DBF

DBF is a small, fast Ruby library for reading dBase, xBase, Clipper, and FoxPro database files.

• Project page: https://github.com/infused/dbf
• API Documentation: https://dbf.infused.org
• Report bugs: https://github.com/infused/dbf/issues
• Questions: Email mailto:keithm@infused.org and put DBF somewhere in the subject line
• Change log: https://github.com/infused/dbf/blob/main/CHANGELOG.md

Compatibility

DBF is tested to work with the following versions of Ruby:

• Ruby 3.3.x, 3.4.x, 4.0.x

Older Rubies are supported by older DBF release lines:

DBF version	Ruby support
5.2+	3.3+
4.3 – 5.1	3.1+
4.0 – 4.2	2.4+
3.x (3_stable)	2.0 – 2.3
2.x (2_stable)	1.8 – 1.9

DBF is Ractor-friendly: all library constants are deeply frozen, so tables can be opened, enumerated, and exported inside non-main Ractors.

Installation

Install the gem manually:

gem install dbf

Or add to your Gemfile:

gem 'dbf'

Basic Usage

Open a DBF file using a path:

require 'dbf'
widgets = DBF::Table.new("widgets.dbf")

Open a DBF file using an IO object:

data = File.open('widgets.dbf', 'rb')
widgets = DBF::Table.new(data)

All errors raised by the library inherit from DBF::Error, so you can rescue the library as a unit:

begin
  widgets = DBF::Table.new("widgets.dbf")
rescue DBF::Error => e
  puts "Unable to read DBF file: #{e.message}"
end

Open a DBF by passing in raw data (wrap the raw data with StringIO):

widgets = DBF::Table.new(StringIO.new('raw binary data'))

Use the block form to close the table automatically, like File.open:

DBF::Table.open('widgets.dbf') do |table|
  table.each { |record| puts record.name }
end

Enumerate all records

widgets.each do |record|
  puts record.name
  puts record.email
end

Find a single record

widget = widgets.find(6)

Note that find() will return nil if the requested record has been deleted and not yet pruned from the database.

The value for an attribute can be accessed via element reference in several ways.

widget.slot_number     # underscored field name as method

widget["SlotNumber"]   # original field name in dbf file
widget['slot_number']  # underscored field name string
widget[:slot_number]   # underscored field name symbol

Get a hash of all attributes. The keys are the original column names.

widget.attributes
# => {"Name" => "Thing1 | SlotNumber" => 1}

Search for records using a simple hash format. Multiple search criteria are ANDed. Use the block form if the resulting record set is too big. Otherwise, all records are loaded into memory.

# find all records with slot_number equal to s42
widgets.find(:all, slot_number: 's42') do |widget|
  # the record will be nil if deleted, but not yet pruned from the database
  if widget
    puts widget.serial_number
  end
end

# find the first record with slot_number equal to s42
widgets.find :first, slot_number: 's42'

# find record number 10
widgets.find(10)

Enumeration

DBF::Table is a Ruby Enumerable, so you get several traversal, search, and sort methods for free. For example, let's get only records created before January 1st, 2015:

widgets.select { |w| w.created_date < Date.new(2015, 1, 1) }

Or custom sorting:

widgets.sort_by { |w| w.created_date }

Encodings (Code Pages)

dBase supports encoding non-english characters with different character sets. Unfortunately, the character set used may not be set explicitly. In that case, you will have to specify it manually. For example, if you know the dbf file is encoded with 'Russian OEM':

table = DBF::Table.new('dbf/books.dbf', nil, 'cp866')

Code Page	Encoding	Description
01	cp437	U.S. MS–DOS
02	cp850	International MS–DOS
03	cp1252	Windows ANSI
04	macRoman	Standard Macintosh
08	cp865	Danish OEM
09	cp437	Dutch OEM
0a	cp850	Dutch OEM*
0b	cp437	Finnish OEM
0d	cp437	French OEM
0e	cp850	French OEM*
0f	cp437	German OEM
10	cp850	German OEM*
11	cp437	Italian OEM
12	cp850	Italian OEM*
13	cp932	Japanese Shift-JIS
14	cp850	Spanish OEM*
15	cp437	Swedish OEM
16	cp850	Swedish OEM*
17	cp865	Norwegian OEM
18	cp437	Spanish OEM
19	cp437	English OEM (Britain)
1a	cp850	English OEM (Britain)*
1b	cp437	English OEM (U.S.)
1c	cp863	French OEM (Canada)
1d	cp850	French OEM*
1f	cp852	Czech OEM
22	cp852	Hungarian OEM
23	cp852	Polish OEM
24	cp860	Portuguese OEM
25	cp850	Portuguese OEM*
26	cp866	Russian OEM
37	cp850	English OEM (U.S.)*
40	cp852	Romanian OEM
4d	cp936	Chinese GBK (PRC)
4e	cp949	Korean (ANSI/OEM)
4f	cp950	Chinese Big5 (Taiwan)
50	cp874	Thai (ANSI/OEM)
57	cp1252	ANSI
58	cp1252	Western European ANSI
59	cp1252	Spanish ANSI
64	cp852	Eastern European MS–DOS
65	cp866	Russian MS–DOS
66	cp865	Nordic MS–DOS
67	cp861	Icelandic MS–DOS
68	cp895	Kamenický (Czech) MS–DOS
69	cp620	Mazovia (Polish) MS–DOS
6a	cp737	Greek MS–DOS (437G)
6b	cp857	Turkish MS–DOS
6c	cp863	French–Canadian MS–DOS
78	cp950	Taiwan Big 5
79	cp949	Hangul (Wansung)
7a	cp936	PRC GBK
7b	cp932	Japanese Shift-JIS
7c	cp874	Thai Windows/MS–DOS
7d	cp1255	Hebrew Windows
7e	cp1256	Arabic Windows
86	cp737	Greek OEM
87	cp852	Slovenian OEM
88	cp857	Turkish OEM
96	macCyrillic	Russian Macintosh
97	macCentEuro	Macintosh EE
98	macGreek	Greek Macintosh
c8	cp1250	Eastern European Windows
c9	cp1251	Russian Windows
ca	cp1254	Turkish Windows
cb	cp1253	Greek Windows
cc	cp1257	Baltic Windows

Migrating to ActiveRecord

An example of migrating a DBF book table to ActiveRecord using a migration:

require 'dbf'

class Book < ActiveRecord::Base; end

class CreateBooks < ActiveRecord::Migration
  def self.up
    table = DBF::Table.new('db/dbf/books.dbf')
    eval(table.schema)

    Book.reset_column_information
    table.each do |record|
      Book.create(title: record.title, author: record.author)
    end
  end

  def self.down
    drop_table :books
  end
end

If you have initialized the DBF::Table with raw data, you will need to set the exported table name manually:

table.name = 'my_table_name'

Migrating to Sequel

An example of migrating a DBF book table to Sequel using a migration:

require 'dbf'

class Book < Sequel::Model; end

Sequel.migration do
  up do
    table = DBF::Table.new('db/dbf/books.dbf')
    eval(table.schema(:sequel, table_only: true)) # limit output to create_table() only

    Book.reset_column_information
    table.each do |record|
      Book.create(title: record.title, author: record.author)
    end
  end

  down do
    drop_table(:books)
  end
end

If you have initialized the DBF::Table with raw data, you will need to set the exported table name manually:

table.name = 'my_table_name'

Command-line utility

A small command-line utility called dbf is installed with the gem.

$ dbf -h
usage: dbf [-h|-s|-a|-c|-r|-j|-J] filename
  -h = print this message
  -v = print the DBF gem version
  -s = print summary information
  -a = create an ActiveRecord::Schema
  -r = create a Sequel migration
  -c = export as CSV
  -j = export as a JSON array
  -J = export as JSON Lines (one record per line)

Create an executable ActiveRecord schema:

dbf -a books.dbf > books_schema.rb

Create an executable Sequel schema:

dbf -r books.dbf > migrate/001_create_books.rb

Dump all records to a CSV file:

dbf -c books.dbf > books.csv

Dump all records as JSON or JSON Lines:

dbf -j books.dbf > books.json
dbf -J books.dbf > books.jsonl

JSON Lines output is streamed record by record, so it works well for very large files and pipelines (for example dbf -J books.dbf | jq or importing into DuckDB).

Reading a Visual Foxpro database (v8, v9)

A special Database::Foxpro class is available to read Visual Foxpro container files (file with .dbc extension). When using this class, long field names are supported, and tables can be referenced without using names.

require 'dbf'

contacts = DBF::Database::Foxpro.new('contact_database.dbc').contacts
my_contact = contacts.record(1).spouses_interests

dBase version compatibility

The basic dBase data types are generally supported well. Support for the advanced data types in dBase V and FoxPro are still experimental or not supported. If you have insight into how any of the unsupported data types are implemented, please open an issue on Github. FoxBase/dBase II files are not supported at this time.

Supported data types by dBase version

Version	Description	C	N	L	D	M	F	B	G	P	Y	T	I	V	X	@	O	+
02	FoxBase	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-	-
03	dBase III without memo file	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-	-
04	dBase IV without memo file	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-	-
05	dBase V without memo file	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-	-
07	Visual Objects 1.x	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-	-
30	Visual FoxPro	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y	Y	N	N	N	N	-
31	Visual FoxPro with AutoIncrement	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y	Y	N	N	N	N	N
32	Visual FoxPro with field type Varchar or Varbinary	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y	Y	N	N	N	N	N
7b	dBase IV with memo file	Y	Y	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-
83	dBase III with memo file	Y	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-
87	Visual Objects 1.x with memo file	Y	Y	Y	Y	Y	-	-	-	-	-	-	-	-	-	-	-	-
8b	dBase IV with memo file	Y	Y	Y	Y	Y	-	-	-	-	-	-	-	-	N	-	-	-
8e	dBase IV with SQL table	Y	Y	Y	Y	Y	-	-	-	-	-	-	-	-	N	-	-	-
f5	FoxPro with memo file	Y	Y	Y	Y	Y	Y	Y	Y	N	Y	Y	Y	N	N	N	N	N
fb	FoxPro without memo file	Y	Y	Y	Y	-	Y	Y	Y	N	Y	Y	Y	N	N	N	N	N

Data type descriptions

• C = Character
• N = Number
• L = Logical
• D = Date
• M = Memo
• F = Float
• B = Binary
• G = General
• P = Picture
• Y = Currency
• T = DateTime
• I = Integer
• V = VariField
• X = SQL compat
• @ = Timestamp
• O = Double
• • = Autoincrement

Processing untrusted files

The parser is hardened against malformed files: crafted headers cannot force huge allocations, unbounded loops, path traversal, or crashes from truncated data. If you accept DBF uploads (for example in a web application), also:

• Treat every column name and value as untrusted text. HTML-escape it before rendering it, and escape JSON before embedding it in a <script> tag.

• Pass the upload as a StringIO (with an explicit memo StringIO if there is one) rather than as a path, so memo files are never looked up in a shared upload directory.

• Limit upload size and processing time. Many records can point at the same large memo, so exported output can be far larger than the uploaded files. Pass max_memo_size: to cut off each memo after that many bytes:

  DBF::Table.new(upload, memo_upload, max_memo_size: 64 * 1024)

  DBF::Database::Foxpro.new takes the same option.

• Rescue DBF::Error, and also StandardError if a malformed file must never fail a request.

• Table#to_csv prefixes a ' to string cells that begin with =, +, -, @, tab, or carriage return so spreadsheets do not run them as formulas.

• Text is passed through as it is, including Unicode bidirectional controls (such as U+202E RIGHT-TO-LEFT OVERRIDE) and zero-width characters. These make displayed text differ from the underlying data (in a web page, a CSV export, or dbf terminal output), but some real data needs them, so the library does not remove them. Strip or flag them yourself where that matters:

  value.gsub(/[​-‏‪-‮⁠-⁤⁦-⁩]/, '')

• Column descriptors that do not fit within the header length or record length are ignored, and the library prints a warning (silenced when $VERBOSE is nil). If a file you trust loses columns, its header has wrong length fields.

Limitations

• DBF is read-only
• Index files are not utilized

License

Copyright (c) 2006-2026 Keith Morrison <keithm@infused.org>

Released under the MIT License. See LICENSE for the full text.