#!/usr/bin/env python
# -*- coding: UTF-8 -*-

from pycen import con_bdcen, con_cad
from os import getlogin
from os.path import join
from sqlalchemy.engine import URL
from sqlalchemy import create_engine
import geopandas as gpd
import pandas as pd

user   = 'postgres'
pwd    = 'postgres'
adr    = '172.17.0.2'
base   = 'postgres'
url = URL.create('postgresql+psycopg2', 
  username=user,
  password=pwd,
  host=adr,
  database=base,
)
con_vm = create_engine(url)

annee = 2023

PATH = '/media/{login}/SRV/FICHIERS/OUTILS/BASES DE DONNEES/BILAN_FEDE_CEN/{an}/'
path_source = join(PATH,'Docs_SIG_joints{an}/').format(login=getlogin(),an=annee)
dico_data = 'Dico_DATA_sites_CEN_v%i_v1.xlsx' % annee
sheet_par = 'parcelles_cen_xx_%i' % annee
sheet_sit = 'sites_cen_xx_%i' % annee

# lv_cen = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=1, usecols='A:D',nrows=25)
# lv_typmaitr = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=1, usecols='F:K',nrows=23)
# lv_typmilie = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=1, usecols='M:N',nrows=16)
# lv_echelnum = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=1, usecols='P:Q',nrows=8)
# lv_typhab   = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=13, usecols='P:Q',nrows=5)
# lv_sourcegeom = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=1, usecols='S:U',nrows=12)
# lv_lienres = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=20, usecols='M:N',nrows=4)
# lv_typprop = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=45, usecols='A:H',nrows=12)
# lv_codgeol = pd.read_excel(path_source+dico_data,sheet_name='listes_valeurs (copy&modif)',header=36, usecols='M:T',nrows=19)
lst_cols_parc = pd.read_excel(path_source+dico_data,sheet_name=sheet_par,header=0, usecols='F',nrows=26)\
    ['nom du champ'].tolist()

lst_prop = ['Acquis par le CEN ou CREN','ACQUIS&CONVENTION', 'PARTIELLEMENT MAITRISE','ACQUIS PAR LE CEN EN BND','ACQUIS PAR LE CEN ISERE']
lst_conv = ["CONVENTION D'USAGE",'CONVENTION']
lst_gere = ['BAIL EMPHYTEOTIQUE', *lst_conv, *lst_prop]

DICT_COLS = {
    'id' : 'id_site_cen_parc',
    'par_id' : 'num_parcelle',
    'section': 'num_section',
    'code_dep':'insee_dep', 
    'code_com':'insee_com',

}

def update_idproprietaire(df):
    if 'id_proprietaire' in df.columns :
        df['id_proprietaire'] = 0

    df.loc[df.classif_prop.str.contains('CEN RA'),'id_proprietaire'] = 28
    df.loc[df.classif_prop.str.contains('CEN ISERE'),'id_proprietaire'] = 14
    df.id_proprietaire.fillna(0,inplace=True)

    return df



def update_typeprop(df):

    DICT = {
        'Etat':'2_1_PUB',
        'CEN RA - BND':'3_CEN',
        'CEN RA':'3_CEN',
        'CEN ISERE - BND':'3_CEN',
        'Privé et entreprises':'1_PRI',
        'Communes et comcom':'2_4_COM',
        'Autres public et assimiliés (syndicats)':'2_0_MIX',
        'Département':'2_3_PUB',
        'Autres asso et fédérations':'2_6_PRI',
        'FRUP':'3_CEN',
    }

    df['type_prop'] = df['classif_prop'].copy()
    df['type_prop'].replace(DICT, inplace=True)
    
    df.loc[
        (df.classif_prop=='Privé et entreprises') &
        (df.gtoper.str.contains('2')),
        'type_prop'] = '2_0_MIX'
    df.loc[
        df.ccogrm == 9,
        'type_prop'] = '2_9_PUB'

    return df


def update_codemfu(df):

    df.loc[df.maitrise.isin(lst_prop),'code_mfu1'] = 'P'
    df.loc[df.maitrise.isin(lst_conv),'code_mfu1'] = 'C'
    df.loc[df.maitrise == 'BAIL EMPHYTEOTIQUE','code_mfu1'] = 'L'
    df.loc[df.maitrise.isin(lst_prop),'code_mfu2'] = 'P1'
    df.loc[df.maitrise.isin(lst_conv),'code_mfu2'] = 'C7'
    df.loc[df.maitrise == 'BAIL EMPHYTEOTIQUE','code_mfu2'] = 'L1'

    df.loc[df.maitrise == 'ACCORD VERBAL','code_mfu1'] = 'C'
    df.loc[df.maitrise == 'ACCORD VERBAL','code_mfu2'] = 'C17'

    df.loc[df.maitrise == 'CONVENTION','code_mfu1'] = 'C'
    df.loc[df.maitrise == 'CONVENTION','code_mfu2'] = 'C7'

    df.loc[df.indivision=='USUFRUIT','code_mfu2'] = 'P2'

    return df


def update_bnd(df):

    df.loc[df.maitrise == 'PARTIELLEMENT MAITRISE','bnd'] = 1
    df.loc[df.maitrise != 'PARTIELLEMENT MAITRISE','bnd'] = 0
    df.loc[df.maitrise == 'ACQUIS&CONVENTION','bnd'] = 1

    return df
    

def get_parc_milli():

    sqlsit = '''
        SELECT * FROM %s.%s 
        WHERE (date_fin is NULL OR date_fin >= '%i-01-01') AND type_site='MILI' ;'''%('sites','c_sites_zonages',annee)
    df1 = gpd.read_postgis(sql=sqlsit,geom_col='geom',con=con_bdcen)

    sql0 = '''SELECT 
    t1.geo_parcelle, substring(t1.geo_parcelle from 1 for 2)||substring(t1.geo_parcelle from 4 for 12) par_id, 
    t1.annee, t1.update_dat, t4.ccopre, t4.ccosec, t4.dcntpa, t1.geom, STRING_AGG(t3.gtoper::text,',') gtoper, STRING_AGG(t3.ccogrm::text,',') ccogrm,
    STRING_AGG(CONCAT(t3.gtoper::text||COALESCE('_'||t3.ccogrm::text,'')),',') type_prop, STRING_AGG(t3.ddenom::text,',') ddenom
    FROM "{sch}"."{t1}" t1 
    LEFT JOIN ("{sch}".parcelle_info t2 
        LEFT JOIN "{sch}".proprietaire t3 USING (comptecommunal))
    USING (geo_parcelle) 
    LEFT JOIN "{sch}".parcelle t4 ON t1.geo_parcelle = t4.parcelle
    '''.format(
        sch='38_%i07'%(annee-1), 
        t1='v_geo_parcelle')
    sql1 = '''
        WHERE ST_INTERSECTS(t1.geom,'SRID={epsg};{poly}')
        GROUP BY 1,2,3,4,5,6,7,8
    '''.format(epsg=2154,poly=df1.unary_union.buffer(-1))
    sql = sql0 + sql1
    df2 = gpd.read_postgis(sql,con_cad)
    df2['code_dep'] = df2.par_id.str[:2]
    df2['code_com'] = df2.par_id.str[2:5]
    df2['code_mfu1'] = 'C'
    df2['code_mfu2'] = 'C12'
    df2['section'] = df2.ccosec.copy()
    df2['id_site_cen_parc'] = 0
    df2['bnd'] = 0
    df2['mesure_compens'] = 0
    df2['pour_part'] = 0

    return df2


def update_mc(df):

    sqlsit = '''
        SELECT * FROM %s.%s 
        WHERE (date_fin is NULL OR date_fin >= '%i-01-01') AND type_site='MC' ;'''%('sites','c_sites_zonages',annee)
    df1 = gpd.read_postgis(sql=sqlsit,geom_col='geom',con=con_bdcen)

    is_mc = df.intersects(df1.unary_union)
    df['mesure_compens'] = 0
    df.loc[
        (is_mc & df.maitrise.isin(lst_conv)
        )|(df.Parc_MCE=='OUI'),
        'mesure_compens'
    ] = 1

    return df


def update_pourpart(df):

    df['pour_part'] = 0
    df.loc[
        df.maitrise=='PARTIELLEMENT MAITRISE',
        'pour_part'
    ] = 1

    return df


def is_rnx(df):

    sqlsit = '''
        SELECT * FROM %s.%s 
        WHERE (date_fin is NULL OR date_fin >= '%i-01-01') AND type_site='RNN' ;'''%('sites','c_sites_zonages',annee)
    df1 = gpd.read_postgis(sql=sqlsit,geom_col='geom',con=con_bdcen)

    isrnx = df.intersects(df1.unary_union.buffer(-1))
    df['parc_gestion_rnx'] = 0
    df.loc[isrnx,'parc_gestion_rnx'] = 1

    return df


def update_datemajparcelle(df,cad):

    tmp = df.merge(
        parcx[['par_id','update_dat']],
        how='left',
        left_on='num_parcelle',
        right_on='par_id')
    tmp.rename(columns={'update_dat':'date_maj_parcelle'},inplace=True)
    
    return tmp


def modif_FRUP2023(df):

    df = df[
        ~df.commentair.str.contains(
            # 'sign&2023|acq&2023',
            r'sign.*2023|acq.*2023',
            na=False,
            case=False,
            regex=True)]
    is_frup = df.maitrise=='FRUP'
    df.loc[is_frup,'classif_prop'] = 'CEN ISERE'
    df.loc[is_frup,'maitrise'] = 'ACQUIS PAR LE CEN ISERE'
    
    return df


def complete_tab(df):

    df['id_site_cen_parc'] = df['id_site_cen_parc'].astype(int)
    df['id_site_fcen_parc'] = None
    df['bnd'] = df['bnd'].astype(int)
    df.loc[df.type_prop.str.contains(','),'type_prop'] = '2_0'

    df['surf_ore_m2'] = 0
    df['date_debut_ore'] = None
    df['date_fin_ore'] = None
    df['doc_foncier'] = 1
    df['source_doc_foncier'] = 0
    df['source_geom_parc_nature'] = 2 
    df['source_geom_parc_annee'] = annee - 1
    df['echelle_num_parc'] = 0
    df['source_surf_parc'] = 1
    df = update_datemajparcelle(df,parcx)
    df['domaine_public'] = 0

    return df



if __name__ == "__main__":

    shplastsend = gpd.read_file(
        join(PATH.format(login=getlogin(),an=annee-1),'ENVOIE','Parcelles_CEN_38_2022.shp')
    )
    lastsend = pd.read_csv(PATH.format(login=getlogin(),an=annee-1)+'ENVOIE/Parcelles_CEN_38_2022.csv')

    sql = '''
        SELECT * FROM foncier.c_anim_foncier_V2
    WHERE maitrise not in ('Acquis par le CDI','Acquis par la commune ou Comcom','CONVENTION DE PASSAGE') AND maitrise is not NULL
        ; '''
    dfparc = gpd.read_postgis(sql,con_bdcen)

    if annee == 2023:
        dfparc = modif_FRUP2023(dfparc)

    dfparc = update_codemfu(dfparc)
    dfparc = update_bnd(dfparc)
    dfparc = update_pourpart(dfparc)
    dfparc = update_mc(dfparc)
    dfparc = is_rnx(dfparc)
    dfparc = update_idproprietaire(dfparc)
    dfparc = update_typeprop(dfparc)


    dfparc['surf_parc_maitrise_m2'] = dfparc.area

    # COLUMNS : 
    # Index(['par_id', 'geom', 'dcntpa', 'typprop_id', 'ccopre', 'section', 'dnupla',
    #    'nom_com', 'numero', 'code_dep', 'code_com', 'gtoper', 'ccogrm',
    #    'type_prop', 'nom_proprio', 'id', 'Parc_MCE', 'docs_off_num',
    #    'Nbr_lots_BND', 'partiel_conv_ou_acq', 'code_mfu1', 'code_mfu2', 'bnd',
    #    'mesure_compens', 'pour_part', 'par_id_v2', 'maitrise', 'date_der_c',
    #    'indivision', 'contact', 'avis_propr', 'commentair', 'nom_proprio_old',
    #    'site_id', 'type_zone', 'last_upd_parc', 'classif_prop', 'date_debut',
    #    'date_fin'],
    #   dtype='object')

    # 
    tmp = dfparc.rename(columns=DICT_COLS).columns.isin(lst_cols_parc)
    lst_col = dfparc.rename(columns=DICT_COLS).columns[tmp]
    parc = dfparc.rename(columns=DICT_COLS)[lst_col].copy()
    shp_parc = dfparc[['par_id','geom']].rename(columns=DICT_COLS).copy()


    sql0 = '''SELECT 
    t1.geo_parcelle, substring(t1.geo_parcelle from 1 for 2)||substring(t1.geo_parcelle from 4 for 12) par_id, 
    t1.annee, t1.update_dat, t4.ccopre, t4.ccosec, t4.dcntpa, t1.geom, STRING_AGG(t3.gtoper::text,',') gtoper, STRING_AGG(t3.ccogrm::text,',') ccogrm,
    STRING_AGG(CONCAT(t3.gtoper::text||COALESCE('_'||t3.ccogrm::text,'')),',') type_prop, STRING_AGG(t3.ddenom::text,',') ddenom
    FROM "{sch}"."{t1}" t1 
    LEFT JOIN ("{sch}".parcelle_info t2 
        LEFT JOIN "{sch}".proprietaire t3 USING (comptecommunal))
    USING (geo_parcelle) 
    LEFT JOIN "{sch}".parcelle t4 ON t1.geo_parcelle = t4.parcelle
    '''.format(
        sch='38_202207', 
        t1='v_geo_parcelle')
    sql1 = '''WHERE substring(t1.geo_parcelle from 1 for 2)||substring(t1.geo_parcelle from 4 for 12) IN ('%s')
    GROUP BY 1,2,3,4,5,6,7,8'''%"','".join(dfparc.par_id.tolist())
    sql = sql0 + sql1

    parcx = gpd.read_postgis(sql,con=con_cad)
    noparcx = dfparc[~dfparc.par_id.isin(parcx.par_id)]

    newparc = parc[~parc.num_parcelle.isin(lastsend.num_parcelle)].copy()
    shp_newparc = shp_parc[~shp_parc.num_parcelle.isin(lastsend.num_parcelle)].copy()
    newparcx = parcx[parcx.par_id.isin(shp_newparc.num_parcelle)].copy()
    shp_newparc = shp_newparc.merge(newparcx[['par_id','update_dat']],how='left',left_on='num_parcelle',right_on='par_id')\
        .drop(columns=['par_id'])\
        .rename(columns={
            'num_parcelle':'num_parc',
            'update_dat':'date_maj_p'
            })\
        .rename_geometry('geometry')


    newparc = complete_tab(newparc)
    parc = complete_tab(parc)
    # mili = get_parc_milli()

    PARC = pd.concat([lastsend,newparc])
    PARC.id_site_cen_parc = PARC.id_site_cen_parc.astype(int)


    # Mise au format des dates
    date_cols = PARC.columns[PARC.columns.str.contains('date')]
    for c in date_cols:
        PARC[c] = pd.to_datetime(PARC[c],yearfirst=True).dt.strftime('%Y/%m/%d')
        # site2023[c] = site2023[c].dt.strftime('%Y/%m/%d')

    


    SHP_PARC = gpd.GeoDataFrame(
        pd.concat([shplastsend,shp_newparc]),
        geometry='geometry',
        crs=2154
    )
    SHP_PARC['date_maj_p'] = pd.to_datetime(SHP_PARC['date_maj_p'],yearfirst=True).dt.strftime('%Y/%m/%d')
    SHP_PARC.geometry = SHP_PARC.buffer(0)

    PARC.to_csv(
        join(
            PATH.format(login=getlogin(),an=annee),
            'Recueil_data',
            'parcelles_cen_38_2023.csv'),
        index=False)
    SHP_PARC.to_file(
        join(
            PATH.format(login=getlogin(),an=annee),
            'Recueil_data',
            'parcelles_cen_38_2023.shp'),
        index=False)