procparse.py: parse linux /proc filesystem and expose as API

procparse.py

#!/usr/bin/env python3

import io, re, os, sys, time, json, socket, threading

def reader(file):
	with open("/proc/{}".format(file)) as file:
		return file.read().split("\n")
	return []

def generic_parser(file):
	return {
		p[0]: float(p[1])
		for p in [
			re.split(r"[\0:=\s]+", line)
			for line in reader(file)
			if len(line) > 0
		]
	}

def flatten(nobj, *args, sep=".", parent=None, final=None, depth=-1, **kwargs):
	if parent is None:
		parent = tuple()
	if final is  None:
		final = dict()
	#print(parent, final)
	if depth == 0:
		final.update({ parent: nobj })
		return nobj
	elif isinstance(nobj, dict):
		for k, v in nobj.items():
			flatten(v, parent=parent+(k,), final=final, depth=depth-1)
	elif isinstance(nobj, list):
		for i, j in enumerate(nobj):
			flatten(j, parent=parent+(i,), final=final, depth=depth-1)
	else:
		final.update({ parent: nobj })
		return nobj
	#print("upd", len(parent), upd)
	return { sep.join([ str(i) for i in k ]): v for k, v in final.items() }


diskstatstab = [
	"empty",
	"major_number",
	"minor_mumber",
	"device_name",
	"reads_completed_successfully",
	"reads_merged",
	"sectors_read",
	"time_spent_reading_ms",
	"writes_completed",
	"writes_merged",
	"sectors_written",
	"time_spent_writing_ms",
	"IOs_currently_in_progress",
	"time_spent_doing_IOs_ms",
	"weighted_time_spent_doing_IOs_ms",
	"discards_completed_successfully",
	"discards_merged",
	"sectors_discarded",
	"time_spent_discarding",
]

loadavgtab = [
	"01mins",
	"05mins",
	"15mins",
	"proc-thread",
	"lastpid",
]

stattab = [
	"user",
	"nice",
	"system",
	"idle",
	"iowait",
	"irq",
	"softirq",
	"steal",
	"guest",
	"guest_nice",
	"10th_col",
	"11th_col",
]

parsertab = {
	"vmstat": generic_parser,
	"meminfo": generic_parser,
	"loadavg": (lambda x: {
		loadavgtab[i]: j
		for i, j in enumerate(reader(x)[0].split(" "))
	}),
	"diskstats": (lambda x: {
		p[3]: {
			diskstatstab[i]: j
			for i, j in enumerate(p)
			if i is not 3 and len(j) > 0
		}
		for p in [
			re.split(r"\s+", disk)
			for disk in reader(x)
			if len(disk) > 0
		]
	}),
	"stat": (lambda x: {
		p[0]: {
			stattab[i]: float(j)
			for i, j in enumerate(p[1:])
		}
		for p in [
			re.split(r"[\0:=\s]+", line)
			for line in reader(x)
			if len(line) > 0
		] if p[0] not in [ "intr" ]
	}),
	"net/netstat": (lambda x: (lambda data: {
		t[0]: {
			data[(i-1 if i > 0 else 0)][j+1]: v
			for j, v in enumerate(t[1:])
		} for i, t in enumerate(data)
	})([
		re.split(r"[\0:=\s]+", line)
		for line in reader(x)
		if len(line) > 0
	])),
}

get_bulk_data = lambda: {
	file.replace("/", "_"): parsertab[file](file) for file in [
		"net/netstat",
		#"diskstats",
		#"mdstat",
		#"schedstat",
		#"stat",
		#"timerstats",
		"diskstats",
		"meminfo",
		"loadavg",
		"vmstat",
		"stat",
	]
}

def publish(baseurl="https://test-c59fa.firebaseio.com"):
	import urllib.request
	url = "{}/{}/{}.json?print=silent".format(
		baseurl,
		socket.gethostname().replace(".", "-"),
		int(time.time()),
	)
	request = lambda *args, **kwargs: (
		#print(kwargs["data"].decode()),
		json.loads(urllib.request.urlopen(urllib.request.Request(*args, **kwargs)).read().decode()),
	)[-1]
	try:
		resp = request(url,
			method="PUT",
			headers={
				"content-type": "application/json"
			},
			data=json.dumps(get_bulk_data(), indent='\t').encode(),
		)
		with open("/tmp/last.txt", "w") as file:
			json.dump(flatten(resp), file, indent='\t')
		#print(json.dumps(flatten(resp), indent='\t'))
	except Exception as e:
		#print(e)
		pass

def main(args):
	import flask
	app = flask.Flask(__name__)
	app.add_url_rule("/", "root", get_bulk_data)
	return app.run(
		debug=False,
		host="0.0.0.0",
		port=8000,
	)

if __name__ == '__main__':
	sys.exit(main(sys.argv[1:]))