mirror of
https://github.com/psycopg/psycopg2.git
synced 2024-11-22 17:06:33 +03:00
125 lines
4.3 KiB
Python
125 lines
4.3 KiB
Python
# -*- python -*-
|
|
#
|
|
# Copyright (C) 2001-2003 Federico Di Gregorio <fog@debian.org>
|
|
#
|
|
# This file is part of the psycopg module.
|
|
#
|
|
# This program is free software; you can redistribute it and/or
|
|
# modify it under the terms of the GNU General Public License
|
|
# as published by the Free Software Foundation; either version 2,
|
|
# or (at your option) any later version.
|
|
#
|
|
# This program is distributed in the hope that it will be useful,
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
# GNU General Public License for more details.
|
|
#
|
|
# You should have received a copy of the GNU General Public License
|
|
# along with this program; if not, write to the Free Software
|
|
# Foundation, 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
|
|
|
# this a little script that analyze a file with (TYPE, NUMBER) tuples
|
|
# and write out C code ready for inclusion in psycopg. the generated
|
|
# code defines the DBAPITypeObject fundamental types and warns for
|
|
# undefined types.
|
|
|
|
import sys, os, string, copy
|
|
from string import split, join, strip
|
|
|
|
|
|
# here is the list of the foundamental types we want to import from
|
|
# postgresql header files
|
|
|
|
basic_types = (['NUMBER', ['INT8', 'INT4', 'INT2', 'FLOAT8', 'FLOAT4',
|
|
'NUMERIC']],
|
|
['LONGINTEGER', ['INT8']],
|
|
['INTEGER', ['INT4', 'INT2']],
|
|
['FLOAT', ['FLOAT8', 'FLOAT4']],
|
|
['DECIMAL', ['NUMERIC']],
|
|
['UNICODE', ['NAME', 'CHAR', 'TEXT', 'BPCHAR',
|
|
'VARCHAR']],
|
|
['STRING', ['NAME', 'CHAR', 'TEXT', 'BPCHAR',
|
|
'VARCHAR']],
|
|
['BOOLEAN', ['BOOL']],
|
|
['DATETIME', ['TIMESTAMP', 'TIMESTAMPTZ',
|
|
'TINTERVAL', 'INTERVAL']],
|
|
['TIME', ['TIME', 'TIMETZ']],
|
|
['DATE', ['DATE']],
|
|
['INTERVAL', ['TINTERVAL', 'INTERVAL']],
|
|
['BINARY', ['BYTEA']],
|
|
['ROWID', ['OID']])
|
|
|
|
# unfortunately we don't have a nice way to extract array information
|
|
# from postgresql headers; we'll have to do it hard-coding :/
|
|
array_types = (['LONGINTEGER', [1016]],
|
|
['INTEGER', [1005, 1006, 1007]],
|
|
['FLOAT', [1017, 1021, 1022]],
|
|
['DECIMAL', [1231]],
|
|
['UNICODE', [1002, 1003, 1009, 1014, 1015]],
|
|
['STRING', [1002, 1003, 1009, 1014, 1015]],
|
|
['BOOLEAN', [1000]],
|
|
['DATETIME', [1115, 1185]],
|
|
['TIME', [1183, 1270]],
|
|
['DATE', [1182]],
|
|
['INTERVAL', [1187]],
|
|
['BINARY', [1001]],
|
|
['ROWID', [1028, 1013]])
|
|
|
|
# this is the header used to compile the data in the C module
|
|
HEADER = """
|
|
typecastObject_initlist typecast_builtins[] = {
|
|
"""
|
|
|
|
# then comes the footer
|
|
FOOTER = """ {NULL, NULL, NULL, NULL}\n};\n"""
|
|
|
|
|
|
# useful error reporting function
|
|
def error(msg):
|
|
"""Report an error on stderr."""
|
|
sys.stderr.write(msg+'\n')
|
|
|
|
|
|
# read couples from stdin and build list
|
|
read_types = []
|
|
for l in sys.stdin.readlines():
|
|
oid, val = split(l)
|
|
read_types.append((strip(oid)[:-3], strip(val)))
|
|
|
|
# look for the wanted types in the read touples
|
|
found_types = {}
|
|
|
|
for t in basic_types:
|
|
k = t[0]
|
|
found_types[k] = []
|
|
for v in t[1]:
|
|
found = filter(lambda x, y=v: x[0] == y, read_types)
|
|
if len(found) == 0:
|
|
error(v+': value not found')
|
|
elif len(found) > 1:
|
|
error(v+': too many values')
|
|
else:
|
|
found_types[k].append(int(found[0][1]))
|
|
|
|
# now outputs to stdout the right C-style definitions
|
|
stypes = "" ; sstruct = ""
|
|
for t in basic_types:
|
|
k = t[0]
|
|
s = str(found_types[k])
|
|
s = '{' + s[1:-1] + ', 0}'
|
|
stypes = stypes + ('static long int typecast_%s_types[] = %s;\n' % (k, s))
|
|
sstruct += (' {"%s", typecast_%s_types, typecast_%s_cast, NULL},\n'
|
|
% (k, k, k))
|
|
for t in array_types:
|
|
kt = t[0]
|
|
ka = t[0]+'ARRAY'
|
|
s = str(t[1])
|
|
s = '{' + s[1:-1] + ', 0}'
|
|
stypes = stypes + ('static long int typecast_%s_types[] = %s;\n' % (ka, s))
|
|
sstruct += (' {"%s", typecast_%s_types, typecast_%s_cast, "%s"},\n'
|
|
% (ka, ka, ka, kt))
|
|
sstruct = HEADER + sstruct + FOOTER
|
|
|
|
print stypes
|
|
print sstruct
|