/ydbf

Pythonic reader and writer for DBF/XBase files

Primary LanguagePythonBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

YDbf

Actions Status PyPI Code style: black

YDbf - reading and writing DBF/XBase files in a pythonic way. The library written in pure Python and have no external dependencies.

YDbf is compatible with Python 3.5+. The last 2.7-compatible version of YDbf is 0.4.1.

What YDbf is good for:

  • export data to a DBF file
  • import data from a DBF file
  • read data from a DBF file as a stream

Where YDbf is not a good fit:

  • random access to records in a DBF file
  • memo fields

Read DBF

The entrypoint of YDbf is open function:

dbf = ydbf.open('simple.dbf')

You can use file name, or already opened in binary mode file:

fh = open('simple.dbf', 'rb')
dbf = ydbf.open(fh)

for record in dbf:
    ...

You may also use with statement:

with ydbf.open('simple.dbf') as dbf:
    for record in dbf:
        ...

Each record is a dict, which keys are names of fields.

Write DBF

YDbf opens file for reading by default, but you may set option mode to open for writing:

dbf = ydbf.open('simple.dbf', ydbf.WRITE, fields)

or open file yourself:

fh = open('simple.dbf', 'wb')
dbf = ydbf.open(fh, ydbf.WRITE, fields)

fields is a structure description of DBF file, it is a required option for write mode. The structure is as sequence of field descriptions, where each fields described by tuple (NAME, TYPE, SIZE, DECIMAL). NAME is a name of field, TYPE -- DBF type of field ('N' for number, 'C' for char, 'D' for date, 'L' for logical), DECIMAL is a precision (useful for 'N' type only).

YDbf offers the field types as constants:

  • ydbf.CHAR
  • ydbf.DATE
  • ydbf.LOGICAL
  • ydbf.NUMERAL

An example of the fields definition:

fields = [
    ('ID',      ydbf.NUMERAL,  4, 0),
    ('VALUE',   ydbf.CHAR, 40, 0),
    ('UPDATE',  ydbf.DATE, 8, 0),
    ('VISIBLE', ydbf.LOGICAL, 1, 0),
]

YDbf uses unicode for 'C' fields by default, so you may want to define encoding which be used forthe DBF file. UTF-8 is not supported, you may use only 8-bit encodings.

dbf = ydbf.open('simple.dbf', ydbf.WRITE, fields, encoding='cp1251')
dbf.write(data)

YDbf gets data as an iterator where each item is a dict, which keys are name of fields. For example,

data = [
    {'ID': 1, 'VALUE': u'ydbf', 'VISIBLE': True,
     'UPDATE': datetime.date(2009, 7, 14)},
    {'ID': 2, 'VALUE': u'ydbf-dev', 'VISIBLE': False,
     'UPDATE': datetime.date(2009, 5, 15)},
    {'ID': 3, 'VALUE': u'pytils', 'VISIBLE': True,
     'UPDATE': datetime.date(2009, 5, 11)},
]