python读写dbf⽂件
最近要完成数据导出dbf格式的⽂件的功能，在⽹上了⼀个这个库，折腾了半天⼜是字符编码问题，后来才搜到了这段代码，⽐较轻量级的，虽然已是⼗年前的代码，暂时没有发现什么⼤的问题，调试起来⽐较⽅便。
Python读写dbf⽂件
# coding=utf8
"""
A reader and writer for dbf file.see code.activestate/recipes/362715/ for moe detail
"""
import struct
import datetime
import decimal
import itertools
def dbfreader(f):
"""Returns an iterator over records in a Xbase DBF file.
python怎么读文件The first row returned contains the field names.
The second row contains field specs: (type, size, decimal places).
Subsequent rows contain the data records.
If a record is marked as deleted, it is skipped.
File should be opened for binary reads.
"""
# See DBF format spec at:
# www.pgts.au/download/public/xbase.htm#DBF_STRUCT
numrec, lenheader = struct.unpack('<xxxxLH22x', f.read(32))
numfields = (lenheader - 33) // 32
fields = []
for fieldno in xrange(numfields):
name, typ, size, deci = struct.unpack('<11sc4xBB14x', f.read(32))
name = place('\0', '') # eliminate NULs from string
fields.append((name, typ, size, deci))
yield [field[0] for field in fields]
yield [tuple(field[1:]) for field in fields]
terminator = f.read(1)
assert terminator == '\r'
fields.insert(0, ('DeletionFlag', 'C', 1, 0))
fmt = ''.join(['%ds' % fieldinfo[2] for fieldinfo in fields])
fmtsiz = struct.calcsize(fmt)
for i in xrange(numrec):
record = struct.unpack(fmt, f.read(fmtsiz))
if record[0] != ' ':
continue # deleted record
result = []
for (name, typ, size, deci), value in itertools.izip(fields, record):
if name == 'DeletionFlag':
continue
if typ == "N":
value = place('\0', '').lstrip()
if value == '':
value = 0
elif deci:
value = decimal.Decimal(value)
else:
else:
value = int(value)
elif typ == 'D':
y, m, d = int(value[:4]), int(value[4:6]), int(value[6:8])
value = datetime.date(y, m, d)
elif typ == 'L':
value = (value in 'YyTt' and 'T') or (value in 'NnFf' and 'F') or '?'
elif typ == 'F':
value = float(value)
result.append(value)
yield result
def dbfwriter(f, fieldnames, fieldspecs, records):
""" Return a string suitable for writing directly to a binary dbf file.
File f should be open for writing in a binary mode.
Fieldnames should be no longer than ten characters and not include \x00.
Fieldspecs are in the form (type, size, deci) where
type is one of:
C for ascii character data
M for ascii character memo data (real memo fields not supported)
D for datetime objects
N for ints or decimal objects
L for logical values 'T', 'F', or '?'
size is the field width
deci is the number of decimal places in the provided decimal object
Records can be an iterable over the records (sequences of field values).
"""
# header info
ver = 3
now = w()
yr, mon, day = ar - 1900, h, now.day
numrec = len(records)
numfields = len(fieldspecs)
lenheader = numfields * 32 + 33
lenrecord = sum(field[1] for field in fieldspecs) + 1
hdr = struct.pack('<BBBBLHH20x', ver, yr, mon, day, numrec, lenheader, lenrecord) f.write(hdr)
# field specs
for name, (typ, size, deci) in itertools.izip(fieldnames, fieldspecs):
name = name.ljust(11, '\x00')
fld = struct.pack('<11sc4xBB14x', name, typ, size, deci)
f.write(fld)
# terminator
f.write('\r')
# records
for record in records:
f.write(' ') # deletion flag
for (typ, size, deci), value in itertools.izip(fieldspecs, record):
if typ == "N":
value = str(value).rjust(size, ' ')
elif typ == 'D':
value = value.strftime('%Y%m%d')
elif typ == 'L':
value = str(value)[0].upper()
else:
value = str(value)[:size].ljust(size, ' ')
assert len(value) == size
f.write(value)
# End of file
f.write('\x1A')
# -------------------------------------------------------
# Example calls
if __name__ == '__main__':
import sys
import csv
from cStringIO import StringIO
from operator import itemgetter
# Read a database
filename = '/pydev/databases/orders.dbf'
if len(sys.argv) == 2:
filename = sys.argv[1]
f = open(filename, 'rb')
db = list(dbfreader(f))
f.close()
for record in db:
print record
fieldnames, fieldspecs, records = db[0], db[1], db[2:]
# Alter the database
del records[4]
records.sort(key=itemgetter(4))
# Remove a field
del fieldnames[0]
del fieldspecs[0]
records = [rec[1:] for rec in records]
# Create a new DBF
f = StringIO()
dbfwriter(f, fieldnames, fieldspecs, records)
# Read the data back from the new DBF
print '-' * 20
f.seek(0)
for line in dbfreader(f):
print line
f.close()
# Convert to CSV
print '.' * 20
f = StringIO()
csv.writer(f).writerow(fieldnames)
csv.writer(f).writerows(records)
value()
f.close()
# Example Output
"""
['ORDER_ID', 'CUSTMR_ID', 'EMPLOY_ID', 'ORDER_DATE', 'ORDER_AMT'] [('C', 10, 0), ('C', 11, 0), ('C', 11, 0), ('D', 8, 0), ('N', 12, 2)]
['10005 ', 'WALNG ', '555 ', datetime.date(1995, 5, 22), Decimal("173.40")] ['10004 ', 'BMARK ', '777 ', datetime.date(1995, 5, 18), Decimal("3194.20")] ['10029 ', 'SAWYH ', '7
77 ', datetime.date(1995, 6, 29), Decimal("97.30")] ['10013 ', 'RITEB ', '777 ', datetime.date(1995, 6, 2), Decimal("560.40")] ['10024 ', 'RATTC ', '444 ', datetime.date(1995, 6, 21), Decimal("2223.50")] ['10018 ', 'RATTC ', '444 ', datetime.date(1995, 6, 12), Decimal("1076.05")] ['10025 ', 'RATTC ', '444 ', datetime.date(1995, 6, 23), Decimal("185.80")] ['10038 ', 'OLDWO ', '111 ', datetime.date(1995, 7, 14), Decimal("863.96")] ['10002 ', 'MTIME ', '333 ', datetime.date(1995, 5, 16), Decimal("731.80")]
['10002 ', 'MTIME ', '333 ', datetime.date(1995, 5, 16), Decimal("731.80")] ['10007 ', 'MORNS ', '444 ', datetime.date(1995, 5, 24), Decimal("1405.00")] ['10026 ', 'MORNS ', '555 ', datetime.date(1995, 6, 26), Decimal("17.40")] ['10030 ', 'LILLO ', '111 ', datetime.date(1995, 7, 3), Decimal("909.91")] ['10022 ', 'LAPLA ', '111 ', datetime.date(1995, 6, 19), Decimal("671.50")] ['10035 ', 'HIGHG ', '111 ', datetime.date(1995, 7, 11), Decimal("1984.83")] ['10033 ', 'FOODG ', '333 ', datetime.date(1995, 7, 6), Decimal("3401.32")] --------------------
['CUSTMR_ID', 'EMPLOY_ID', 'ORDER_DATE', 'ORDER_AMT']
[('C', 11, 0), ('C', 11, 0), ('D', 8, 0), ('N', 12, 2)]
['MORNS ', '555 ', datetime.date(1995, 6, 26), Decimal("17.40")]
['SAWYH ', '777 ', datetime.date(1995, 6, 29), Decimal("97.30")]
['WALNG ', '555 ', datetime.date(1995, 5, 22), Decimal("173.40")]
['RATTC ', '444 ', datetime.date(1995, 6, 23), Decimal("185.80")]
['RITEB ', '777 ', datetime.date(1995, 6, 2), Decimal("560.40")]
['LAPLA ', '111 ', datetime.date(1995, 6, 19), Decimal("671.50")]
['MTIME ', '333 ', datetime.date(1995, 5, 16), Decimal("731.80")]
['OLDWO ', '111 ', datetime.date(1995, 7, 14), Decimal("863.96")]
['LILLO ', '111 ', datetime.date(1995, 7, 3), Decimal("909.91")]
['RATTC ', '444 ', datetime.date(1995, 6, 12), Decimal("1076.05")]
['MORNS ', '444 ', datetime.date(1995, 5, 24), Decimal("1405.00")]
['HIGHG ', '111 ', datetime.date(1995, 7, 11), Decimal("1984.83")]
['BMARK ', '777 ', datetime.date(1995, 5, 18), Decimal("3194.20")]
['FOODG ', '333 ', datetime.date(1995, 7, 6), Decimal("3401.32")]
....................
CUSTMR_ID,EMPLOY_ID,ORDER_DATE,ORDER_AMT
MORNS ,555 ,1995-06-26,17.40
SAWYH ,777 ,1995-06-29,97.30
WALNG ,555 ,1995-05-22,173.40
RATTC ,444 ,1995-06-23,185.80
RITEB ,777 ,1995-06-02,560.40
LAPLA ,111 ,1995-06-19,671.50
MTIME ,333 ,1995-05-16,731.80
OLDWO ,111 ,1995-07-14,863.96
LILLO ,111 ,1995-07-03,909.91
RATTC ,444 ,1995-06-12,1076.05
MORNS ,444 ,1995-05-24,1405.00
HIGHG ,111 ,1995-07-11,1984.83
BMARK ,777 ,1995-05-18,3194.20
FOODG ,333 ,1995-07-06,3401.32
"""
转载于:my.oschina/kinegratii/blog/394562
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论