#!/usr/bin/env python
#
# Pull the pDAQ hit rates from all stringHub-*.moni files in the specified
# directory.
#
# John Kelley
# jkelley@icecube.wisc.edu
#
#-------------------------------------------------

import numpy as np
import sys
import glob
import re
import time
import datetime
import string
import json
import operator

# Naming convention for stringHub moni files
FILEPAT = "stringHub-*.moni"
   
# Use HLC?  Otherwise use SLC
HLC = False

#-------------------------------------------------
# Rate reporting in the log follows this format:
#
# DataCollectorMonitor-00A: 2012-04-27 02:11:32.712034:
#	 NumHits: 8761
#	 RunState: RUNNING
#	 NumTcal: 17
#	 MainboardId: 12b78d1cea33
#	 LBMOverflowCount: 0
#	 AcquisitionLoopCount: 264
#	 HitRate: 760.1
#	 HitRateLC: 25.3
#	 NumMoni: 152
#	 NumSupernova: 1
#
# although field order can be different in older runs!!!
#
#-------------------------------------------------

# MB lookup file
NICKNAMES = "nicknames.txt"

#-------------------------------------------------

if (len(sys.argv) != 2):
    print("Usage: %s <log directory>" % (sys.argv[0]))
    sys.exit(0)

dirName = sys.argv[1]
m = re.match("daqrun(\d+)", dirName)
if m:
    runNumber = m.group(1)
else:
    print("ERROR: log directory should be daqrun######")
    sys.exit(1)

#-------------------------------------------------
# Look up the mbid to position mapping

try:
    f = open(NICKNAMES, "r")
except:
    print("ERROR: couldn't open nicknames file for MBID mapping.")
    sys.exit(1)

nameDict = {}
posDict = {}
# Skip header
f.readline() 
for line in f.readlines():
    vals = line.split()
    if len(vals) >= 4:
        mbid = vals[0]
        name = vals[2]
        pos = vals[3]
        nameDict[mbid] = name
        posDict[mbid] = pos        
f.close()

#-------------------------------------------------

# Get the filenames from the directory
files = glob.glob(dirName+"/"+FILEPAT)
if not files:
    print("ERROR: couldn't find stringHub log files in directory",dirName,".")
    sys.exit(1)

# Iterate through the files
rateDict = {}

# TEMP FIX ME
rateArr = []
timeArr = []

for fileName in files:
    try:
        f = open(fileName, "r")        
    except:
        print("WARNING: couldn't open file",f,"for reading: skipped!")
        continue

    startTime = None

    foundMoni = False
    for line in f.readlines():
        # Reset on newline
        m = re.match("^\s*$", line)
        if m:
            foundMoni = False
            continue
        
        m = re.match("DataCollectorMonitor-([0-9][0-9][AB]):\s*(.*):", line)
        if m:
            timeStr = m.group(2)
            mbid = ""
            rate = 0.
            foundMoni = True
            continue

        if foundMoni:
            m = re.match("\s*MainboardId:\s*([0-9a-f]+)\s*", line)
            if m:
                mbid = m.group(1)
                # Fix mainboard id if not zero-padded (moni bug)
                if len(mbid) != 12:
                    z = "0"*(12-len(mbid))
                    mbid = z+mbid
                continue
        
            if not HLC:
                m = re.match("\s*HitRate:\s*([0-9.]+)\s*", line)
            else:
                m = re.match("\s*HitRateLC:\s*([0-9.]+)\s*", line)

            if m:
                rate = float(m.group(1))
                if mbid in rateDict:
                    rateDict[mbid].append(rate)
                else:
                    if mbid != "":
                        rateDict[mbid] = [rate]
                    else:
                        print("WARNING: found rate without mbid!")
                continue
        
    f.close()

# Calculate median and stddev
ratestats = {}
for mbid,pos in sorted(posDict.items(), key=operator.itemgetter(1)):
    if mbid in rateDict:
        rate_median = np.median(rateDict[mbid])
        rate_std = np.std(rateDict[mbid])
        poskey = pos.replace("-", ",")
        ratestats[poskey] = {"rate_hz": rate_median, "stddev_hz": rate_std}
        
# Find longest name for pretty-printing
longName = max(nameDict.items(), key=lambda a:len(a[1]))
longLen = len(longName[1])

# Save the results as text, sorted by position
with open('rates_%s.txt' % runNumber, 'w') as outfile:
    print("Pos \t%sRate[Hz]\tStdDev[Hz]" % ("Name".ljust(longLen)), file=outfile)
    for mbid,pos in sorted(posDict.items(), key=operator.itemgetter(1)):
        poskey = pos.replace("-", ",")
        if poskey in ratestats:
            print("%s\t%s\t%.2f\t%.2f" % (pos, nameDict[mbid].ljust(longLen),
                                          ratestats[poskey]["rate_hz"],
                                          ratestats[poskey]["stddev_hz"]), file=outfile)

# Also dump the result as JSON
with open('rates_%s.json' % runNumber, 'w') as outfile:
    json.dump(ratestats, outfile, sort_keys = True, indent = 4, ensure_ascii=False)