#!/usr/bin/env python3
"""Emit mapping-category totals without path names from local Linux /proc/PID/smaps."""
import argparse
from collections import defaultdict
import json
from pathlib import Path
import re

HEADER = re.compile(r"^[0-9a-f]+-[0-9a-f]+\s+(\S+)\s+\S+\s+\S+\s+\S+\s*(.*)$")
METRIC = re.compile(r"^(Size|Rss|Pss|Private_Dirty|Private_Clean|Anonymous):\s+(\d+) kB$")


def summarize(path):
    groups = defaultdict(lambda: defaultdict(int))
    group = None
    for line in path.read_text().splitlines():
        header = HEADER.match(line)
        if header:
            permissions, name = header.groups()
            if "doublemapper" in name:
                group = "jit-doublemapper"
            elif name == "[heap]":
                group = "native-heap"
            elif name.startswith("[stack"):
                group = "stack"
            elif not name:
                group = "anonymous-exec" if "x" in permissions else "anonymous-write" if "w" in permissions else "anonymous-read"
            elif name.endswith(".dll"):
                group = "managed-assembly-files"
            elif ".so" in name:
                group = "native-library-files"
            elif name.startswith("["):
                group = "other-kernel-mappings"
            else:
                group = "other-files"
        elif group and (metric := METRIC.match(line)):
            groups[group][metric[1]] += int(metric[2]) * 1024
    return dict(groups)


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("files", nargs="+", type=Path)
    args = parser.parse_args()
    print(json.dumps([{"sample": index, "groups": summarize(path)} for index, path in enumerate(args.files)], indent=2))