Skip to content

armgit / xerj

Public Imported from GitHub
0

AI search engine and MCP server for bots and agents. One Rust binary indexes code, docs, logs, JSONL/CSV and PDFs for BM25 full-text, vector and hybrid search, code search without grep, agent memory and RAG. Elasticsearch-compatible API; Corpus Hub serves hundreds of reference memory corpora.

test(es-yaml): ml datafeed action arms + 10_datafeed.yml

xerj-org2fc3a1a6257b
diff --git a/engine/tests/es-compat-yaml/src/main.rs b/engine/tests/es-compat-yaml/src/main.rs
index b753b09a3..374910f79 100644
--- a/engine/tests/es-compat-yaml/src/main.rs
+++ b/engine/tests/es-compat-yaml/src/main.rs
@@ -1358,6 +1358,62 @@ fn resolve_action(action: &str, params: &serde_yaml::Mapping) -> (String, String
// Security — API keys
"security.create_api_key" => ("POST".into(), "/_security/api_key".into(), body),
+ // ML — anomaly detectors, datafeeds, results
+ "ml.put_job" => {
+ let job = get_param_str(params, "job_id").unwrap_or_default();
+ (
+ "PUT".into(),
+ format!("/_ml/anomaly_detectors/{}", job),
+ body,
+ )
+ }
+ "ml.delete_job" => {
+ let job = get_param_str(params, "job_id").unwrap_or_default();
+ (
+ "DELETE".into(),
+ format!("/_ml/anomaly_detectors/{}", job),
+ None,
+ )
+ }
+ "ml.get_records" => {
+ let job = get_param_str(params, "job_id").unwrap_or_default();
+ (
+ "GET".into(),
+ format!("/_ml/anomaly_detectors/{}/results/records", job),
+ body,
+ )
+ }
+ "ml.put_datafeed" => {
+ let feed = get_param_str(params, "datafeed_id").unwrap_or_default();
+ ("PUT".into(), format!("/_ml/datafeeds/{}", feed), body)
+ }
+ "ml.get_datafeeds" | "ml.get_datafeed" => match get_param_str(params, "datafeed_id") {
+ Some(feed) if !feed.is_empty() => {
+ ("GET".into(), format!("/_ml/datafeeds/{}", feed), None)
+ }
+ _ => ("GET".into(), "/_ml/datafeeds".into(), None),
+ },
+ "ml.delete_datafeed" => {
+ let feed = get_param_str(params, "datafeed_id").unwrap_or_default();
+ ("DELETE".into(), format!("/_ml/datafeeds/{}", feed), None)
+ }
+ "ml.start_datafeed" => {
+ let feed = get_param_str(params, "datafeed_id").unwrap_or_default();
+ (
+ "POST".into(),
+ format!("/_ml/datafeeds/{}/_start", feed),
+ None,
+ )
+ }
+ "ml.stop_datafeed" => {
+ let feed = get_param_str(params, "datafeed_id").unwrap_or_default();
+ (
+ "POST".into(),
+ format!("/_ml/datafeeds/{}/_stop", feed),
+ None,
+ )
+ }
+
// Catch-all
_ => {
let path = format!("/{}", action.replace('.', "/"));
diff --git a/engine/tests/es-compat-yaml/yaml/ml/10_datafeed.yml b/engine/tests/es-compat-yaml/yaml/ml/10_datafeed.yml
new file mode 100644
index 000000000..2b6f37831
--- /dev/null
+++ b/engine/tests/es-compat-yaml/yaml/ml/10_datafeed.yml
@@ -0,0 +1,129 @@
+setup:
+ - do:
+ indices.create:
+ index: ml_cpu
+ body:
+ mappings:
+ properties:
+ "@timestamp":
+ type: date
+ cpu:
+ type: double
+
+ # One reading per minute; a runaway spike at minute 12 (96 vs the ~20 baseline).
+ - do:
+ bulk:
+ index: ml_cpu
+ refresh: true
+ body:
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:00:00Z", "cpu": 19}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:01:00Z", "cpu": 21}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:02:00Z", "cpu": 20}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:03:00Z", "cpu": 20}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:04:00Z", "cpu": 22}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:05:00Z", "cpu": 18}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:06:00Z", "cpu": 21}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:07:00Z", "cpu": 20}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:08:00Z", "cpu": 19}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:09:00Z", "cpu": 21}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:10:00Z", "cpu": 20}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:11:00Z", "cpu": 22}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:12:00Z", "cpu": 96}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:13:00Z", "cpu": 20}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:14:00Z", "cpu": 21}'
+ - '{"index": {}}'
+ - '{"@timestamp": "2026-07-06T00:15:00Z", "cpu": 19}'
+
+---
+"Continuous datafeed scores a spike and supports the start/stop/delete lifecycle":
+ # Create the anomaly-detection job (detector).
+ - do:
+ ml.put_job:
+ job_id: cpu-spike
+ body:
+ source_index: ml_cpu
+ time_field: "@timestamp"
+ function: mean
+ field: cpu
+ bucket_span: 1m
+ - match: { job_id: cpu-spike }
+
+ # Create a continuous datafeed for it (short frequency).
+ - do:
+ ml.put_datafeed:
+ datafeed_id: cpu-feed
+ body:
+ job_id: cpu-spike
+ frequency: 1s
+ - match: { datafeed_id: cpu-feed }
+ - match: { job_id: cpu-spike }
+ - match: { state: stopped }
+
+ # Datafeed for an unknown job must 404.
+ - do:
+ catch: missing
+ ml.put_datafeed:
+ datafeed_id: bad-feed
+ body:
+ job_id: no-such-job
+
+ # Start it — runs one synchronous scoring pass now.
+ - do:
+ ml.start_datafeed:
+ datafeed_id: cpu-feed
+ - match: { started: true }
+
+ # The spike must have produced at least one anomaly record.
+ - do:
+ ml.get_records:
+ job_id: cpu-spike
+ - gte: { count: 1 }
+ - match: { records.0.job_id: cpu-spike }
+
+ # Datafeed now reports started.
+ - do:
+ ml.get_datafeeds:
+ datafeed_id: cpu-feed
+ - match: { datafeeds.0.state: started }
+
+ # Deleting a started datafeed must conflict.
+ - do:
+ catch: conflict
+ ml.delete_datafeed:
+ datafeed_id: cpu-feed
+
+ # Stop, then delete cleanly.
+ - do:
+ ml.stop_datafeed:
+ datafeed_id: cpu-feed
+ - match: { stopped: true }
+
+ - do:
+ ml.get_datafeeds:
+ datafeed_id: cpu-feed
+ - match: { datafeeds.0.state: stopped }
+
+ - do:
+ ml.delete_datafeed:
+ datafeed_id: cpu-feed
+ - match: { acknowledged: true }
+
+ - do:
+ ml.delete_job:
+ job_id: cpu-spike
+ - match: { acknowledged: true }