import os
import sys
import pandas as pd
from pandas import Series
from datetime import datetime
pd.options.mode.chained_assignment = None  # default='warn'
pd.set_option('precision', 1)
pd.set_option('display.max_rows', None)

def lat_lon_to_float(v):
#	Convert strings from NHC file (Lat Lon with N/S and W/E to floats)
	if (v[-1] == 'S') or (v[-1] =='W'):
		multiplier = -1
	else:
		multiplier = 1
	return float(v[:-1]) * multiplier

AnnualList = [['YEAR','NAME','Lat','Lon','Reg']]
SummaryList = []
SummDFCols = ['YEAR', 'MDR', 'GMX', 'EXT','Total', 'MDRFrac','GMXFrac','EXTFrac']
data = []
data2= []
with open('atl_hurdat.txt','r') as f:
	for line in f.readlines():
		if line.startswith('AL'):
			storm_id =line.split(',')
			storm_id.pop()
#			storm_id_str=','.join([str(elem) for elem in storm_id])
			storm_numberFull = storm_id[0].strip()
			storm_name   = storm_id[1].strip()
			basin = storm_numberFull[0:2]
			storm_number = int(storm_numberFull[2:4])
			storm_year   = int(storm_numberFull[4:])
		else:
			location_line = line.split(',')
			location_line.pop()
#			location_line_str=','.join([str(elem) for elem in location_line])
			#print(location_line)
			dt = datetime.strptime(location_line[0]+ location_line[1], '%Y%m%d %H%M')
			dy = datetime.strptime(location_line[0], '%Y%m%d')
			hr = int(location_line[1][1:3])
			mns = int(location_line[1][3:5])
			lfall = location_line[2].strip()
			storm_status = location_line[3].strip()
			storm_lat = lat_lon_to_float(location_line[4].strip())
			storm_lon = lat_lon_to_float(location_line[5].strip())
			max_wind  = int(location_line[6].strip())
			data.append([basin, storm_number, storm_year, storm_name,storm_status, storm_lat, storm_lon, lfall, dt, dy, hr, mns, max_wind])
#			data2.append([storm_id_str,location_line_str])

#bigdf=pd.DataFrame(data2,columns=['id','locline'])
#s=bigdf['id']
#p=s.str.split(pat=",",expand=True)
#s=bigdf['locline']
#t=s.str.split(pat=",",expand=True)
#newBigdf=pd.concat([p,t],axis=1,join="inner")
#print(newBigdf)
#newBigdf.to_csv('./atl_hurdat.csv')

df = pd.DataFrame(data, columns=['Basin','StormNumber', 'StormYear', 'StormName', 'StormStatus', 'Lat', 'Lon', 'LandFall','Date_Time', 'Day', 'Hour', 'Minutes','MaxWind'])
df['MaxWSqd']=df['MaxWind']*df['MaxWind']
df.drop(df[df['StormYear']<1950].index, inplace=True)
#dfPost1950=df
for status in ['TD','EX', 'LO','DB', 'SD', 'WV']:
	df.drop(df[(df['StormStatus'] == status)].index,inplace=True)

years= df['StormYear'].unique()
#years=['2019','2020']
for yr in years:
#	print("\n"+str(yr))
	dfYear=df[df['StormYear']==int(yr)]
#	print(dfYear['StormNumber'].unique().max())
	#print(dfYear.drop_duplicates(subset='StormName'))
	MDRCount = 0
	GMXCount = 0
	EXTCount = 0
	for num in dfYear['StormNumber'].unique():
		dfByNum=dfYear[dfYear['StormNumber']==num]
		#print(dfByNum)
		Name=dfByNum['StormName'].iloc[0]
		Lat=dfByNum['Lat'].iloc[0]
		Lon=dfByNum['Lon'].iloc[0]

		if ((0 < Lat < 21) and (-90 < Lon < -20)):
			FormReg = "MDR"
			MDRCount = MDRCount + 1
		elif ((Lat >= 21) and (Lon < -81)):
			FormReg = "GMX"
			GMXCount = GMXCount + 1
		elif ((Lat >= 18) and (-100 < Lon < -90)):
			FormReg = "GMX"
			GMXCount = GMXCount + 1
		else:
			FormReg = "EXT"
			EXTCount = EXTCount + 1
		AnnualList.append([yr, Name, Lat, Lon, FormReg])
	AllCount = MDRCount + GMXCount + EXTCount
	MDRFrac=(MDRCount/AllCount)*100
	GMXFrac=(GMXCount/AllCount)*100
	EXTFrac=(EXTCount/AllCount)*100
	SummaryList.append([yr, MDRCount, GMXCount, EXTCount,AllCount, MDRFrac, GMXFrac, EXTFrac])
#print(AnnualList)
#print(SummaryList)

SummaryDF=pd.DataFrame(SummaryList, columns=SummDFCols)
SummaryDF.loc['mean']=SummaryDF.mean()
SummaryDF.loc['median']=SummaryDF.median()
print(SummaryDF.describe())

SummaryDF.to_csv("./ATL_1950_Present_FormationRegionStats.csv", index=False)
#os.system('more ./ATL_1950_Present_Stats.csv')