MarkStreamlit.py

import streamlit as st
import pandas as pd

uploaded_file = st.file_uploader("Choose product file", type="csv")

if uploaded_file:
    #df = pd.read_excel(uploaded_file)
    df = pd.read_csv(uploaded_file, encoding='utf8')
    #st.dataframe(df)

uploaded_file2 = st.file_uploader("Choose inventory file", type="csv")

if uploaded_file2:
    #df2 = pd.read_excel(uploaded_file2)
    df2 = pd.read_csv(uploaded_file2, encoding='utf8')
    
    #st.dataframe(df2)
    
    #st.table(df2)

def ConvertCitrus(df,df2):
    # Load pandas
    import re as re
    import RemoveHTMLtags as RHT
    #INPUT FILE

    #df = pd.read_csv('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/products_export_1 21-10-22.csv', encoding='utf8')


    #df2 = pd.read_csv('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/inventory_export_1 21-10-22.csv', encoding='utf8')
    df.to_excel('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/products_export_1.xlsx',index=False)
    df2.to_excel('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/inventory_export_1.xlsx',index=False)

    tagsp=str('<style type=')+str('"')+str('"')+str('text/css')+str('"')+str('"')+str('><!--')
    tags_list = ['<p class=','"p1"', 'data-mce-fragment="1">,','<b data-mce-fragment="1">','<i data-mce-fragment="1">','<p>' ,'</p>' , '<p*>',
                  '<ul>','</ul>',
                  '</i>','</b>','</p>','</br>',
                  '<li>','</li>',
                  '<br>',
                  '<strong>','</strong>',
                  '<span*>','</span>', '"utf-8"','UTF-8',
                  '<a href*>','</a>','<meta charset=utf-8>',';;',
                  '<em>','</em>','"','<meta charset=','utf-8>','<p>','<p','data-mce-fragment=1',';','<style type=','<style type=','><!--','text/css','<style type=\"\"text/css\"\"><!--','--></style>','td {border: 1px solid #ccc','}br {mso-data-placement:same-cell','}','>']




    def remove_html_tags(text):
        """Remove html tags from a string"""
        import re
        clean = re.compile('<.*?>')
        return re.sub(clean, '', text)
    #for tag in tags_list:
    ##      df['overview_copy'] = df['overview_copy'].str.replace(tag, '')
    #   df.replace(to_replace=tag, value='', regex=True, inplace=True)

    for index, row in df.iterrows():
        df.iloc[index,2]=RHT.remove_tags(str(df.iloc[index,2]))

    print(df.iloc[:,2])

    df.to_excel('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/products_export_1-nohtml.xlsx')

    #df.fillna('', inplace=True)
    df.iloc[:,2] = pd.Series(df.iloc[:,2],dtype="string")
    print(df.iloc[:,2].dtype)
    #s = pd.Series(['a', 'b', 'c'], dtype="string")
    #s.dtype

    #CONVERT FORMATS

    #Column A(0) – Ignore
    #Column B(1) “Title” > Column B(1) “Product Name”
    #Column C(2) – Ignore
    #Column D(3) “Vendor” > Column K(10) “Brand”
    #Column F(5) “Custom Product Type” > Column AF(31) “Short Description”
    #Column J(9) “Option1 Value” > Column I(8) “Size 1”
    #Column L(11) “Option2 Value” > Column H(7) > Colour
    #Column M(12) - Ignore
    #Column N(13) “Option 3 Value” > Column A(0) “Style Number”
    #1.	Problems in Column N.  Some codes do not stay as a number when the Citrus Lime csv is re-opened (8.05652E+12 instead of 8056516179091) The saved csv keeps turning this column back to “general’ format column when I re-open it, even after I save it as number column. The upload must keep this as a number formatted column.
 
    #Column O(14) - Ignore
    #Column P(15) “Variant Grams” > Column AE (30) “Weight (grams)”
    #Column R(17) “Variant Inventory  Qty” > Column R (17) “Stock Count”. THIS IS THE KEY TO THE DAILY UPLOAD
    #Column U(20) “Variant Price” > Column F (5) “Unit MSRP”

    #Column Y > C&D
    #################################################################################################
    temp_cols=df.columns.tolist()
    new_cols=temp_cols.copy()
    new_cols[1]=temp_cols[1]

    new_cols[17]=temp_cols[17]

    #################################################################################################
    #THERE IS NO EXISTING COLUMN ON THE SHOPIFY EXPORT TO DIRECTLY PROVIDE DATA FOR COLUMN E ON THE CITRUS LIME CSV (which is the wholesale price ex VAT to the retailer). However – Column U “ Variant Price” can provide the information for Column E with the following formula:
 
    #((Column U/1.2)/1.6)*0.96
 
    #Column Y “Variant Barcode” > Column C “Vendor SKU” (2) (and D "UPC/EAN" (3)??)
 
    #There are 2 problems with converting Column Y to Column C.  
    #2.	Shopify exports the UPC data and adds an apostrophe. This fails the SIM process. We need to get data without the apostrophe.
    #3.	Vendor SKU.  The CSV file keeps switching the data to a non-number eg 8056516178308 shows as 8.05652E+12. The saved csv keeps turning this column to “general’ format column when I re-open it, even after I save it as number column. The upload must keep this as a number formatted column.
 
    #This is where it gets complicated…
 
    #Shopify exports the image file as https:// links in an odd way. Instead of attributing image 1, image 2, and image 3 etc in dedicated and separate columns, it spreads them across the sizes for the related product in the same column (Column Z “Image Src”). Column AA in the Shopify export csv just shows the image position instead.  We need to find a solution. We need to be able to provide https// image links in separate columns for each product and size. For example, if a product has 3 images, these need to be converted into Citrus Lime CSV columns Column Z “Image 1”, Column AA “Image 2”, Column AB “Image 3”, Column AC “Image 4” etc.
    #new_cols[4]=((temp_cols[20]/1.2)/1.96)*0.96

    #Column C “Body (HTML)” > Column AG “Long Description” (32)


    df_copy=df[new_cols].copy(deep=True)
    print("SKU")
    print(df.iloc[:,24])

    local_df = df.copy(deep=True)

    df_copy.iloc[:,0]=local_df.iloc[:,13].copy(deep=True)
    df_copy.iloc[:,5]=local_df.iloc[:,20].copy(deep=True)
    df_copy.iloc[:,7]=local_df.iloc[:,11].copy(deep=True)
    #24 is variant Bar code
    df_copy.iloc[:,2]=local_df.iloc[:,24].copy(deep=True)

    df_copy.iloc[:,8]=local_df.iloc[:,9].copy(deep=True)
    df_copy.iloc[:,10]=local_df.iloc[:,3].copy(deep=True)
    df_copy.rename(columns={df_copy.columns[10]: 'Brand'},inplace=True)
    df_copy.columns.values[10] = 'Brand'

    df_copy.iloc[:,30]=local_df.iloc[:,15].copy(deep=True)
    df_copy.iloc[:,31]=local_df.iloc[:,5].copy(deep=True)
    df_copy.iloc[:,32]=local_df.iloc[:,2].copy(deep=True)

    df_copy.rename(columns={df_copy.columns[8]: 'Size 1'},inplace=True)

    print(list(df_copy.columns.values))

    #WE CONVERT COLUMN 20 to numeric (in case it's read as string)
    df_copy.iloc[:,20] = df_copy.iloc[:,20].astype(float)

    df_copy.iloc[:,4]=(((df_copy.iloc[:,20]/1.2)/1.96)*0.96)
    from babel.numbers import format_currency
    df_copy.iloc[:,4] = df_copy.iloc[:,4].apply(lambda x: format_currency(x, currency="GBP", locale="en_GB"))
    df_copy.iloc[:,5] = df_copy.iloc[:,5].apply(lambda x: format_currency(x, currency="GBP", locale="en_GB"))

    print(((df_copy.iloc[:,20]/1.2)/1.96)*0.96)
    #df_copy.iloc[:,2]=df_copy.iloc[:,2].str.replace("'","")
    df_copy.iloc[:,2] = df_copy.iloc[:,2].astype(str).str.replace("'","")


    #df_copy.iloc[:,24]=df_copy.iloc[:,24].str.replace("'","")
    df_copy.iloc[:,24] = df_copy.iloc[:,24].astype(str).str.replace("'","")

    print("SKU")
    print(df_copy.iloc[:,2])






    #rename specific column names

    #df_copy.rename(columns = {'Variant Inventory Qty':'Stock Count','Variant Grams' : 'Weight (grams)'}, inplace = True)

    #df_copy.rename(columns = {'Option2 Value':'Colour','Option1 Value' : 'Size 1'}, inplace = True)

    #df_copy.rename(columns = {'Vendor':'Brand','Title' : 'Product Name'}, inplace = True)
    #df_copy.rename(columns = {'Body (HTML)':'Long Description'}, inplace = True)

    #df_copy.rename(columns={df_copy.columns[4]: 'Unit Cost'},inplace=True)


    print(list(df_copy.columns.values))


    #df_copy.rename(columns={df_copy.columns[31]: 'Short Description'},inplace=True)
    #df_copy.rename(columns={df_copy.columns[2]: 'Vendor SKU'},inplace=True)
    df_copy.rename(columns={df_copy.columns[6]: 'Colour Code (Simple Colour)'},inplace=True)
    ##IN COLUMN H (6), WE HAVE SOME TAGS AND WE WANT TO GET THE TAG "MEN, WOMEN, LADY OR BOTH (UNISEX)"
    #WE ARE GETTING THAT INFO BEFORE REMOVING DATA FROM 6
    for index, row in df_copy.iterrows():
        if index==0:
            print(row['Colour Code (Simple Colour)'])
        if " mens" in str(row['Colour Code (Simple Colour)']):
            if " womens" in str(row['Colour Code (Simple Colour)']):
                df_copy.iloc[index,12]="Unisex"
            else:
                df_copy.iloc[index,12]="Mens"
            
        if " womens" in str(row['Colour Code (Simple Colour)']):
            if " mens" in str(row['Colour Code (Simple Colour)']):
                df_copy.iloc[index,12]="Unisex"
            else:
                df_copy.iloc[index,12]="Womens"
        if " ladys" in str(row['Colour Code (Simple Colour)']):
                df_copy.iloc[index,12]="Ladys"
        if index==0:
            print(row[12])
    print(df_copy.iloc[:,12])



    df_copy.iloc[:,6] = ""
    #Style Number	Product Name	Vendor SKU	UPC/EAN	Unit Cost	Unit MSRP 	Colour Code (Simple Colour)	Colour	
    df_copy.rename(columns={df_copy.columns[0]: 'Style Number'},inplace=True)
    df_copy.rename(columns={df_copy.columns[1]: 'Product Name'},inplace=True)
    df_copy.rename(columns={df_copy.columns[2]: 'Vendor SKU'},inplace=True)
    df_copy.rename(columns={df_copy.columns[3]: 'UPC/EAN'},inplace=True)
    df_copy.rename(columns={df_copy.columns[4]: 'Unit Cost'},inplace=True)
    df_copy.rename(columns={df_copy.columns[5]: 'Unit MSRP'},inplace=True)
    df_copy.rename(columns={df_copy.columns[6]: 'Colour Code (Simple Colour)'},inplace=True)
    print(df_copy.columns[6])
    df_copy.rename(columns={df_copy.columns[7]: 'Colour'},inplace=True)
    #Size 1	Size 2	Brand 	Year or Season	Gender	Manufacturer Part Code	Other Barcode	VAT	Pack Qty	
    df_copy.rename(columns={df_copy.columns[8]: 'Size 1'},inplace=True)
    df_copy.rename(columns={df_copy.columns[9]: 'Size 2'},inplace=True)
    df_copy.rename(columns={df_copy.columns[10]: 'Brand'},inplace=True)
    df_copy.rename(columns={df_copy.columns[11]: 'Year of Season'},inplace=True)
    df_copy.rename(columns={df_copy.columns[12]: 'Gender'},inplace=True)
    df_copy.rename(columns={df_copy.columns[13]: 'Manufacturer Part Code'},inplace=True)
    df_copy.rename(columns={df_copy.columns[14]: 'Other Bar Code'},inplace=True)
    df_copy.rename(columns={df_copy.columns[15]: 'VAT'},inplace=True)
    df_copy.rename(columns={df_copy.columns[16]: 'Pack Qty'},inplace=True)
    #Stock Count	Price Band 1	Price Band 2 	IE VAT	Unit Cost in Euros	MSRP in Euros	
    df_copy.rename(columns={df_copy.columns[17]: 'Stock Count'},inplace=True)
    df_copy.rename(columns={df_copy.columns[18]: 'Price Band 1'},inplace=True)
    df_copy.rename(columns={df_copy.columns[19]: 'Price Band 2'},inplace=True)
    df_copy.rename(columns={df_copy.columns[20]: 'IE VAT'},inplace=True)
    df_copy.rename(columns={df_copy.columns[21]: 'Unit Cost in Euros'},inplace=True)
    df_copy.rename(columns={df_copy.columns[22]: 'MSRP in Euros'},inplace=True)
    #Commodity Codes	Country of Origin	Image (multiple images can be added in separate columns if available)	
    df_copy.rename(columns={df_copy.columns[23]: 'Commodity Codes'},inplace=True)
    df_copy.rename(columns={df_copy.columns[24]: 'Country of Origin'},inplace=True)
    #Weight	Short Description	Long Description	Video Link		
    df_copy.rename(columns={df_copy.columns[30]: 'Weight'},inplace=True)
    df_copy.rename(columns={df_copy.columns[31]: 'Short Description'},inplace=True)
    df_copy.rename(columns={df_copy.columns[32]: 'Long Description'},inplace=True)
    df_copy.rename(columns={df_copy.columns[33]: 'Video Link'},inplace=True)







    df_copy.iloc[:,9] = ""

    df_copy.iloc[:,13] = ""

    df_copy.iloc[:,14] = ""

    df_copy.iloc[:,16] = ""

    df_copy.iloc[:,18] = ""

    df_copy.iloc[:,19] = ""

    df_copy.iloc[:,20] = ""

    df_copy.iloc[:,21] = ""

    df_copy.iloc[:,22] = ""
    #df_copy.rename(columns={df_copy.columns[26]: 'Weight (Grams)'},inplace=True)

    #df_copy.iloc[:,26] = ""

    df_copy.iloc[:,33] = ""



    #df_copy.iloc[:,5] = " "
    df_copy.iloc[:,15] = "20"

    print(list(df_copy.columns.values))

    #Column Y in the export and this code should go into both Columns C and D in the conversion with the titles “Vendor SKU” and  “UPC/EAN” It is replicated for a complicated reason that I won’t explain here, but Column Y in the export should go into both Column C and D in the conversion
    df_copy.iloc[:,3] = df_copy.iloc[:,2]
    df_copy.columns.values[10] = 'Brand'
    df_copy.iloc[:,11] = ""
    df_copy.iloc[:,22] = ""
    #df_copy.rename(columns={df_copy.columns[30]: 'Weight (Grams)'},inplace=True)


    print("SKU")
    print(df_copy.iloc[:,2])


    #DATA COMING FROM THE OTHER CSV FILE

    df_copy.iloc[:,23] = ""


    df_copy.iloc[:,24] = ""

    #WARNING: HEADER IS IN SECOND ROW. WE DONT HAVE INTO ACCOUNT FIRST ROW
    #df2 = pd.read_excel('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/inventory_export_12.xlsx',engine="openpyxl", header=1)


    #WE HAVE TO REORDER COLUMNS COO and HS Code in df2 in order to match the index order of df 
    #list1=df_copy.set_index('Vendor SKU').T.to_dict('list')
    #print(list1)
    new_index=df['Variant SKU']
    boolean = df['Variant SKU'].duplicated().any()
    #print(boolean)
    boolean = df2['SKU'].duplicated().any()
    #print(boolean)
    duplicateRows2 = df2[df2.duplicated(['SKU'],keep = False)]
    #print(duplicateRows2['SKU'])

    duplicateRows = df[df.duplicated(['Variant SKU'],keep = False)]
    #print(duplicateRows)
    #print(duplicateRows['Variant SKU'])
    #print(new_index)
    df2=df2.set_index('SKU')
    #print(df2)
    #i=df2.index
    #for x in i:
    #    print(x)
    df2.reindex(new_index)
    #i=df2.index
    #for x in i:
    #    print(x)
    #print(df2)
    #print(df2.index)
    #df3 = pd.DataFrame(students, index=['a', 'b', 'c', 'd', 'e'])
    #print("Original DataFrame: ")
    #print(df)








    print("TERMINE")

    df_copy.iloc[:,24] = df2.loc[:,'COO']
    df_copy.iloc[:,23] = df2.loc[:,'HS Code']

    df_copy['Commodity Codes']=df2['HS Code'].values
    df_copy['Country of Origin']=df2['COO'].values


    #print(df2.loc[:,'COO'])
    #print(df2.loc[:,'HS Code'])
    #print(df_copy.iloc[:,24])
    #print(df_copy.iloc[:,23])
    print("SKU")
    print(df_copy.iloc[:,2])



    #WE COMPLETE THE DATAFRMAE WITH DUMMY COLUMNS TILL THE MAXIMUM DESIRED NUMBER
    header_list=[]
    for i in range(49,58):
        #df.insert(i, "Dummy", [], True)
        header_list.append(str(i))
        df_copy[str(i)]=''



    column_indices=[]
    for i in range(0,24):
        column_indices.append(34+i)

    #Tech Specs	Size Chart	Geometry Chart	Frame	Rear Shock	Fork
    #Headset	Stem	Handlebar	Bar Tape / Grip	Brakes Levers	Brake Calipers	Tyres	Wheels	Front Derailleur	
    #Rear Derailleur	Shift Levers	Chain	Cassette	Chainset	Bottom Bracket	Pedals	Saddle	Seatpost

    old_names = df_copy.columns[column_indices]
    new_names = ['Tech Specs','Size Chart','Geometry Chart','Frame',	'Rear Shock',	'Fork',	'Headset',	'Stem',	'Handlebar',	'Bar Tape / Grip',	'Brakes Levers',	'Brake Calipers',	'Tyres',	'Wheels',	'Front Derailleur',	'Rear Derailleur',	'Shift Levers'	,'Chain'	,'Cassette'	,'Chainset'	,'Bottom Bracket',	'Pedals',	'Saddle',	'Seatpost']
    old_names = df_copy.columns[column_indices]
    df_copy.rename(columns=dict(zip(old_names, new_names)), inplace=True)


    df_copy.iloc[:,34:58]=''


    print("SKUf")
    print(df_copy.iloc[:,2])
    #print(df_copy.iloc[:,3])

    ## Rename all columns with list
    #cols = ['Courses','Courses_Fee','Courses_Duration']
    #df_copy.columns = cols
    #print(df.columns)


    ###################
    #PUT IMAGES IN A SIGNLE ROW: WE LOOK FOR IMAGES COMING FROM COMMON NAMES
    #Shopify exports the image file as https:// links in an odd way. Instead of attributing image 1, image 2, and image 3 etc in dedicated 
    #and separate columns, it spreads them across the sizes for the related product in the same column (Column Z “Image Src”). 
    #Column AA in the Shopify export csv just shows the image position instead.  We need to find a solution. 
    #We need to be able to provide https// image links in separate columns for each product and size. For example, if a product has 3 images, 
    #these need to be converted into Citrus Lime CSV columns Column Z “Image 1”, Column AA “Image 2”, Column AB “Image 3”, Column AC “Image 4” 
    #etc
    ####################
    #region imagesRow2Column
    #We get the list of rows with NAN data in Product Name column (same product name but different sizes (XS, XL...). Each of these rows has a image scr link
    list_col=df_copy.loc[pd.isna(df_copy.loc[:,'Product Name']), :].index
    images=df_copy.loc[list_col,'Image Src']
    list_end=[]
    for row in df_copy.index:
        #NotNA gets rows where Product Name column has a name in it (first image and row where we should add the images)
        if pd.notna(df_copy.loc[row,'Product Name']):
            #print(df_copy.loc[row,'Product Name'])
            rowNotNa=row
            i=1
            #j=1
            list_img=[]  
            #WE INCLUDE IN THE LIST THE FIRST IMAGE
            list_img.append(df_copy.loc[row,'Image Src'])
            while pd.isna(df_copy.loc[row+i,'Product Name']) and row+i<len(df_copy.index)-1:
                #WE ADD THE REST OF THE IMAGES (FOLLOWING ROWS)
                if "http" in str(df_copy.loc[row+i,'Image Src']):
                    list_img.append(df_copy.loc[row+i,'Image Src'])      
                i=i+1
            list_end.append(list_img)

    #IN list_end WE HAVE ALL OF THE IMAGES FOR EACH PRODUCT NAME
    index_nonnan=df_copy.loc[pd.notna(df_copy.loc[:,'Product Name']), :].index
    max=0
    for i in range(len(list_end)):
        if max<len(list_end[i]):
            max=len(list_end[i])
    print("SKUf")
    print(df_copy.iloc[:,2])

    #WE CHANGE THE COLUMN NAME OF THE COLUMNS WHERE THERE ARE IMAGES: EACH COLUMN IS CALLED "Image x"
    #We first delete old values in the Image columns
    for j in range(max):
            df_copy.iloc[:,25+j]=''

    counter=0
    for index in index_nonnan:   
        for j in range(len(list_end[counter])):
       
        
            if list_end[counter][j]!='nan':
                df_copy.iloc[index,25+j]=list_end[counter][j]
                df_copy.rename(columns={df_copy.columns[25+j]: 'Image'+str(j+1)},inplace=True)

        counter=counter+1
    print("SKUf")
    print(df_copy.iloc[:,2])
    #WE HAVE TO FILL NAN ROWS (SAME PRODUCT BUT DIFFERENT SIZES) WITH THE SAME IMAGES THAT IN NON NAN ROWS (MAIN PRODUCT-SIZE)
    listImages=[None] * max
    list1=[None] * max
    list2=[None] * max
    list3=[None] * max
    list4=[None] * max
    list5=[None] * max
    for index, row in df_copy.iterrows():
        #NotNA gets rows where Product Name column has a name in it (first image and row where we should add the images)
        #print(df_copy.iloc[index,1])
        if pd.notna(df_copy.iloc[index,1]):
            for j in range(0,max):
                listImages[j]=str((df_copy.iloc[index,25+j]))
                #list1[j]=str((df_copy.iloc[index,1+j]))
                #list2[j]=str((df_copy.iloc[index,10+j]))
                #list3[j]=str((df_copy.iloc[index,12+j]))
                #list4[j]=str((df_copy.iloc[index,31+j]))
                #list5[j]=str((df_copy.iloc[index,32+j]))
                list1[j]=str((df_copy.iloc[index,1]))
                list2[j]=str((df_copy.iloc[index,10]))
                list3[j]=str((df_copy.iloc[index,12]))
                list4[j]=str((df_copy.iloc[index,31]))
                list5[j]=str((df_copy.iloc[index,32]))
            
        else:
            for j in range(0,max):
                df_copy.iloc[index,25+j]=listImages[j]
                #df_copy.iloc[index,1+j]=list1[j]
                #df_copy.iloc[index,10+j]=list2[j]
                #df_copy.iloc[index,12+j]=list3[j]
                #df_copy.iloc[index,31+j]=list4[j]
                #df_copy.iloc[index,32+j]=list5[j]
                df_copy.iloc[index,1]=list1[j]
                df_copy.iloc[index,10]=list2[j]
                df_copy.iloc[index,12]=list3[j]
                df_copy.iloc[index,31]=list4[j]
                df_copy.iloc[index,32]=list5[j]

    #endregion

    print("SKUf")
    print(df_copy.iloc[:,2])
    #print(df_copy.iloc[:,3])

    ###################################################################################
    df_copy.to_excel('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/OCCHIO-Cycle-Data-File_st.xlsx',index=False)



    #df_copy.to_csv('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/OCCHIO-Cycle-Data-File.csv',index=False, encoding='utf-8')
    df_copy.to_csv('C:/Users/15572890/Desktop/I+D/MarksCsvConversion/Validation2/OCCHIO-Cycle-Data-File_st.csv',index=False, encoding='utf_8_sig')


if uploaded_file and uploaded_file2:
    ConvertCitrus(df,df2)