import collections
import csv
import json
import os
import urllib.request

HERE = os.path.dirname(os.path.abspath(__file__))

CSV_URL = "https://raw.githubusercontent.com/elastic/ecs/main/generated/csv/fields.csv"
CSV_PATH = os.path.join(HERE, "ecs_fields.csv")
OUT_PATH = os.path.join(HERE, "ecs_log.json")


def fetch_csv():
    if not os.path.exists(CSV_PATH):
        urllib.request.urlretrieve(CSV_URL, CSV_PATH)
    return CSV_PATH


def kind_for(ecs_type):
    t = (ecs_type or "").strip()
    if t in ("long", "integer"):
        return "integer"
    if t in ("float", "double", "scaled_float", "half_float"):
        return "float"
    if t == "boolean":
        return "boolean"
    if t in ("object", "flattened", "nested", "geo_point"):
        return "json"
    # keyword, text, match_only_text, wildcard, constant_keyword, ip, date,
    # version, ... -> string
    return "string"


def build_values(csv_path):
    values = collections.OrderedDict()
    with open(csv_path, newline="") as f:
        for row in csv.DictReader(f):
            name = row["Field"].strip()
            key = name.replace(".", "/")
            if key in values:
                continue
            kind = kind_for(row["Type"])
            entry = {"kind": kind}
            leaf = name.rsplit(".", 1)[-1]
            if leaf in ("id", "name") and kind == "string":
                entry["identifier"] = True
            values[key] = entry
    return values


def build_format(values):
    return {
        "$schema": "https://lnav.org/schemas/format-v1.schema.json",
        "ecs_log": {
            "title": "ecs",
            "description": "Elastic Common Schema (ECS) defines a common set of fields for ingesting data into Elasticsearch including log records",
            "url": "https://www.elastic.co/guide/en/ecs-logging/overview/current/intro.html",
            "file-type": "json",
            "line-format": [
                {"field": "__timestamp__", "auto-width": True, "align": "left",
                 "timestamp-format": "%y-%m-%d %H:%M:%S.%L"},
                " ",
                {"field": "log/logger", "auto-width": True, "align": "left"},
                " ",
                {"field": "log/level", "align": "left", "min-width": 4, "max-width": 4,
                 "overflow": "truncate", "prefix": "[", "suffix": "]"},
                " ",
                {"field": "message"},
            ],
            "value": values,
            "level-field": "log/level",
            "thread-id-field": "process/thread/id",
            "timestamp-field": "@timestamp",
            "convert-to-local-time": True,
            "hide-extra": True,
        },
    }


def main():
    csv_path = fetch_csv()
    values = build_values(csv_path)
    fmt = build_format(values)
    with open(OUT_PATH, "w") as f:
        json.dump(fmt, f, indent=4)
        f.write("\n")
    print(f"fields: {len(values)}")
    print(f"written: {OUT_PATH}")


if __name__ == "__main__":
    main()