Provide a program to read a .DBF version 3 file and output the results to a .CSV file that can be read by MS-Excel. No library functions can be used other than the standard I/O routines. I can supply sample .DBF files. (pseudo code below)

# Python program to read a .DBF version 3 file

# then output the results to a .CSV file that can be read by MS-Excel.

#

# DBF file format explained at https://en.wikipedia.org/wiki/.dbf

#

# Input: .DBF file

filename = input("Please input the file name you would like to convert: ")

f = open(filename, 'rb') # read binary allows reading byte at a time

# The header is 32 bytes:

# byte 0 = version, bytes 1-3 = YMD

# bytes 4-7 = #records; bytes 8-9 = #headerlength; bytes 10-11 = #recordlength

# ignore bytes 12-31

#

dbase_ver = f.read(1)

last_yy = int.from_bytes(f.read(1), byteorder='little') % 100

last_mm = int.from_bytes(f.read(1), 'little')

last_dd = int.from_bytes(f.read(1), 'little')

numrec = int.from_bytes(f.read(4),'little') # get 4 byte integer as number of records

hdrlen = int.from_bytes(f.read(2),'little') # get header length, used to calc num of fields

reclen = int.from_bytes(f.read(2),'little') # get record length

num_fields = int(hdrlen/32)

f.read(20) # skip reserved section of 32 byte header

# after the 32 byte header above, is an array of "num_fields" field descriptors

# bytes 0-11=fieldname, byte 11=fieldtype, bytes12-15=ignore, byte 16=fieldlength, bytes17-31=ignore

#

# for each field (i.e. num_fields)

# -- save the following information

# read fieldname

# read fieldtype

# read fieldsize

#

# after all the field descriptors, there is one byte with x'0d'

# then the data records begin

# note: each record includes a byte at the beginning, ' '=valid, '*'=deleted

#

# skip the x'0d' byte

#

# for each record (i.e. numrec)

# read byte to determine if valid or deleted

# for each field

# read fieldsize bytes

# -- if not deleted, print this field

#

f.close()