Script that will extract a ProgressDB binary dump file and convert it to a comma separate file. Tested on v8.3

readprogressdumps.py

#!/usr/bin/python

#Example usage
#python readprogressdumps.py -i city.bd -o city.csv -b 3,4,14,16,23 -p 36 -h branch,city-nm,city-no,entry-dt,entry-prsn,rte,state,tax-matl,tax-other,tax-serv,terr-no,tax-cd,area-cd,county-no,zip-cd,bill-to-only,low-zip,high-zip,estimator,salesman,technician,service-rep,service-area,CASS-city-nm,country,township-no,county-nm,township-nm,sales-rep
# -i input file is a binary dump file from Progress Software database
# -o output file
# -b columns that contain binary integers
# -p columns that contain packed decimals
# -h column headers  (first row)
# --debug  (make first row contain column number instead of column header

import struct 
from decimal import Decimal
import csv
import sys,getopt

def unpack(val):
	fld  = struct.unpack('B',val)[0]
	return int(fld)

def getSize(fileobject):
    fileobject.seek(0,2) # move the cursor to the end of the file
    size = fileobject.tell()
    return size

def processfile (inputfile, outputfile, header, int_list, packedlist, show_field_no):
	with open(inputfile, 'rb') as f:
		w = open(outputfile, 'wb')
		writer = csv.writer(w) #,quoting=csv.QUOTE_NONNUMERIC

		print 'Input file is ', inputfile
		print 'Output file is ', outputfile
		
		filesize = getSize(f)
		print 'filesize is %dkb' % filesize

		f.seek(312) #field sepearater
		sep = f.read(1)
		f.seek(320)
		fldcount = unpack(f.read(1))
		print 'field count is %d' % fldcount

		f.seek(636) #jump to first record
		while  f.read(1) != sep  and f.tell() < filesize:
				pass

		
		writer.writerow(header)
		#create header based on field number
		if show_field_no:
			currentfld=1
			header=[]
			while currentfld < fldcount:
				header.append("%d" %(currentfld))
				currentfld+=1
			writer.writerow(header)
		

		lineno = 0		
		#process file
	 	while  f.tell() < filesize:
			lineno +=1
			val = f.read(1) 		
			fld  = struct.unpack('B',val)[0]
			num = int(fld)
			if num == 231:
					val =f.read(1)
					fld1  = struct.unpack('B',val)[0]
					val =f.read(1)
					fld2  = struct.unpack('B',val)[0]
					num = 256 * fld1 + fld2
					f.read(num)
			if num != 1:
				val = f.read(1) 		
				fld  = struct.unpack('B',val)[0]
				num = int(fld)
			val = f.read(num)
			if  val != sep:
				print 'invalid start character found on line number %d ...attempting to recover' % lineno
				f.seek(f.tell()-num-12)
				while  int(struct.unpack('B',f.read(1))[0]) != 231  and f.tell() < filesize:
					pass
				f.seek(f.tell()-1)
				continue

		 	array=[]
		 	currentfld=1
		 	while currentfld < fldcount and f.tell() < filesize:	 		
		 		val = f.read(1) 		
				fld  = struct.unpack('B',val)[0]
				num = int(fld)
				#char string
				if num == 250:
					val =f.read(2)
					charlen  = struct.unpack('>h',val)[0]
					charcount  = struct.unpack('>b',f.read(1))[0]
					if charlen > 0: 
						encvalue = ''
						for x in range(1, charlen):
							val = f.read(1)
							num = struct.unpack('>b',val)[0]
							if num == -1:
								pass
							elif num < 32:
								encvalue += str(num)
							else:
								encvalue += val
						array.append(encvalue)
					else :
						array.append('')
				elif num == 230: 
					val =f.read(2)
					charlen  = struct.unpack('>h',val)[0]
					if charlen > 0:
						array.append(f.read(charlen))
					else :
						array.append('')
				elif num > 0 and num !=253:
					val = f.read(num)

					packed = ord(struct.unpack('c', val[0])[0])
					#binary values
					if  int_list and num == 3 and currentfld in int_list:					
							array.append(struct.unpack('>bh', val)[0])
					elif int_list and num == 2 and currentfld in int_list:					
							array.append(struct.unpack('>h', val)[0])
					elif int_list and num == 1 and currentfld in int_list:					
							array.append(struct.unpack('>b', val)[0])
					#packed decimal-1	
					elif (packed >= 128 and packed <= 137):
						encvalue = ''
						for x in range(1, val.__len__()):
							encvalue += struct.unpack('c', val[x])[0].encode('hex')
						encvalue = Decimal(encvalue.replace('f',''))
						for x in range(128, packed):
							encvalue = encvalue/10
						array.append(encvalue)
					#packed decimal-2 signed
					elif packedlist and currentfld in packedlist:
						encvalue = ''
						for x in range(1, val.__len__()):
							encvalue += struct.unpack('c', val[x])[0].encode('hex')
						encvalue = Decimal(encvalue.replace('f',''))
						for x in range(0, packed):
							encvalue = encvalue/10
						encvalue = encvalue*-1
						array.append(encvalue)
					else:
						encvalue = ''
						for x in range(0, val.__len__()):
							num = struct.unpack('>b',val[x])[0]
							if num == -1:
								pass
							elif num < 32:
								encvalue += str(num)
							else:
								encvalue += val[x]
						array.append(encvalue)
				else:
					array.append('')
				currentfld += 1
			writer.writerow(array)
			
			f.read(12) #stuff at end of line, I don't know what it is for
			
		print 'finished exporting %d rows' % lineno
		print 'done'

		
def main(argv):
	inputfile = None
	header = None
	intlist = None
	outputfile = None
	packedlist = None
	show_field_no = None
	try:
		opts, args = getopt.getopt(argv,"b:p:h:i:o:",["ifile=","ofile=", "binary=", "packed=","debug="])
	except getopt.GetoptError:
		print ' -i <inputfile> -o <outputfile> -b <binary field numbers> -h <header record>'
		sys.exit(2)
	for opt, arg in opts:
		if opt == '-h':
			header = arg.split(',')		
		elif opt in ("--debug"):
			show_field_no = 1
		elif opt in ("-i", "--ifile"):
			inputfile = arg
		elif opt in ("-o", "--ofile"):
			outputfile = arg
		elif opt in ("-b","--binary"):
			intlist = map(int,arg.split(','))
		elif opt in ("-p","--packed"):
			packedlist = map(int,arg.split(','))
	if inputfile == None or outputfile == None :
		print ' -i <inputfile> -o <outputfile> required'
		sys.exit()
	processfile(inputfile,outputfile, header, intlist, packedlist, show_field_no)
	


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