hpos-web/scripts/merge_csv.py

import glob
import pandas as pd
import openpyxl
import os
from tqdm import tqdm
import platform
from PyQt5.QtWidgets import QApplication, QWidget, QFileDialog

def consolidate_and_perform_calculations(curdir, rootdir, path_delim, validation_file):

    print(os.path.split(curdir)[1])
    
    # list all csv files only  
    csv_files = glob.glob(curdir + path_delim + '/*.{}'.format('csv'))
    if len(csv_files) == 0:
        print("no csv files in the sub folder")
    # print(csv_files)

    df_csv_append = pd.DataFrame()
    
    first = True

    # merge the CSV files
    for file in csv_files:
        if first:
            df_csv_append = pd.read_csv(file)
            colname = file.split('.')[0].split(path_delim)[-1] #file.split('.')[0]
            df_csv_append.rename(columns={'ca': colname}, inplace = True)
            df_csv_append = df_csv_append.drop(['1'], axis=1)
            first = False
        else:
            df = pd.read_csv(file)
            colname = file.split('.')[0].split(path_delim)[-1] #file.split('.')[0]
            df.rename(columns={'ca': colname}, inplace = True)
            df = df.drop(['1'], axis=1)
            df_csv_append = df_csv_append.merge(df, on='tv')
    
    df_csv_append = df_csv_append[df_csv_append['tv'].between(300, 700)]

    wavelength_col = "123_tv" # to make sorting columns simpler

    df_csv_append.rename(columns={'tv': wavelength_col}, inplace = True)

    df_csv_append = df_csv_append.reindex(sorted(df_csv_append.columns), axis=1)

    outfile = rootdir + path_delim + os.path.split(curdir)[1] + "_analysis.xlsx"

    # df_csv_append.to_excel(outfile, sheet_name="merged_data", index=False)
    # df_csv_append.to_csv("D8 Merged.csv", index=False)

    # calculations
    df_427 = df_csv_append.loc[(df_csv_append[wavelength_col] >= 427) & (df_csv_append[wavelength_col] < 428)]

    df_555 = df_csv_append.loc[(df_csv_append[wavelength_col] >= 555) & (df_csv_append[wavelength_col] < 556)]

    df_validation = pd.read_excel(validation_file)

    df_analysis = df_427.iloc[0] + df_555.iloc[0]
    # print(df_analysis)
    # print(min(df_csv_append[0:5]))

    midpoint1 = 427
    midpoint2 = 555
    bandwidth1 = 25
    bandwidth2 = 10

    df = df_analysis.rename(columns = {"NM":"Wavelength","CA":"Absorbance"}, inplace = True)

    # 427 nm range
    df1 = df[ (df['Wavelength'] > (midpoint1-bandwidth1)) & (df['Wavelength'] < (midpoint1+bandwidth1)) ]

    # 555 nm range
    df2 = df[ (df['Wavelength'] > (midpoint2-bandwidth2)) & (df['Wavelength'] < (midpoint2+bandwidth2)) ]
    
    procData.append({"Sample ID": sampleID, 
                     "max_427": round(df1["Absorbance"].max() , 3),
                     "wvmax_427": round(df1.at[df1["Absorbance"].idxmax(),"Wavelength"], 3),
                     "avg_427": round(df1["Absorbance"].mean(), 3),
                     "max_555": round(df2["Absorbance"].max(), 3),
                     "wvmax_555": round(df2.at[df2["Absorbance"].idxmax(),"Wavelength"], 3),
                     "avg_555": round(df2["Absorbance"].mean(), 3),
                     "ratio_max": round(df2["Absorbance"].max()/df1["Absorbance"].max(), 3),
                     "ratio_avg": round(df2["Absorbance"].mean()/df1["Absorbance"].mean(), 3)
                    })

    writer = pd.ExcelWriter(outfile, engine = 'openpyxl')
    df_analysis.to_excel(writer, sheet_name = 'analysis', index=False)
    df_csv_append.to_excel(writer, sheet_name = 'merged_data', index=False)
    writer.close()


sampleID = 1
allDF = pd.DataFrame()
procData = []


if __name__ == "__main__":
    environment = "dev" # dev, QC
    rootdir = os.getcwd()
    validation_file = rootdir + '/validation.xlsx'
    if environment != "dev":
        app = QApplication([])

        # window = QWidget()
        # layout = QVBoxLayout(window)
        rootdir = QFileDialog.getExistingDirectory(None, 'Select main folder')
        choose_validation_file = QFileDialog.getOpenFileName(None, "Select validation excel")
        validation_file = choose_validation_file[0]
    
    path_delim = ''
    if platform.system() == 'Darwin':
        path_delim = '/'
    else:
        path_delim = '\\'
    
    for file in tqdm(os.listdir(rootdir)):
        curdir = os.path.join(rootdir, file)
        if os.path.isdir(curdir):
            # os.chdir(d)
            consolidate_and_perform_calculations(curdir, rootdir, path_delim, validation_file)