From a411306fba0dc87778e64392b57c844e0477df2b Mon Sep 17 00:00:00 2001
From: faye-sama <32477052+faye-sama@users.noreply.github.com>
Date: Mon, 13 Nov 2017 23:06:47 +0000
Subject: [PATCH] Fail metrics parsing on unescaped quotes (#3409)

Before this change Fields() method on a metric parsed from a line with
unescaped quotes could panic. This change makes such line unparseable.

Fixes #3326
---
 metric/parse.go                       | 20 +++++++++++++-------
 plugins/parsers/influx/parser_test.go | 14 ++++++++++++++
 2 files changed, 27 insertions(+), 7 deletions(-)

diff --git a/metric/parse.go b/metric/parse.go
index 1acf30094..80fe2ef3f 100644
--- a/metric/parse.go
+++ b/metric/parse.go
@@ -326,7 +326,9 @@ func scanTagsValue(buf []byte, i int) (int, int, error) {
 func scanFields(buf []byte, i int) (int, []byte, error) {
 	start := skipWhitespace(buf, i)
 	i = start
-	quoted := false
+
+	// track how many '"" we've seen since last '='
+	quotes := 0
 
 	// tracks how many '=' we've seen
 	equals := 0
@@ -350,13 +352,17 @@ func scanFields(buf []byte, i int) (int, []byte, error) {
 		// Only quote values in the field value since quotes are not significant
 		// in the field key
 		if buf[i] == '"' && equals > commas {
-			quoted = !quoted
 			i++
+			quotes++
+			if quotes > 2 {
+				break
+			}
 			continue
 		}
 
 		// If we see an =, ensure that there is at least on char before and after it
-		if buf[i] == '=' && !quoted {
+		if buf[i] == '=' && quotes != 1 {
+			quotes = 0
 			equals++
 
 			// check for "... =123" but allow "a\ =123"
@@ -398,19 +404,19 @@ func scanFields(buf []byte, i int) (int, []byte, error) {
 			}
 		}
 
-		if buf[i] == ',' && !quoted {
+		if buf[i] == ',' && quotes != 1 {
 			commas++
 		}
 
 		// reached end of block?
-		if buf[i] == ' ' && !quoted {
+		if buf[i] == ' ' && quotes != 1 {
 			break
 		}
 		i++
 	}
 
-	if quoted {
-		return i, buf[start:i], makeError("unbalanced quotes", buf, i)
+	if quotes != 0 && quotes != 2 {
+		return i, buf[start:i], makeError("unescaped/ublanaced quotes", buf, i)
 	}
 
 	// check that all field sections had key and values (e.g. prevent "a=1,b"
diff --git a/plugins/parsers/influx/parser_test.go b/plugins/parsers/influx/parser_test.go
index 6b2ba8d56..67833fc56 100644
--- a/plugins/parsers/influx/parser_test.go
+++ b/plugins/parsers/influx/parser_test.go
@@ -1,6 +1,7 @@
 package influx
 
 import (
+	"fmt"
 	"io/ioutil"
 	"testing"
 	"time"
@@ -24,6 +25,8 @@ const (
 	validInfluxNoNewline = "cpu_load_short,cpu=cpu0 value=10 1257894000000000000"
 	invalidInflux        = "I don't think this is line protocol\n"
 	invalidInflux2       = "{\"a\": 5, \"b\": {\"c\": 6}}\n"
+	invalidInflux3       = `name text="unescaped "quote" ",value=1 1498077493081000000`
+	invalidInflux4       = `name text="unbalanced "quote" 1498077493081000000`
 )
 
 const influxMulti = `
@@ -221,10 +224,21 @@ func TestParseInvalidInflux(t *testing.T) {
 	assert.Error(t, err)
 	_, err = parser.Parse([]byte(invalidInflux2))
 	assert.Error(t, err)
+	_, err = parser.Parse([]byte(invalidInflux3))
+	assert.Error(t, err)
+	fmt.Printf("%+v\n", err) // output for debug
+	_, err = parser.Parse([]byte(invalidInflux4))
+	assert.Error(t, err)
+
 	_, err = parser.ParseLine(invalidInflux)
 	assert.Error(t, err)
 	_, err = parser.ParseLine(invalidInflux2)
 	assert.Error(t, err)
+	_, err = parser.ParseLine(invalidInflux3)
+	assert.Error(t, err)
+	_, err = parser.ParseLine(invalidInflux4)
+	assert.Error(t, err)
+
 }
 
 func BenchmarkParse(b *testing.B) {