What's the easiest way to read a FoxPro DBF file from Python?

Question

I've got a bunch of FoxPro (VFP9) DBF files on my Ubuntu system, is there a library to open these in Python? I only need to read them, and would preferably have access to the memo fields too.

Update: Thanks @cnu, I used Yusdi Santoso's dbf.py and it works nicely. One gotcha: The memo file name extension must be lower case, i.e. .fpt, not .FPT which was how the filename came over from Windows.

Brunson

Answer 1

You can try this recipe on Active State.

There is also a DBFReader module which you can try.

For support for memo fields.

Allocution

The DBFReader module is now a dead link. – Illaudable

In memo fields program its give the number not exact data how to retrieve the data based on that index number where that stored . – Marji

Since Sep 2008 (time of the answer), there are many more options (Nov 2015), for example: pypi.python.org/pypi/dbf, pypi.python.org/pypi/dbfread, sourceforge.net/projects/dbfpy. I personally chose dbfread because it is well documented. – Maniac

Answer 2

I prefer dbfpy. It supports both reading and writing of .DBF files and can cope with most variations of the format. It's the only implementation I have found that could both read and write the legacy DBF files of some older systems I have worked with.

Dric

Answer 3

I was able to read a DBF file (with associated BAK, CDX, FBT, TBK files**) using the dbf package from PyPI http://pypi.python.org/pypi/dbf . I am new to python and know nothing about DBF files, but it worked easily to read a DBF file from my girlfriend's business (created with a music store POS application called AIMsi).

After installing the dbf package (I used aptitude and installed dbf version 0.88 I think), the following python code worked:

from dbf import *
test = Table("testfile.dbf")
for record in test:
    print record
    x = raw_input("")  # to pause between showing records

That's all I know for now, but hopefully it's a useful start for someone else who finds this question!

April 21, 2012 SJK Edit: Per Ethan Furman's comment, I should point out that I actually don't know which of the data files were necessary, besides the DBF file. The first time I ran the script, with only the DBF available, it complained of a missing support file. So, I just copied over the BAK, CDX, FPT (not FBT as I said before edit), TBK files and then it worked.

Platter

This package doesn't yet support IDX/CDX files, although I hope to have that in place this year. – Caddell

Are you sure it was FBT and not FPT? FPT files are the memo files. CDX is the index file although @Ethan says these aren't supported so that was unlikely to be the missing file. Along with the DBF these are the only 3 files you should need to read a DBF/Table. – Illaudable

This package is capable but documentation is not newbie friendly, dbfread provides more examples. – Maniac

You might need to add test.open() before the for loop for this to work – Tercet

@Steve are you sure it works with FPT files too? Or anyone else? I am trying to read FPT files but I am new to DBF and do not know what to do now. HELP please – Meemeece

Answer 4

If you're still checking this, I have a GPL FoxPro-to-PostgreSQL converter at https://github.com/kstrauser/pgdbf . We use it to routinely copy our tables into PostgreSQL for fast reporting.

Billat

I coincidentally saw this while checking for something else. Now a postgreSQL lover, coming from VFP and MS SQL, this is much appreciated. Thanks! – Skepticism

Answer 5

Check out http://groups.google.com/group/python-dbase

It currently supports dBase III and Visual Foxpro 6.0 db files... not sure if the file layout change in VFP 9 or not...

Prudence

Answer 6

It's 2016 now and I had to fiddle with the dbf package to get it to work. Here is a python3 version to just export a dbf file to a csv

import dbf

d=dbf.Table('mydbf.dbf')
d.open()
dbf.export(d, filename='mydf_exported.csv', format='csv', header=True)

I had some unicode error at first, but got around that by turning off memos.

import dbf

d=dbf.Table('mydbf.dbf', ignore_memos=True)
d.open()
dbf.export(d, filename='mydf_exported.csv', format='csv', header=True)

Rheotaxis