Read DBF Files with Python

2.0.7 · active · verified Fri Apr 10

dbfread is a pure Python library designed to read DBF files (used by databases like dBase, Visual FoxPro, and FoxBase+). It processes the data and returns it as native Python data types, making it ideal for batch jobs and one-off scripts. The current version is 2.0.7, and while stable, it is not actively developed, with its last update being in late 2016.

Warnings

breaking The functions `dbfread.open()` and `dbfread.read()` are deprecated since version 2.0 and are scheduled for removal in version 2.2. Direct list-like access on the `DBF` object (e.g., `table[0]`) no longer works by default since version 1.1.0.
Fix: Always instantiate the `DBF` class directly (e.g., `table = DBF('filename.dbf')`). For list-like access, pass `load=True` to the `DBF` constructor, then access records via `table.records[index]` (e.g., `DBF('filename.dbf', load=True).records[0]`).
gotcha Character encoding issues are common with DBF files. While `dbfread` attempts to detect the encoding, it may fall back to `ASCII` (since 1.1.0) or `latin1` (older versions), leading to `UnicodeDecodeError` or incorrect characters.
Fix: If character encoding issues arise, explicitly specify the correct encoding using the `encoding` parameter in the `DBF` constructor (e.g., `DBF('file.dbf', encoding='cp1252')`). Common encodings include 'latin1', 'cp1252', or 'utf-8'.
gotcha Memo files (`.fpt`, `.dbt`) associated with DBF files can cause issues if missing or if file casing on non-Windows systems (e.g., Linux) prevents detection (as Windows filenames are case-preserving).
Fix: To handle missing memo files, pass `ignore_missing_memofile=True` to the `DBF` constructor to suppress `MissingMemoFile` exceptions. If case sensitivity is an issue, consider passing `ignorecase=False` if you know the exact casing of your memo file.
gotcha DBF field names are typically uppercase. When accessing fields as dictionary keys, this might be inconvenient in Python where lowercase conventions are common.
Fix: Use the `lowernames=True` parameter in the `DBF` constructor to automatically convert all field names to lowercase, allowing access like `record['fieldname']` instead of `record['FIELDNAME']`.

Install

pip install dbfread Install stable version

Imports

DBF
```
from dbfread import DBF
```
open
```
from dbfread import DBF
```
The `dbfread.open()` function is deprecated since version 2.0 and will be removed in 2.2. Instantiate the `DBF` class directly instead.
read
```
from dbfread import DBF
```
The `dbfread.read()` function is deprecated since version 2.0 and will be removed in 2.2. Instantiate the `DBF` class directly instead, using `load=True` if random access to records is needed.

Quickstart

This quickstart demonstrates how to open a DBF file and iterate through its records. It also shows how to load all records into memory if random access by index is required. It assumes a 'people.dbf' file exists for demonstration.

import datetime
from dbfread import DBF

# For this example to run, ensure 'people.dbf' exists in the same directory
# or provide a full path. A dummy 'people.dbf' might look like:
# NAME      BIRTHDATE
# Alice     1987-03-01
# Bob       1980-11-12

# Open the DBF file and iterate through records (streaming, memory-efficient)
print("Streaming records:")
try:
    for record in DBF('people.dbf'):
        print(record)
except Exception as e:
    print(f"Error reading DBF: {e}. Please ensure 'people.dbf' exists in the current directory.")

# Load all records into memory for random access (if needed)
print("\nLoading all records into memory for random access:")
try:
    table = DBF('people.dbf', load=True)
    print(f"First record name: {table.records[0]['NAME']}")
    print(f"Total records: {len(table.records)}")
except Exception as e:
    print(f"Error reading DBF with load=True: {e}. Please ensure 'people.dbf' exists.")

view raw JSON →