from etl.src import data
from etl.src.data import etl_pipelines as ep
import polars as pl
from datetime import datetime, timedelta
import pandas as pd


'''
# Get Data from DB
inst = data.load()

df = inst.extractions().pl()
df = ep.expansion_Pipeline(df)

df.write_csv('dok/flatDates.csv')
print(df)
'''
'''
#Load Data from DF 
dfLive = pl.read_csv('dok/liveDates.csv')
dfFlat = pl.read_csv('dok/flatDates.csv')


# Step 1 Get all occupied dates in live data
dfLive = dfLive.filter(pl.col("calendar_value") == 0)
dfLive = dfLive.with_columns(pl.col("created_at").str.to_date("%Y-%m-%d"))
dfLive = dfLive.with_columns(pl.col("calendar_date").str.to_date("%Y-%m-%d"))
#print(dfLive)

dfFlat = dfFlat.with_columns(pl.col("created_at").str.to_date("%Y-%m-%d"))
dfFlat = dfFlat.with_columns(pl.col("calendar_date").str.to_date("%Y-%m-%d"))

propIds = dfLive.get_column('property_id').unique()
createdAt = dfLive.get_column('created_at').unique()
#print(createdAt)

fullPreorderMatrix = []

for propId in propIds:
    curPreorderList = []
    print("Property ID = " + str(propId))
    tempPropFlatDf = dfFlat.filter(pl.col("property_id") == propId)
    tempPropLiveDf = dfLive.filter(pl.col("property_id") == propId)
    allLiveOccupiedDates = tempPropLiveDf.filter(pl.col("calendar_value") == 0).get_column('created_at')
    #print("allLiveOccupiedDates = ",allLiveOccupiedDates)
    for date in allLiveOccupiedDates:
        calLiveDate = tempPropLiveDf.filter(pl.col("created_at") == date).get_column('calendar_date')[0]
        #print("Occupied Date = " + str(date), "with Calendar Date =", str(calLiveDate))
        numOfScrapedPreordered = 0
        foundLastDate = False
        for createDate in createdAt:
            if date > createDate:
                #print("Finding Flat Date with CreateDate =",createDate, "and Calendar Date =", calLiveDate)
                tempFlatDf = tempPropFlatDf.filter(pl.col("created_at") == createDate)
                tempFlatDf = tempFlatDf.filter(pl.col("calendar_date") == calLiveDate)
                #print("tempLiveDf = ", tempFlatDf)

                calVal = tempFlatDf.get_column('calendar_value')
                if len(calVal) > 0:
                    if calVal[0] == 0:
                        # Still Occupied
                        if not foundLastDate:
                            numOfScrapedPreordered += 1
                    else:
                        # Found last Date where not occupied
                        foundLastDate = True
                        #print("number of Scrapes already occupied =", numOfScrapedPreordered)
                        #break
            #else:
                #print("Skipped: Live Date = ",date, "Flat Date =",createDate)
        #print(propId, date, numOfScrapedPreordered)
        curPreorderList.append(numOfScrapedPreordered)
    if len(curPreorderList) > 0:
        mean = sum(curPreorderList) / len(curPreorderList)
    else: mean = 0
    #fullPreorderMatrix.append([propId, mean, curPreorderList])
    fullPreorderMatrix.append([propId, mean])

print(fullPreorderMatrix)
fullPreoDF = pl.DataFrame(fullPreorderMatrix,orient="row")
fullPreoDF.write_csv('dok/fullPreoDF.csv')
print(fullPreoDF)
'''

# Filter Props to locations and calculate Means per location
inst = data.load()

propDf = inst.propIds_with_region().pl()
print(propDf)
propDf = propDf.select(
    pl.col("id").cast(pl.Int64),
    pl.col("seed_id").cast(pl.Int64),
)

preoDF = pl.read_csv('dok/fullPreoDF.csv')
preoDF = preoDF.rename({"column_0": "id", "column_1": "meanPreorderScrapeNum"})


merge = preoDF.join(propDf, how='inner', on='id')
print(merge)

print("Global meanPreorderTime = ",round(merge.get_column("meanPreorderScrapeNum").mean()*3,2))

# 1 = Heidiland
heidi = merge.filter(pl.col("seed_id") == 1)
print("Heidiland meanPreorderTime = ", round(heidi.get_column("meanPreorderScrapeNum").mean()*3,2))
# 2 = Davos
Davos = merge.filter(pl.col("seed_id") == 2)
print("Davos meanPreorderTime = ", (Davos.get_column("meanPreorderScrapeNum").mean()*3,2))
# 3 = Engadin
Engadin = merge.filter(pl.col("seed_id") == 3)
print("Engadin meanPreorderTime = ", round(Engadin.get_column("meanPreorderScrapeNum").mean()*3,2))
# 4 = St. Moritz
Moritz = merge.filter(pl.col("seed_id") == 4)
print("St. Moritz meanPreorderTime = ", round(Moritz.get_column("meanPreorderScrapeNum").mean()*3,2))