Receive all logs if no attributes are specified

.chloggen/receive-all-cloudflare.yaml

@@ -0,0 +1,27 @@
+# Use this changelog template to create an entry for release notes.
+
+# One of 'breaking', 'deprecation', 'new_component', 'enhancement', 'bug_fix'
+change_type: enhancement
+
+# The name of the component, or a single word describing the area of concern, (e.g. filelogreceiver)
+component: cloudflarereceiver
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Ingest all attributes by default when attributes map is empty
+
+# Mandatory: One or more tracking issues related to the change. You can use the PR number here if no issue exists.
+issues: [37720]
+
+# (Optional) One or more lines of additional information to render under the primary note.
+# These lines will be padded with 2 spaces and then inserted directly into the document.
+# Use pipe (|) for multiline entries.
+subtext:
+
+# If your change doesn't affect end users or the exported elements of any package,
+# you should instead start your pull request title with [chore] or use the "Skip Changelog" label.
+# Optional: The change log or logs in which this entry should be included.
+# e.g. '[user]' or '[user, api]'
+# Include 'user' if the change is relevant to end users.
+# Include 'api' if there is a change to a library API.
+# Default: '[user]'
+change_logs: []

receiver/cloudflarereceiver/README.md

@@ -48,6 +48,7 @@ If the receiver will be handling TLS termination:
   - This receiver was built with the Cloudflare `http_requests` dataset in mind, but should be able to support any Cloudflare dataset. If using another dataset, you will need to set the `timestamp_field` appropriately in order to have the log record be associated with the correct timestamp. the timestamp must be formatted RFC3339, as stated in the Getting Started section.
 - `attributes`
   - This parameter allows the receiver to be configured to set log record attributes based on fields found in the log message. The fields are not removed from the log message when set in this way. Only string, boolean, integer or float fields can be mapped using this parameter.
+  - When the `attributes` configuration is empty, the receiver will automatically ingest all fields from the log messages as attributes, using the original field names as attribute names.
 
 
 ### Example:
@@ -66,3 +67,18 @@ receivers:
         ClientIP: http_request.client_ip
         ClientRequestURI: http_request.uri
 ```
+
+### Example with automatic attribute ingestion:
+```yaml
+receivers:
+  cloudflare:
+    logs:
+      tls:
+        key_file: some_key_file
+        cert_file: some_cert_file
+      endpoint: 0.0.0.0:12345
+      secret: 1234567890abcdef1234567890abcdef
+      timestamp_field: EdgeStartTimestamp
+      attributes:
+        # Specifying no attributes ingests them all
+```

receiver/cloudflarereceiver/logs.go

@@ -272,23 +272,36 @@ func (l *logsReceiver) processLogs(now pcommon.Timestamp, logs []map[string]any)
 			}
 
 			attrs := logRecord.Attributes()
-			for field, attribute := range l.cfg.Attributes {
-				if v, ok := log[field]; ok {
-					switch v := v.(type) {
-					case string:
-						attrs.PutStr(attribute, v)
-					case int:
-						attrs.PutInt(attribute, int64(v))
-					case int64:
-						attrs.PutInt(attribute, v)
-					case float64:
-						attrs.PutDouble(attribute, v)
-					case bool:
-						attrs.PutBool(attribute, v)
-					default:
-						l.logger.Warn("unable to translate field to attribute, unsupported type", zap.String("field", field), zap.Any("value", v), zap.String("type", fmt.Sprintf("%T", v)))
+			for field, v := range log {
+				attrName := field
+				if len(l.cfg.Attributes) != 0 {
+					// Only process fields that are in the config mapping
+					if mappedAttr, ok := l.cfg.Attributes[field]; ok {
+						attrName = mappedAttr
+					} else {
+						// Skip fields not in mapping when we have a config
+						continue
 					}
 				}
+				// else if l.cfg.Attributes is empty, default to processing all fields with no renaming
+
+				switch v := v.(type) {
+				case string:
+					attrs.PutStr(attrName, v)
+				case int:
+					attrs.PutInt(attrName, int64(v))
+				case int64:
+					attrs.PutInt(attrName, v)
+				case float64:
+					attrs.PutDouble(attrName, v)
+				case bool:
+					attrs.PutBool(attrName, v)
+				default:
+					l.logger.Warn("unable to translate field to attribute, unsupported type", 
+						zap.String("field", field), 
+						zap.Any("value", v), 
+						zap.String("type", fmt.Sprintf("%T", v)))
+				}
 			}
 
 			err := logRecord.Body().SetEmptyMap().FromRaw(log)

receiver/cloudflarereceiver/logs_test.go

@@ -353,6 +353,90 @@ func TestHandleRequest(t *testing.T) {
 	}
 }
 
+func TestEmptyAttributes(t *testing.T) {
+	now := time.Time{}
+
+	testCases := []struct {
+		name         string
+		payload      string
+		attributes	map[string]string
+	}{
+		{
+			name: "Empty Attributes Map",
+			payload: `{ "ClientIP": "89.163.253.200", "ClientRequestHost": "www.theburritobot0.com", "ClientRequestMethod": "GET", "ClientRequestURI": "/static/img/testimonial-hipster.png", "EdgeEndTimestamp": "2023-03-03T05:30:05Z", "EdgeResponseBytes": "69045", "EdgeResponseStatus": "200", "EdgeStartTimestamp": "2023-03-03T05:29:05Z", "RayID": "3a6050bcbe121a87" }
+{ "ClientIP" : "89.163.253.201", "ClientRequestHost": "www.theburritobot1.com", "ClientRequestMethod": "GET", "ClientRequestURI": "/static/img/testimonial-hipster.png", "EdgeEndTimestamp": "2023-03-03T05:30:05Z", "EdgeResponseBytes": "69045", "EdgeResponseStatus": "200", "EdgeStartTimestamp": "2023-03-03T05:29:05Z", "RayID": "3a6050bcbe121a87" }`,
+			attributes: map[string]string{},
+		},
+		{
+			name: "nil Attributes",
+			payload: `{ "ClientIP": "89.163.253.200", "ClientRequestHost": "www.theburritobot0.com", "ClientRequestMethod": "GET", "ClientRequestURI": "/static/img/testimonial-hipster.png", "EdgeEndTimestamp": "2023-03-03T05:30:05Z", "EdgeResponseBytes": "69045", "EdgeResponseStatus": "200", "EdgeStartTimestamp": "2023-03-03T05:29:05Z", "RayID": "3a6050bcbe121a87" }
+{ "ClientIP" : "89.163.253.201", "ClientRequestHost": "www.theburritobot1.com", "ClientRequestMethod": "GET", "ClientRequestURI": "/static/img/testimonial-hipster.png", "EdgeEndTimestamp": "2023-03-03T05:30:05Z", "EdgeResponseBytes": "69045", "EdgeResponseStatus": "200", "EdgeStartTimestamp": "2023-03-03T05:29:05Z", "RayID": "3a6050bcbe121a87" }`,
+			attributes: nil,
+		},
+	}
+
+	expectedLogs := func(t *testing.T, payload string) plog.Logs {
+		logs := plog.NewLogs()
+		rl := logs.ResourceLogs().AppendEmpty()
+		sl := rl.ScopeLogs().AppendEmpty()
+		sl.Scope().SetName("github.com/open-telemetry/opentelemetry-collector-contrib/receiver/cloudflarereceiver")
+
+		for idx, line := range strings.Split(payload, "\n") {
+			lr := sl.LogRecords().AppendEmpty()
+
+			require.NoError(t, lr.Attributes().FromRaw(map[string]any{
+				"ClientIP": fmt.Sprintf("89.163.253.%d", 200+idx),
+				"ClientRequestHost": fmt.Sprintf("www.theburritobot%d.com", idx),
+				"ClientRequestMethod": "GET",
+				"ClientRequestURI": "/static/img/testimonial-hipster.png",
+				"EdgeEndTimestamp": "2023-03-03T05:30:05Z",
+				"EdgeResponseBytes": "69045",
+				"EdgeResponseStatus": "200",
+				"EdgeStartTimestamp": "2023-03-03T05:29:05Z",
+				"RayID": "3a6050bcbe121a87",
+			}))
+
+			lr.SetObservedTimestamp(pcommon.NewTimestampFromTime(now))
+			ts, err := time.Parse(time.RFC3339, "2023-03-03T05:29:05Z")
+			require.NoError(t, err)
+			lr.SetTimestamp(pcommon.NewTimestampFromTime(ts))
+			lr.SetSeverityNumber(plog.SeverityNumberInfo)
+			lr.SetSeverityText(plog.SeverityNumberInfo.String())
+
+			var log map[string]any
+			err = json.Unmarshal([]byte(line), &log)
+			require.NoError(t, err)
+
+			payloadToExpectedBody(t, line, lr)
+		}
+
+		return logs
+	}
+
+	for _, tc := range testCases {
+		t.Run(tc.name, func(t *testing.T) {
+			recv := newReceiver(t, &Config{
+				Logs: LogsConfig{
+					Endpoint:       "localhost:0",
+					TLS:            &configtls.ServerConfig{},
+					TimestampField: "EdgeStartTimestamp",
+					Attributes: tc.attributes,
+				},
+			},
+				&consumertest.LogsSink{},
+			)
+			var logs plog.Logs
+			rawLogs, err := parsePayload([]byte(tc.payload))
+			if err == nil {
+				logs = recv.processLogs(pcommon.NewTimestampFromTime(time.Now()), rawLogs)
+			}
+			require.NoError(t, err)
+			require.NotNil(t, logs)
+			require.NoError(t, plogtest.CompareLogs(expectedLogs(t, tc.payload), logs, plogtest.IgnoreObservedTimestamp()))
+		})
+	}
+}
+
 func gzippedMessage(message string) string {
 	var b bytes.Buffer
 	w := gzip.NewWriter(&b)