Merge pull request #2857 from influxdb/jw-commas

Fix parsing commas in string field values
influxdata · Jun 9, 2015 · cd3e758 · cd3e758
2 parents 22e6cbc + 9f444a6
commit cd3e758
Show file tree

Hide file tree

Showing 3 changed files with 62 additions and 1 deletion.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -9,6 +9,7 @@
 - [2829](https://github.com/influxdb/influxdb/pull/2829) -- Re-enable Graphite support as a new Service-style component.
 - [2814](https://github.com/influxdb/influxdb/issues/2814) -- Convert collectd to a service.
 - [2852](https://github.com/influxdb/influxdb/pull/2852) -- Don't panic when altering retention policies. Thanks for the report @huhongbo
+- [2857](https://github.com/influxdb/influxdb/issues/2857) -- Fix parsing commas in string field values.
 
 ## v0.9.0-rc32 [2015-06-07]
 

diff --git a/tsdb/points.go b/tsdb/points.go
@@ -436,6 +436,33 @@ func scanTagValue(buf []byte, i int) (int, []byte) {
 	return i, buf[start:i]
 }
 
+func scanFieldValue(buf []byte, i int) (int, []byte) {
+	start := i
+	quoted := false
+	for {
+		if i >= len(buf) {
+			break
+		}
+
+		if buf[i] == '"' {
+			i += 1
+			quoted = !quoted
+			continue
+		}
+
+		if buf[i] == '\\' {
+			i += 2
+			continue
+		}
+
+		if buf[i] == ',' && !quoted {
+			break
+		}
+		i += 1
+	}
+	return i, buf[start:i]
+}
+
 func escape(in []byte) []byte {
 	for b, esc := range escapeCodes {
 		in = bytes.Replace(in, []byte{b}, esc, -1)
@@ -700,7 +727,7 @@ func newFieldsFromBinary(buf []byte) Fields {
 			continue
 		}
 
-		i, valueBuf = scanTo(buf, i+1, ',')
+		i, valueBuf = scanFieldValue(buf, i+1)
 		if len(valueBuf) == 0 {
 			fields[string(name)] = nil
 			continue

diff --git a/tsdb/points_test.go b/tsdb/points_test.go
@@ -334,6 +334,39 @@ func TestParsePointWithStringWithSpaces(t *testing.T) {
 	)
 }
 
+func TestParsePointWithStringWithCommas(t *testing.T) {
+	// escaped comma
+	test(t, `cpu,host=serverA,region=us-east value=1.0,str="foo\,bar" 1000000000`,
+		NewPoint(
+			"cpu",
+			Tags{
+				"host":   "serverA",
+				"region": "us-east",
+			},
+			Fields{
+				"value": 1.0,
+				"str":   "foo,bar", // commas in string value
+			},
+			time.Unix(1, 0)),
+	)
+
+	// non-escaped comma
+	test(t, `cpu,host=serverA,region=us-east value=1.0,str="foo,bar" 1000000000`,
+		NewPoint(
+			"cpu",
+			Tags{
+				"host":   "serverA",
+				"region": "us-east",
+			},
+			Fields{
+				"value": 1.0,
+				"str":   "foo,bar", // commas in string value
+			},
+			time.Unix(1, 0)),
+	)
+
+}
+
 func TestParsePointWithStringWithEquals(t *testing.T) {
 	test(t, `cpu,host=serverA,region=us-east str="foo=bar",value=1.0, 1000000000`,
 		NewPoint(