Add telemetry (initialised when intake.cat.access_nri entrypoint acce…

…ssed)

pyproject.toml

@@ -22,6 +22,7 @@ dependencies = [
     "jsonschema",
     "pooch",
     "xarray",
+    "httpx>=0.28"
 ]
 dynamic = ["version"]
 

src/access_nri_intake/data/__init__.py

@@ -5,13 +5,34 @@
 
 import intake
 import intake.catalog
+from IPython import get_ipython
+from IPython.core.magic import register_line_magic
 
 from access_nri_intake.utils import get_catalog_fp
 
+from .telemetry import capture_datastore_searches
+
 CATALOG_NAME_FORMAT = (
     r"^v(?P<yr>2[0-9]{3})\-(?P<mon>1[0-2]|0[1-9])\-(?P<day>0[1-9]|[1-2][0-9]|3[0-1])$"
 )
 
+
+def load_ipython_extension(ipython):
+    @register_line_magic("capture_func_calls")
+    def capture_func_calls(info):
+        """
+        Returns the function calls from the code in the cell
+        """
+        ipython.events.register("pre_run_cell", capture_datastore_searches)
+        print("Function calls will be captured for the next cell execution")
+
+
+# Register the extension
+ip = get_ipython()
+if ip:
+    load_ipython_extension(ip)
+    ip.run_line_magic("capture_func_calls", "")
+
 try:
     data = intake.open_catalog(get_catalog_fp()).access_nri
 except FileNotFoundError:

src/access_nri_intake/data/telemetry.py

@@ -0,0 +1,82 @@
+import ast
+import asyncio
+import warnings
+
+import httpx
+from IPython import get_ipython
+
+TELEMETRY_SERVER_URL = "https://intake-telemetry-bb870061f91a.herokuapp.com"
+
+
+def send_api_request(function_name, kwargs):
+    telemetry_data = {
+        "name": f"CT_testing_{function_name}_ipy_extensions",
+        "search": kwargs,
+    }
+
+    endpoint = f"{TELEMETRY_SERVER_URL}/telemetry/update"
+
+    async def send_telemetry(data):
+        headers = {"Content-Type": "application/json"}
+        async with httpx.AsyncClient() as client:
+            try:
+                response = await client.post(endpoint, json=data, headers=headers)
+                response.raise_for_status()
+
+                print(f"Telemetry data sent: {response.json()}")
+            except httpx.RequestError as e:
+                warnings.warn(
+                    f"Request failed: {e}", category=RuntimeWarning, stacklevel=2
+                )
+
+    # Schedule the telemetry data to be sent in the background
+    asyncio.create_task(send_telemetry(telemetry_data))
+    return None
+
+
+def capture_datastore_searches(info):
+    """
+    Use the AST module to parse the code that we are executing & send an API call
+    if we
+
+    """
+    code = info.raw_cell
+
+    # Remove lines that contain IPython magic commands
+    code = "\n".join(
+        line for line in code.splitlines() if not line.strip().startswith("%")
+    )
+
+    tree = ast.parse(code)
+    user_namespace = get_ipython().user_ns
+
+    for node in ast.walk(tree):
+        if isinstance(node, ast.Call):
+            if isinstance(node.func, ast.Name):
+                func_name = node.func.id
+            elif isinstance(node.func, ast.Attribute):
+                # Check if the attribute is a method call on a class instance or a module function
+                if isinstance(node.func.value, ast.Name):
+                    instance_name = node.func.value.id
+                    method_name = node.func.attr
+                    try:
+                        # Evaluate the instance to get its class name
+                        instance = eval(instance_name, globals(), user_namespace)
+                        class_name = instance.__class__.__name__
+                        func_name = f"{class_name}.{method_name}"
+                    except Exception as e:
+                        print(f"Error evaluating instance: {e}")
+                        continue
+
+            if func_name in [
+                "esm_datastore.search",
+            ]:
+                # args = [ast.dump(arg) for arg in node.args]
+                kwargs = {kw.arg: ast.literal_eval(kw.value) for kw in node.keywords}
+                send_api_request(
+                    func_name,
+                    # args,
+                    kwargs,
+                )
+            else:
+                print(f"Function call: {func_name}; not tracking")