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('display.precision', 1)
pd.set_option('display.max_rows', None)

YearStart=sys.argv[1]
YearEnd=sys.argv[2]
#InMonth=sys.argv[3]



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.csv','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')
			mody= int(location_line[0][4:8])
			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, mody, 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', 'MODY','Hour', 'Minutes','MaxWind'])
df['MaxWSqd']=df['MaxWind']*df['MaxWind']
df.drop(df[df['StormYear']<int(YearStart)].index, inplace=True)
#dfPost1950=df
for status in ['EX', 'LO','DB', 'SD', 'WV']:
	df.drop(df[(df['StormStatus'] == status)].index,inplace=True)
#df.drop_duplicates(['StormYear','StormNumber'],inplace=True)
df['FormationDelta']=df['Day'].diff()
df['FormationDelta']=df['FormationDelta'].dt.days
#print(df)

SummaryList=[]
years= df['StormYear'].unique()
#years=['1950'] #,'2020']
for yr in years:
#	print("\n"+str(yr))
	dfYear=df[df['StormYear']==int(yr)]
	Nums=dfYear['StormNumber'].unique()
	for num in Nums:
		TDLat,TDLon,TDDay,TSLat,TSLon,TSDay,HULat,HULon,HUDay,MHLat,MHLon,MHDay = ["Null","Null","Null","Null","Null","Null","Null","Null","Null","Null","Null","Null"]
		dfName=dfYear[dfYear['StormNumber']==num]
#		print(dfName)
		StormName=dfName['StormName'].iloc[0]
		try:
			IndexFirstTD=dfName[dfName['StormStatus']=='TD'].index[0]
			TDLat= dfName.loc[[IndexFirstTD],['Lat']].loc[IndexFirstTD][0]
			TDLon= dfName.loc[[IndexFirstTD],['Lon']].loc[IndexFirstTD][0]
			TDDay= datetime.strftime(dfName.loc[[IndexFirstTD],['Date_Time']].loc[IndexFirstTD][0],"%m%d%H%M")
		except:
			pass
		try:
			IndexFirstTS=dfName[dfName['StormStatus']=='TS'].index[0]
			TSLat= dfName.loc[[IndexFirstTS],['Lat']].loc[IndexFirstTS][0]
			TSLon= dfName.loc[[IndexFirstTS],['Lon']].loc[IndexFirstTS][0]
			TSDay= datetime.strftime(dfName.loc[[IndexFirstTS],['Date_Time']].loc[IndexFirstTS][0],"%m%d%H%M")
		except:
			pass
		try:
			IndexFirstHU=dfName[dfName['StormStatus']=='HU'].index[0]
			HULat= dfName.loc[[IndexFirstHU],['Lat']].loc[IndexFirstHU][0]
			HULon= dfName.loc[[IndexFirstHU],['Lon']].loc[IndexFirstHU][0]
			HUDay= datetime.strftime(dfName.loc[[IndexFirstHU],['Date_Time']].loc[IndexFirstHU][0],"%m%d%H%M")
#			try:
#						StormName
#					except:
#						StormName=dfName.loc[[IndexFirstHU],['StormName']].loc[IndexFirstHU][0]
		except:
			pass
#			print(StormName + " didn't find HU")

		try:
			IndexFirstMH=dfName[dfName['MaxWind']>95].index[0]
			MHLat= dfName.loc[[IndexFirstMH],['Lat']].loc[IndexFirstMH][0]
			MHLon= dfName.loc[[IndexFirstMH],['Lon']].loc[IndexFirstMH][0]
			MHDay= datetime.strftime(dfName.loc[[IndexFirstMH],['Date_Time']].loc[IndexFirstMH][0],"%m%d%H%M")
		except:
			pass
#			print(StormName + " didn't find HU and > 95")

		IndexMaxWnd=dfName['MaxWind'].idxmax()
		MW=dfName.loc[[IndexMaxWnd],['MaxWind']].loc[IndexMaxWnd][0]
		MWLat= dfName.loc[[IndexMaxWnd],['Lat']].loc[IndexMaxWnd][0]
		MWLon= dfName.loc[[IndexMaxWnd],['Lon']].loc[IndexMaxWnd][0]
		MWDay= datetime.strftime(dfName.loc[[IndexMaxWnd],['Date_Time']].loc[IndexMaxWnd][0],"%m%d%H%M")
#		print(yr,num,StormName,MW, MWLat, MWLon, MWDay, TDLat,TDLon,TDDay,TSLat,TSLon,TSDay,HULat,HULon,HUDay,MHLat,MHLon,MHDay)
		

		SummaryList.append([yr,num,StormName,MW, MWLat, MWLon, MWDay, TDLat,TDLon,TDDay,TSLat,TSLon,TSDay,HULat,HULon,HUDay,MHLat,MHLon,MHDay])
#print(AnnualList)
#print(SummaryList)
SummDFCols=['Year','StmNum','Name','MaxWnd','MxWndMODATIME','MxWndLat','MxWndLon','DepMODATIME', 'DepLat','DepLon','NSMODATIME','NSLat','NSLon','HRMODATIME','HRLat','HRLON','MHMODATIME','MHLat','MHLON']
SummaryDF=pd.DataFrame(SummaryList, columns=SummDFCols)
#SummaryDF.loc['mean']=SummaryDF.mean()
#SummaryDF.loc['median']=SummaryDF.median()
print(SummaryDF)

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