#!/usr/bin/env python
# Wenchang Yang (wenchang@princeton.edu)
# Wed Oct  5 11:09:59 EDT 2022
if __name__ == '__main__':
    import sys
    from misc.timer import Timer
    tt = Timer('start ' + ' '.join(sys.argv))
import sys, os.path, os, glob, datetime
import xarray as xr, numpy as np, pandas as pd, matplotlib.pyplot as plt
#more imports
from modelout import get_modelout_data, update_modelout_data
from modelout.getdata import funcs
import xfilter
nwindow, dimlp = 9, 'year'
#lowpass = lambda x: x.filter.lowpass(1/nwindow, dim=dimlp, padtype='odd')
lowpass = lambda x: x.rolling(year=nwindow, center=True, min_periods=1).mean() if x.year.size>9 else x
import geoxarray
#
if __name__ == '__main__':
    tt.check('end import')
#
#start from here
daname = 't_surf'
#func = lambda x: x.load().geo.fldmean()
#funcname = 'glbmean'
func = funcs['nino34']
funcname = 'nino34'

cleanup = False #remove cache or not

das = dict()

model = 'FLOR'
#ctl1990_v201905
label = 'FLOR_ctl_1990_v201905'
expname = 'CTL1990_v201905_tigercpu_intelmpi_18_576PE'
da = get_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname, years=range(1901,2000+1))
das[label] = da
#ctl1990_v201905_FA
label = 'FLOR_ctl_1990_v201905_FA'
expname = 'CTL1990_v201905_FA_tigercpu_intelmpi_18_576PE'
da = update_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname)#, years=range(100,201))
das[label] = da
#ctl1990_v201905_FAtrop
label = 'FLOR_ctl_1990_v201905_FAtrop'
expname = 'CTL1990_v201905_FAtrop_tigercpu_intelmpi_18_576PE'
da = update_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname)#, years=range(100,201))
das[label] = da

#ctl1860_v202407
label = 'FLOR_ctl_1860_v202407'
expname = 'CTL1860_v202407_tigercpu_intelmpi_18_576PE'
da = get_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname, years=range(1901,2000+1))
das[label] = da
#ctl1860_v202407_FA
label = 'FLOR_ctl_1860_v202407_FA'
expname = 'CTL1860_v202407_FA_tigercpu_intelmpi_18_576PE'
da = get_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname, years=range(1901,2000+1))
das[label] = da
#ctl1860_v202407_FAtrop
label = 'FLOR_ctl_1860_v202407_FAtrop'
expname = 'CTL1860_v202407_FAtrop_tigercpu_intelmpi_18_576PE'
da = update_modelout_data(daname=daname, model=model, expname=expname, func=func, funcname=funcname) #, years=range(1901,2000+1))
das[label] = da


#HadISST nino34
#ifile = '/home/wenchang//sGEOCLIM/CM2.5_data/input/SST_TARGETS/temp_restore_HadISST_1870-2017.nc'
ifile = '/scratch/gpfs/GEOCLIM/wenchang/modelInput/CM2.5/input/SST_TARGETS/temp_restore_HadISST_1870-2017.nc' # tiger3
da_obs = xr.open_dataarray(ifile, decode_times=False) \
    .rename(TIME='time', LONGITUDE='lon', LATITUDE='lat') \
    .assign_coords(time=pd.date_range('1870-01', '2017-06',freq='MS')) \
    .sel(time=slice('1980', '2016')) 
nino34_obs = da_obs.sel(lon=slice(-170,-120), lat=slice(-5,5)).geo.fldmean()
 
def wyplot(da, label,**kws):
    #da = da.groupby('time.year').mean('time')
    da = da - 273.15 #K -> degC
    #da = da.pipe(lowpass)
    #da = da.sel(year=slice(1900,2000))
    da.plot(label=label, **kws)
def wyplot_std_mclim(da, label,**kws):
    ax = kws.pop('ax', plt.gca())
    #da = da.groupby('time.year').mean('time')
    #da = da - 273.15 #K -> degC
    #da = da.pipe(lowpass)
    #da = da.sel(year=slice(1900,2000))
    da.groupby('time.month').std('time').roll(month=-5).assign_coords(month=range(6, 18)).plot(label=label, **kws)
    ax.set_xticks(range(6,18))
    ax.set_xticklabels(['Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec', 'Jan', 'Feb', 'Mar', 'Apr', 'May'])
    ax.set_xlabel('')
    
 
if __name__ == '__main__':
    from wyconfig import * #my plot settings
    """
    import cftime
    year_xlim = [1910, None][0] #might be overridden later
    fig,ax = plt.subplots(figsize=(11,4))
    for label,da in das.items():
        wyplot(da, label, ax=ax, alpha=0.6)
    ax.legend(ncol=2)
    ax.set_ylabel(f'Nino3.4 [degC]')
    #ax.set_xlim(1800, 2000)
    if year_xlim is not None:
        ax.set_xlim(None, cftime.datetime(year_xlim, 12, 31, calendar='noleap'))
        ax.set_title('')

    #savefig
    if 'savefig' in sys.argv or 's' in sys.argv:
        figname = __file__.replace('.py', f'.png')
        if year_xlim is not None: figname = figname.replace('.png', f'__1901-{year_xlim}.png')
        if 'overwritefig' in sys.argv or 'o' in sys.argv:
            wysavefig(figname, overwritefig=True)
        else:
            wysavefig(figname)
    """
    #monthly std
    fig,ax = plt.subplots()
    #obs
    nino34_obs.groupby('time.month').std('time').roll(month=-5).assign_coords(month=range(6, 18)).plot(label='HadISST Nino34', color='k')
    #model results
    colors = ['C0', 'C1', 'C2']*2
    linestyles = ['-']*3 + ['--']*3
    for (label,da),color,linestyle in zip(das.items(), colors, linestyles):
        wyplot_std_mclim(da, label, ax=ax, ls=linestyle, color=color)
    ax.legend(ncol=2)
    ax.set_ylabel(f'Nino3.4 STD [degC]')
    #ax.set_xlim(1800, 2000)
    ax.set_title('')
    #savefig
    if 'savefig' in sys.argv or 's' in sys.argv:
        figname = __file__.replace('.py', f'__std.png')
        if 'overwritefig' in sys.argv or 'o' in sys.argv:
            wysavefig(figname, overwritefig=True)
        else:
            wysavefig(figname)

    tt.check(f'**Done**')
    print()
    if 'notshowfig' in sys.argv:
        pass
    else:
        plt.show()