[receiver/sqlquery] add support for logs

.chloggen/sqlquery-receiver-add-logs.yaml

@@ -0,0 +1,16 @@
+# 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: sqlqueryreceiver
+
+# A brief description of the change.  Surround your text with quotes ("") if it needs to start with a backtick (`).
+note: Add support for logs
+
+# One or more tracking issues related to the change
+issues: [20284]
+
+# (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:

receiver/sqlqueryreceiver/README.md

@@ -1,7 +1,7 @@
 # SQL Query Receiver (Alpha)
 
 | Status                   |           |
-|--------------------------|-----------|
+| ------------------------ | --------- |
 | Stability                | [alpha]   |
 | Supported pipeline types | metrics   |
 | Distributions            | [contrib] |
@@ -21,29 +21,58 @@ The configuration supports the following top-level fields:
   a driver-specific string usually consisting of at least a database name and connection information. This is sometimes
   referred to as the "connection string" in driver documentation.
   e.g. _host=localhost port=5432 user=me password=s3cr3t sslmode=disable_
-- `queries`(required): A list of queries, where a query is a sql statement and one or more metrics (details below).
+- `queries`(required): A list of queries, where a query is a sql statement and one or more `logs` and/or `metrics` sections (details below).
 - `collection_interval`(optional): The time interval between query executions. Defaults to _10s_.
 
 ### Queries
 
-A _query_ consists of a sql statement and one or more _metrics_, where each metric consists of a
+A _query_ consists of a sql statement and one or more `logs` and/or `metrics` section.
+At least one `logs` or one `metrics` section is required.
+Note that technically you can put both `logs` and `metrics` sections in a single query section,
+but it's probably not a real world use case, as the requirements for logs and metrics queries
+are quite different.
+
+Example:
+
+```yaml
+receivers:
+  sqlquery:
+    driver: postgres
+    datasource: "host=localhost port=5432 user=postgres password=s3cr3t sslmode=disable"
+    queries:
+      - sql: "select * from my_logs"
+        logs:
+          - {}
+      - sql: "select count(*) as count, genre from movie group by genre"
+        metrics:
+          - metric_name: movie.genres
+            value_column: "count"
+```
+
+#### Logs Queries
+
+The `logs` section currently has no options. This section is in development.
+
+#### Metrics queries
+
+Each `metrics` section consists of a
 `metric_name`, a `value_column`, and additional optional fields.
 Each _metric_ in the configuration will produce one OTel metric per row returned from its sql query.
 
-* `metric_name`(required): the name assigned to the OTel metric.
-* `value_column`(required): the column name in the returned dataset used to set the value of the metric's datapoint.
+- `metric_name`(required): the name assigned to the OTel metric.
+- `value_column`(required): the column name in the returned dataset used to set the value of the metric's datapoint.
   This may be case-sensitive, depending on the driver (e.g. Oracle DB).
-* `attribute_columns`(optional): a list of column names in the returned dataset used to set attibutes on the datapoint.
+- `attribute_columns`(optional): a list of column names in the returned dataset used to set attibutes on the datapoint.
   These attributes may be case-sensitive, depending on the driver (e.g. Oracle DB).
-* `data_type` (optional): can be `gauge` or `sum`; defaults to `gauge`.
-* `value_type` (optional): can be `int` or `double`; defaults to `int`.
-* `monotonic` (optional): boolean; whether a cumulative sum's value is monotonically increasing (i.e. never rolls over
+- `data_type` (optional): can be `gauge` or `sum`; defaults to `gauge`.
+- `value_type` (optional): can be `int` or `double`; defaults to `int`.
+- `monotonic` (optional): boolean; whether a cumulative sum's value is monotonically increasing (i.e. never rolls over
   or resets); defaults to false.
-* `aggregation` (optional): only applicable for `data_type=sum`; can be `cumulative` or `delta`; defaults
+- `aggregation` (optional): only applicable for `data_type=sum`; can be `cumulative` or `delta`; defaults
   to `cumulative`.
-* `description` (optional): the description applied to the metric.
-* `unit` (optional): the units applied to the metric.
-* `static_attributes` (optional): static attributes applied to the metrics
+- `description` (optional): the description applied to the metric.
+- `unit` (optional): the units applied to the metric.
+- `static_attributes` (optional): static attributes applied to the metrics
 
 ### Example
 
@@ -53,27 +82,29 @@ receivers:
     driver: postgres
     datasource: "host=localhost port=5432 user=postgres password=s3cr3t sslmode=disable"
     queries:
+      - sql: "select * from my_logs"
+        logs: {}
       - sql: "select count(*) as count, genre from movie group by genre"
         metrics:
           - metric_name: movie.genres
             value_column: "count"
-            attribute_columns: [ "genre" ]
+            attribute_columns: ["genre"]
             static_attributes:
               dbinstance: mydbinstance
 ```
 
 Given a `movie` table with three rows:
 
 | name      | genre  |
-|-----------|--------|
+| --------- | ------ |
 | E.T.      | sci-fi |
 | Star Wars | sci-fi |
 | Die Hard  | action |
 
 If there are two rows returned from the query `select count(*) as count, genre from movie group by genre`:
 
 | count | genre  |
-|-------|--------|
+| ----- | ------ |
 | 2     | sci-fi |
 | 1     | action |
 
@@ -87,7 +118,7 @@ Descriptor:
 NumberDataPoints #0
 Data point attributes:
      -> genre: STRING(sci-fi)
-     -> dbinstance: STRING(mydbinstance)     
+     -> dbinstance: STRING(mydbinstance)
 Value: 2
 
 Metric #1
@@ -115,6 +146,5 @@ The Oracle DB driver documentation can be found [here.](https://github.com/sijms
 Another usage example is the `go_ora`
 example [here.](https://blogs.oracle.com/developers/post/connecting-a-go-application-to-oracle-database)
 
-[alpha]:https://github.com/open-telemetry/opentelemetry-collector#alpha
-
-[contrib]:https://github.com/open-telemetry/opentelemetry-collector-releases/tree/main/distributions/otelcol-contrib
+[alpha]: https://github.com/open-telemetry/opentelemetry-collector#alpha
+[contrib]: https://github.com/open-telemetry/opentelemetry-collector-releases/tree/main/distributions/otelcol-contrib

receiver/sqlqueryreceiver/config.go

@@ -52,15 +52,21 @@ func (c Config) Validate() error {
 type Query struct {
 	SQL     string      `mapstructure:"sql"`
 	Metrics []MetricCfg `mapstructure:"metrics"`
+	Logs    []LogsCfg   `mapstructure:"logs"`
 }
 
 func (q Query) Validate() error {
 	var errs error
 	if q.SQL == "" {
 		errs = multierr.Append(errs, errors.New("'query.sql' cannot be empty"))
 	}
-	if len(q.Metrics) == 0 {
-		errs = multierr.Append(errs, errors.New("'query.metrics' cannot be empty"))
+	if len(q.Logs) == 0 && len(q.Metrics) == 0 {
+		errs = multierr.Append(errs, errors.New("at least one of 'query.logs' and 'query.metrics' must not be empty"))
+	}
+	for _, logs := range q.Logs {
+		if err := logs.Validate(); err != nil {
+			errs = multierr.Append(errs, err)
+		}
 	}
 	for _, metric := range q.Metrics {
 		if err := metric.Validate(); err != nil {
@@ -70,6 +76,14 @@ func (q Query) Validate() error {
 	return errs
 }
 
+type LogsCfg struct {
+}
+
+func (config LogsCfg) Validate() error {
+	var errs error
+	return errs
+}
+
 type MetricCfg struct {
 	MetricName       string            `mapstructure:"metric_name"`
 	ValueColumn      string            `mapstructure:"value_column"`

receiver/sqlqueryreceiver/config_test.go

@@ -104,15 +104,34 @@ func TestLoadConfig(t *testing.T) {
 			errorMessage: "'driver' cannot be empty",
 		},
 		{
-			fname:        "config-invalid-missing-metrics.yaml",
+			fname:        "config-invalid-missing-logs-metrics.yaml",
 			id:           component.NewIDWithName(typeStr, ""),
-			errorMessage: "'query.metrics' cannot be empty",
+			errorMessage: "at least one of 'query.logs' and 'query.metrics' must not be empty",
 		},
 		{
 			fname:        "config-invalid-missing-datasource.yaml",
 			id:           component.NewIDWithName(typeStr, ""),
 			errorMessage: "'datasource' cannot be empty",
 		},
+		{
+			fname: "config-logs.yaml",
+			id:    component.NewIDWithName(typeStr, ""),
+			expected: &Config{
+				ScraperControllerSettings: scraperhelper.ScraperControllerSettings{
+					CollectionInterval: 10 * time.Second,
+				},
+				Driver:     "mydriver",
+				DataSource: "host=localhost port=5432 user=me password=s3cr3t sslmode=disable",
+				Queries: []Query{
+					{
+						SQL: "select * from test_logs",
+						Logs: []LogsCfg{
+							{},
+						},
+					},
+				},
+			},
+		},
 		{
 			fname:        "config-unnecessary-aggregation.yaml",
 			id:           component.NewIDWithName(typeStr, ""),
@@ -121,7 +140,7 @@ func TestLoadConfig(t *testing.T) {
 	}
 
 	for _, tt := range tests {
-		t.Run(tt.id.String(), func(t *testing.T) {
+		t.Run(tt.fname, func(t *testing.T) {
 			cm, err := confmaptest.LoadConf(filepath.Join("testdata", tt.fname))
 			require.NoError(t, err)
 

receiver/sqlqueryreceiver/testdata/config-logs.yaml

@@ -0,0 +1,8 @@
+sqlquery:
+  collection_interval: 10s
+  driver: mydriver
+  datasource: "host=localhost port=5432 user=me password=s3cr3t sslmode=disable"
+  queries:
+    - sql: "select * from test_logs"
+      logs:
+      - {}