#!/usr/bin/env python3
# -*- coding: UTF-8 -*-.

import geopandas as gpd
from pycen import con
from os import path,listdir
from re import findall

PATH = '/media/colas/SRV/FICHIERS/OUTILS/CARTOGRAPHIE/ESPACE DE TRAVAIL/ETUDES/Zones humides'
lst_layer = [x for x in listdir(PATH) if (x.endswith('.shp') or x.endswith('.gpkg')) and x.startswith('ZHP')]

def upper_first_letter(s):
    if s is None or gpd.pd.isna(s):
        return s
    else:
        return s[0].upper() + s[1:]


def manual_replace(s, old, new, index):
    if s is None or gpd.pd.isna(s):
        return s
    else:
        if old == s[index]:
            return s[:index] + new + s[index +1:]
        else :
            return s


def extract_info(df,value):

    t1 = df[0].str.lower().str.contains(value,na=False)
    t2 = df[1].str.lower().str.contains(value,na=False)
    t3 = df[2].str.lower().str.contains(value,na=False)
    t4 = df[3].str.lower().str.contains(value,na=False)
    t5 = df[4].str.lower().str.contains(value,na=False)
    df[5] = None

    for i,t in enumerate([t1,t2,t3,t4,t5]):
        if t.any():
            df.loc[t,5] = df[t][i]
    
    return df[5].str.strip()


def extract_descripti2(df):
    df['statut'] = extract_info(df,'isit|possible ?|absence zh|non cartographier')
    df['source'] = extract_info(df,'photointe|ortho|scan|information communale')
    return df


if __name__ == "__main__":

    gdf = gpd.GeoDataFrame()
    for layer in lst_layer:
        print(layer)
        data = gpd.read_file(path.join(PATH,layer))
        gdf = gpd.pd.concat([gdf,data])

    gdf.DATEMODIF = gpd.pd.to_datetime(gdf.DATEMODIF,format='mixed')
    gdf.columns = gdf.columns.str.lower()
    gdf.sort_values(['datemodif'],inplace=True,ignore_index=True,ascending=False)
    gdf.drop(columns=['X','Y'],inplace=True,errors='ignore')
    gdf.rename_geometry('geom',inplace=True)
    gdf.drop_duplicates(subset='geom',inplace=True,ignore_index=True,keep='first')

    tmp = (gdf.descripti2
        .str.replace('/','|')
        .str.replace('06|19','06/19')
        .str.replace('04|12|2019','04/12/2019')
        .str.replace('vistier','visiter')
        .str.replace('AbsenceZH','Absence ZH')
        .str.replace('Zone humide non cartographiée et','non cartographier | ZH')
        .str.replace('continuité zh de la gère, à visiter','continuité zh de la gère | à visiter')
        .str.split('\\|| - ',expand=True)
        .replace('',None))
    res = extract_descripti2(tmp)
    gdf[['statut','source']] = res[['statut','source']]
    
    gdf['comment'] = (gdf.descripti2
        .str.replace('vistier','visiter')
        .str.replace('AbsenceZH','Absence ZH')
        .str.replace('Zone humide non cartographiée et','non cartographier | ZH')
        .str.replace('continuité zh de la gère, à visiter','continuité zh de la gère | à visiter')
        .replace([x for x in gdf.statut.unique() if x is not None],'',regex=True)
        .replace([x for x in gdf.source.unique() if x is not None],'',regex=True)
        .str.strip()
        .replace('',None)
        .apply(lambda x: manual_replace(x,'|','',0)).str.strip()
        .apply(lambda x: manual_replace(x,'|','',0)).str.strip()
        .replace('',None)
        .apply(lambda x: manual_replace(x,'/','',0)).str.strip()
        .apply(lambda x: manual_replace(x,'|','',len(x)-1) if x is not None else None)
        .apply(lambda x: manual_replace(x,',','',len(x)-1) if x is not None else None)
        .replace('',None)
        .apply(lambda x: manual_replace(x,'/','',len(x)-1) if x is not None else None)
        .str.strip()
        .str.replace('| | ','|')
        .replace('',None)
        .apply(lambda x: manual_replace(x,'?','',len(x)-1) if x is not None else None)
        .str.strip()
        .apply(lambda x: manual_replace(x,'|','',len(x)-1) if x is not None else None)
        .str.strip()
        .str.replace('Extension zh amont - peupleraie','Peupleraie - extension zh amont')
        .str.replace("Zone d'effondrement - source","Source - zone d'effondrement")
        )
    
    gdf['statut'] = (gdf['statut']
        .apply(lambda x: upper_first_letter(x))
        .str.replace(' ?','?')
        .str.replace('?',' ?')
        .replace('À visiter','A visiter'))
    gdf['source'] = (gdf['source']
        .apply(lambda x: upper_first_letter(x))
        .str.replace('Interprétation','Photointerprétation'))
    gdf['type']  = (gdf.comment
        .str.replace('Prairie humide','Prairiehumide')
        .str.split('\\|| ',n=1)
        .str[0]
        .str.replace('Prairiehumide','Prairie humide')
        .str.strip()
        .replace('Roselières','Roselière')
        .apply(lambda x: upper_first_letter(x))
        )
    gdf['comment']  = (gdf.comment
        .replace([x for x in gdf.type.unique() if x is not None or x.contains('INVZH')],'',regex=True)
        .str.replace('Prairie humide','Prairiehumide')
        .str.split('\\|| ',n=1)
        .str[1]
        .str.replace('Prairiehumide','Prairie humide')
        .str.strip()
        .replace('',None)
        .apply(lambda x: manual_replace(x,'-','',0)).str.strip()
        )
    et = gdf.comment.str.startswith('et ',na=False)
    de = gdf.comment.str.startswith('de ',na=False)
    gdf.loc[et,'comment'] = gdf.loc[et,'comment'].str.replace('et ','',n=1)
    gdf.loc[de,'comment'] = gdf.loc[de,'comment'].str.replace('de ','',n=1)
    
    invzh = gdf['type'].str.contains('INVZH',na=False)
    gdf.loc[invzh,'comment'] = gdf.loc[invzh,'type']
    gdf.loc[invzh,'type'] = None

    gdf.comment = (gdf.comment
        .str.strip()
        .replace('',None)
        .apply(lambda x: upper_first_letter(x)))
    
    gdf.to_postgis('zh_ponctuelles',con,'zones_humides',if_exists='replace',index=True,index_label='gid')

    sql = """
        ALTER TABLE zones_humides.zh_ponctuelles ADD PRIMARY KEY (gid);
        GRANT ALL ON TABLE zones_humides.zh_ponctuelles TO grp_admin;
        GRANT SELECT ON TABLE zones_humides.zh_ponctuelles TO grp_consult;
    """
    with con.begin() as cnx:
        cnx.execute(sql)


    test = """
    SELECT * FROM zones_humides.zh_ponctuelles;
    """
    data = gpd.pd.read_sql_query(
        sql = test,
        con = con)