source: main/trunk/model-sites-dev/heritage-nz/collect/reports-2019/prepare/xls-to-csv.py@ 33024

import xlrd
import csv
import os

## worksheet_name = "Archaeological reports"


# https://code.activestate.com/recipes/546518-simple-conversion-of-excel-files-into-csv-and-yaml/download/1/
def format_excelval(book, type, value, wanttupledate):
    """ Clean up the incoming excel data """
    ##  Data Type Codes:
    ##  EMPTY   0
    ##  TEXT    1 a Unicode string 
    ##  NUMBER  2 float 
    ##  DATE    3 float 
    ##  BOOLEAN 4 int; 1 means TRUE, 0 means FALSE 
    ##  ERROR   5 
    returnrow = []
    if   type == 2: # TEXT
        if value == int(value): value = int(value)
    elif type == 3: # NUMBER
        datetuple = xlrd.xldate_as_tuple(value, book.datemode)
        value = datetuple if wanttupledate else tupledate_to_isodate(datetuple)
    elif type == 5: # ERROR
        value = xlrd.error_text_from_code[value]
    return value


letter_to_folder = {
  'a': "A - B",
  'b': "A - B",
  'c': "C - D",
  'd': "C - D",
  'e': "E - F",
  'f': "E - F",
  'g': "G - H",
  'h': "G - H",
  'i': "I - J",
  'j': "I - J",
  'k': "K - L",
  'l': "K - L",
  'm': "M - N",
  'n': "M - N",
  'o': "O - P",
  'p': "O - P",
  'q': "Q - R",
  'r': "Q - R",
  's': "S - T",
  't': "S - T",
  'u': "U - V",
  'v': "U - V",
  'w': "W - Z",
  'x': "W - Z",
  'y': "W - Z",
  'z': "W - Z"
}

def id_to_relative_pdf_file(id):
    first_char = id[0:1]
    first_char_lower = first_char.lower()

    file_tail = letter_to_folder[first_char_lower]
    file_root = "Reports " + file_tail

    pdf_file = os.path.join("Digital-Library-PDFs",file_root,id+".pdf")

    return pdf_file


def excel_to_bound_pdf_csv(excel_filename):
    workbook = xlrd.open_workbook(excel_filename, formatting_info=True)
    ## worksheet = workbook.sheet_by_name(worksheet_name)
    worksheet = workbook.sheet_by_index(0)

    excel_filename_split = os.path.split(excel_filename)
    excel_dirname = excel_filename_split[0]
    excel_file = excel_filename_split[1]
    excel_file_splitext = os.path.splitext(excel_file)
    excel_file_root = excel_file_splitext[0];

    pdfbound_filename = os.path.join(excel_dirname,"PDF-BOUND "+excel_file_root+".csv")
    unbound_filename  = os.path.join(excel_dirname,"UNBOUND "+excel_file_root+".csv")

    # print "Worksheet: " + worksheet_name
    csv_pdfbound_ofile = open(pdfbound_filename, 'wb')
    #csv_unbound_ofile  = open('{}.csv'.format(excel_file_root), 'wb')
    csv_unbound_ofile  = open(unbound_filename, 'wb')

    pdfbound_wr = csv.writer(csv_pdfbound_ofile, quoting=csv.QUOTE_ALL)
    unbound_wr  = csv.writer(csv_unbound_ofile, quoting=csv.QUOTE_ALL)

    # 1. Skip lines until Header encountered (as defined by encountering "Report Identifier")
    # 2. Write out Header line as Greenstone friendly Metadata terms 
    #      => remove sp spaces, change "Report Identifier" to hnz.Identifier
    # 3. Process the rest of the file, checking if accompanying 
    #     PDF file present or not

    num_rows = worksheet.nrows

    # 1, Skip lines until Header encountered
    row_i = 0
    found_header = False
    while row_i<num_rows:
        first_cell = worksheet.cell_value(row_i,0);
        if first_cell == "Report Identifier":
            found_header = True
            break
        print "Skipping row {} as not yet encountered 'Report Identifier' metadata label in column 0".format(row_i)
        row_i = row_i + 1

    # 2. Process Header into Greenstone friendly metadata terms     
    if found_header:
        entry_utf8_row = []
        for entry in worksheet.row_values(row_i):
            if entry == "Report Identifier":
                entry = "hnz.Identifier"

            if entry == "Title":
                entry = "dc.Title"

            entry_utf8 = unicode(entry).encode("utf-8")
            metadata_entry_utf8 = entry_utf8.replace(" ", "")

            entry_utf8_row.append(metadata_entry_utf8)
        unbound_wr.writerow(entry_utf8_row)

        # Add in 'Filename' as first column in pdfbound CSV file
        entry_utf8_row.insert(0, "Filename")
        pdfbound_wr.writerow(entry_utf8_row)


        row_i = row_i + 1
        while row_i<num_rows:
            num_cols = worksheet.row_len(row_i)
            formatted_utf8_row = []

            found_pdf = False

            for col_j in range(num_cols):
                cell = worksheet.cell(row_i,col_j)

                cell_type  = worksheet.cell_type(row_i,col_j)
                cell_value = worksheet.cell_value(row_i,col_j)

                formatted_cell = format_excelval(workbook,cell_type,cell_value,False)

                if col_j == 0:
                    # Check to see if companion PDF file present
                    # pdf_filename = os.path.join(excel_dirname,"pdfs",formatted_cell+".pdf")
                    id = formatted_cell.replace(" ","")

                    pdf_file = id_to_relative_pdf_file(id)
                    pdf_file_root, pdf_ext = os.path.splitext(pdf_file)
                    PDF_file = pdf_file_root + ".PDF"

                    pdf_filename = os.path.join(excel_dirname,pdf_file)
                    PDF_filename = os.path.join(excel_dirname,PDF_file)

                    if os.path.exists(pdf_filename):
                        found_pdf = True
                        formatted_utf8_row.insert(0, pdf_file)
                    elif os.path.exists(PDF_filename):
                        found_pdf = True
                        formatted_utf8_row.insert(0, PDF_file)
                    else:
                        print "Unbound id: '" + id + "'"

                if isinstance(formatted_cell, basestring):
                    # Remove any trailing whitespace.  
                    # Newline at end particular harmful for a entry in the CSV file
                    formatted_cell = formatted_cell.rstrip();

                formatted_cell_utf8 = unicode(formatted_cell).encode("utf-8")
                formatted_utf8_row.append(formatted_cell_utf8)
            if found_pdf:
                pdfbound_wr.writerow(formatted_utf8_row)
            else:
                unbound_wr.writerow(formatted_utf8_row)

            row_i = row_i + 1
    else:
        print "Failed to find metadata label 'Report Identifier' in column 0"


    csv_pdfbound_ofile.close()
    csv_unbound_ofile.close()

if __name__ == "__main__":
    excel_to_bound_pdf_csv(os.sys.argv[1])