import os
import pandas as pd
import numpy as np

import re

class state_gen_old:
    """"""
    def __init__(self,pattern_start,pattern_stop,null_state="Null"):
        self.previous = null_state
        self.null_state=null_state
        self.pattern_start=pattern_start
        self.pattern_stop=pattern_stop
        
    def __call__(self,actual):
        
        if(not(pd.isnull(actual))):
            
            #Start condition
            if((re.search(self.pattern_start,actual))!=None):
                self.previous=actual
                
            #End Condition
            elif((re.search(self.pattern_stop,actual))!=None):
                self.previous = self.null_state
            
        return self.previous
            
            
class state_gen_new:
    """"""
    def __init__(self,pattern_start,pattern_stop,split = False,null_state="Null"):
        self.previous = null_state
        self.null_state = null_state
        self.pattern_start = pattern_start
        self.pattern_stop = pattern_stop
        self.split = split
        
    def process_state(self,state):
        if (self.split):
            #Keep the root of the state (eg. _Y6014 IntervalStart --> _Y6014)
            state =  state.split()[0]
        return state

    def __call__(self,actual):
        if(not(pd.isnull(actual))):
            #Start condition
            if((re.search(self.pattern_start,actual))!=None):
                self.previous = self.process_state(actual)
                            
            #End Condition
            elif((re.search(self.pattern_stop,actual))!=None):
                self.previous = self.null_state
                            
        return self.previous

path_old= r"C:\Users\jeuux\Desktop\Carrera\MoAI\TFM\AnnotatedData\Accelerometer_Data\Participants\0404b\CompleteData\segments_data.npy"
path_new = r"C:\Users\jeuux\Desktop\Carrera\MoAI\TFM\AnnotatedData\FinalDatasets\Participants\0404b\FullDataset\segments_data.npy"

# # df_old = pd.read_csv(path_old)
# df_old = pd.read_csv(path_new)
# df_new = pd.read_csv(path_new)
targets_old = np.load(path_old)
targets_new = np.load(path_new)

targets_old[0] == targets_new[0]

targets_old == targets_new

##get targets and valid intervals
generator_target_old=state_gen_old("AG","Behaviour End")
df_old["target"]=df_old["Event"].apply(lambda event: generator_target_old(event))

df_old.target.value_counts()

##get targets and valid intervals
generator_target_new=state_gen_new("AG","Behaviour End")
df_new["target"]=df_new["Event"].apply(lambda event: generator_target_new(event))

df_new.target.value_counts() ==df_old.target.value_counts()

df_new[df_new["Recording timestamp"]==0]

df0_path = r"C:\Users\jeuux\Desktop\Carrera\MoAI\TFM\AnnotatedData\Accelerometer_Data\Participants\0705b\CompleteData\CompleteData_raw_0705b.csv"
df1_path = r"C:\Users\jeuux\Desktop\Carrera\MoAI\TFM\AnnotatedData\FinalDatasets\Participants\0705b\FullDataset\FullDataset_raw_0705b.csv"

df0 = pd.read_csv(df0_path)
df1 = pd.read_csv(df1_path)

df0[df0["Recording timestamp"]==0]

df1[df1["Recording timestamp"]==0]