components/databricks/actions/list-runs/list-runs.mjs (1)

42-47: Nice-to-have: respect maxResults when paging to avoid over-fetching.
Throttle params.limit per-iteration to remaining budget.

-    do {
-      const { runs } = await this.databricks.listRuns({
+    do {
+      params.limit = Math.min(this.maxResults - allRuns.length, params.limit);
+      const { runs } = await this.databricks.listRuns({
         params,
         $,
       });

components/databricks/databricks.app.mjs (1)

70-77: Optional: Broaden workspace host support
Hardcoding .cloud.databricks.com restricts this client to AWS; Azure workspaces use *.azuredatabricks.net and GCP uses *.gcp.databricks.com. Accept a full host in $auth.domain if it contains a dot, otherwise append the AWS pattern.

- _baseUrl() {
-   return `https://${this.$auth.domain}.cloud.databricks.com/api/2.0`;
- },
+ _baseUrl() {
+   const host = this.$auth.domain.includes('.')
+     ? this.$auth.domain
+     : `${this.$auth.domain}.cloud.databricks.com`;
+   return `https://${host}/api/2.0`;
+ },

components/databricks/actions/create-endpoint/create-endpoint.mjs (3)

28-31: Harden success summary against missing id

Some responses may not include id (or use a different field). Fall back to the provided name so the summary is still useful.

-      if (response) {
-        $.export("$summary", `Successfully created endpoint with ID ${response.id}.`);
-      }
+      if (response) {
+        const idOrName = response?.id ?? response?.endpoint_id ?? this.name;
+        $.export("$summary", `Successfully created endpoint "${idOrName}".`);
+      }

---

18-26: Validate input early

Reject empty or whitespace-only names before making the API call.

-  async run({ $ }) {
-    try {
+  async run({ $ }) {
+    const name = this.name?.trim();
+    if (!name) {
+      throw new ConfigurationError("Endpoint Name is required.");
+    }
+    try {
       const response = await this.databricks.createEndpoint({
         data: {
-          name: this.name,
+          name,
           endpoint_type: "STANDARD",
         },
         $,
       });

---

10-17: Consider exposing endpoint type as a prop (default STANDARD)

Let users pick the endpoint type if Databricks adds more types, while defaulting to "STANDARD".

   props: {
     databricks,
     name: {
       type: "string",
       label: "Endpoint Name",
       description: "The name of the vector search endpoint",
     },
+    endpointType: {
+      type: "string",
+      label: "Endpoint Type",
+      description: "Type of vector search endpoint.",
+      options: [
+        { label: "STANDARD", value: "STANDARD" },
+      ],
+      default: "STANDARD",
+    },
   },

And pass it:

-          endpoint_type: "STANDARD",
+          endpoint_type: this.endpointType,

components/databricks/actions/delete-endpoint/delete-endpoint.mjs (1)

18-26: Wrap in try/catch and return a stable shape

Delete APIs often return empty bodies. Wrap errors for clearer UX and always return a predictable object.

-  async run({ $ }) {
-    const response = await this.databricks.deleteEndpoint({
-      endpointName: this.endpointName,
-      $,
-    });
-
-    $.export("$summary", `Successfully deleted endpoint "${this.endpointName}".`);
-    return response;
-  },
+  async run({ $ }) {
+    try {
+      const response = await this.databricks.deleteEndpoint({
+        endpointName: this.endpointName,
+        $,
+      });
+      $.export("$summary", `Successfully deleted endpoint "${this.endpointName}".`);
+      return response ?? { success: true, endpointName: this.endpointName };
+    } catch (err) {
+      const message = err?.response?.data?.message
+        || err?.message
+        || `Failed to delete endpoint "${this.endpointName}".`;
+      throw new Error(message);
+    }
+  },

Optional: swap Error for ConfigurationError and import it from @pipedream/platform for consistency.

components/databricks/actions/get-endpoint/get-endpoint.mjs (1)

18-29: Add defensive error handling and resilient summary

Guard against missing bodies and surface better errors.

-  async run({ $ }) {
-    const response = await this.databricks.getEndpoint({
-      endpointName: this.endpointName,
-      $,
-    });
-
-    if (response) {
-      $.export("$summary", `Successfully retrieved endpoint "${this.endpointName}".`);
-    }
-
-    return response;
-  },
+  async run({ $ }) {
+    try {
+      const response = await this.databricks.getEndpoint({
+        endpointName: this.endpointName,
+        $,
+      });
+      $.export("$summary", `Successfully retrieved endpoint "${this.endpointName}".`);
+      return response ?? {};
+    } catch (err) {
+      const message = err?.response?.data?.message
+        || err?.message
+        || `Failed to retrieve endpoint "${this.endpointName}".`;
+      throw new Error(message);
+    }
+  },

components/databricks/actions/list-endpoints/list-endpoints.mjs (2)

11-16: Constrain maxResults

Add a minimal bound to avoid confusion from zero or negative values.

     maxResults: {
       type: "integer",
       label: "Max Results",
       description: "Maximum number of endpoints to return",
       default: 100,
+      min: 1,
     },

---

35-43: Clarify truncation in the summary

Make it explicit when results were truncated to maxResults.

-    $.export("$summary", `Successfully retrieved ${allEndpoints.length} endpoint${allEndpoints.length === 1
-      ? ""
-      : "s"}.`);
+    const truncated = allEndpoints.length > this.maxResults ? " (truncated)" : "";
+    $.export("$summary", `Successfully retrieved ${allEndpoints.length} endpoint${allEndpoints.length === 1 ? "" : "s"}${truncated}.`);

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

💡 Knowledge Base configuration:

• MCP integration is disabled by default for public repositories
• Jira integration is disabled by default for public repositories
• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

Learnt from: jcortes
PR: PipedreamHQ/pipedream#14935
File: components/sailpoint/package.json:15-18
Timestamp: 2024-12-12T19:23:09.039Z
Learning: When developing Pipedream components, do not add built-in Node.js modules like `fs` to `package.json` dependencies, as they are native modules provided by the Node.js runtime.