from fastapi import APIRouter, Response
from database import fetch_all
from kubernetes import client, config
import yaml
import json
import logging

router = APIRouter()

# Helper functions for conversions
def convert_cpu_to_cores(cpu):
    """
    Convert CPU usage to cores for human-readable format.
    Handles units: n (nano), u (micro), m (milli), or none (cores).
    Returns float values for cores, rounded appropriately.
    """
    if "n" in cpu:  # Nanocores to cores
        return round(int(cpu.replace("n", "")) / 1e9, 4)
    elif "u" in cpu:  # Microcores to cores
        return round(int(cpu.replace("u", "")) / 1e6, 4)
    elif "m" in cpu:  # Millicores to cores
        return round(int(cpu.replace("m", "")) / 1000, 4)
    return float(cpu)  # Already in cores

def convert_memory_to_mib(memory):
    """
    Convert memory to MiB (mebibytes).
    Handles units: Ki (kibibytes), Mi (mebibytes), Gi (gibibytes).
    """
    if "Ki" in memory:
        return int(memory.replace("Ki", "")) / 1024
    elif "Mi" in memory:
        return int(memory.replace("Mi", ""))
    elif "Gi" in memory:
        return int(memory.replace("Gi", "")) * 1024
    return float(memory)

# Fetch Kubernetes data with namespace resource usage
def fetch_k8s_data_with_usage():
    config.load_incluster_config()
    v1 = client.CoreV1Api()
    metrics_client = client.CustomObjectsApi()

    # Fetch nodes
    nodes = [{
        "node_name": node.metadata.name,
        "cpu": node.status.capacity.get("cpu"),
        "memory": round(convert_memory_to_mib(node.status.capacity.get("memory")), 2),  # Convert to MiB
        "pods_allocatable": node.status.allocatable.get("pods"),
    } for node in v1.list_node().items]

    # Fetch namespaces
    namespaces = [ns.metadata.name for ns in v1.list_namespace().items]

    # Fetch pod metrics and calculate namespace resource usage
    namespace_usage = {}
    pod_metrics = metrics_client.list_cluster_custom_object(
        group="metrics.k8s.io", version="v1beta1", plural="pods"
    )
    for pod in pod_metrics["items"]:
        pod_namespace = pod["metadata"]["namespace"]
        if pod_namespace not in namespace_usage:
            namespace_usage[pod_namespace] = {"cpu": 0, "memory": 0}

        for container in pod["containers"]:
            cpu_usage = container["usage"]["cpu"]
            memory_usage = container["usage"]["memory"]

            # Convert CPU to cores and memory to MiB
            namespace_usage[pod_namespace]["cpu"] += convert_cpu_to_cores(cpu_usage)
            namespace_usage[pod_namespace]["memory"] += convert_memory_to_mib(memory_usage)

    # Round and format usage for readability
    namespace_usage = {
        ns: {
            "cpu": round(usage["cpu"], 4),  # Round to 4 decimal places
            "memory": round(usage["memory"], 2),  # Memory in MiB
        }
        for ns, usage in namespace_usage.items()
    }

    return {"nodes": nodes, "namespaces": namespaces, "namespace_usage": namespace_usage}


# Export endpoint
@router.get("/export")
def export_data(format: str = "yaml"):
    data = {
        "metal_nodes": fetch_all("metal_nodes"),
        "virtual_machines": fetch_all("virtual_machines"),
        "kubernetes": fetch_k8s_data_with_usage(),
    }

    if format.lower() == "yaml":
        yaml_data = yaml.safe_dump(data, sort_keys=False)
        return Response(content=yaml_data, media_type="text/yaml")

    return Response(content=json.dumps(data, indent=2), media_type="application/json")