla/5.0" "-"`
+
+	httpRequest, err := logParser.Parse(line)
+	failNowIfErr(t, err, line)
+
+	expectedHost := "prom-grafana.tyranus.de"
+	assert.Equal(t, expectedHost, httpRequest.Host, "The hostname should have ANSI codes removed")
+}