[7512] | 1 | ## |
---|
| 2 | ## fix_import_file.py |
---|
| 3 | ## Login : <uli@pu.smp.net> |
---|
| 4 | ## Started on Wed Jan 25 17:08:30 2012 Uli Fouquet |
---|
| 5 | ## $Id$ |
---|
| 6 | ## |
---|
| 7 | ## Copyright (C) 2012 Uli Fouquet |
---|
| 8 | ## This program is free software; you can redistribute it and/or modify |
---|
| 9 | ## it under the terms of the GNU General Public License as published by |
---|
| 10 | ## the Free Software Foundation; either version 2 of the License, or |
---|
| 11 | ## (at your option) any later version. |
---|
| 12 | ## |
---|
| 13 | ## This program is distributed in the hope that it will be useful, |
---|
| 14 | ## but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
| 15 | ## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
| 16 | ## GNU General Public License for more details. |
---|
| 17 | ## |
---|
| 18 | ## You should have received a copy of the GNU General Public License |
---|
| 19 | ## along with this program; if not, write to the Free Software |
---|
| 20 | ## Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA |
---|
| 21 | ## |
---|
| 22 | """ |
---|
| 23 | Fix exports from old SIRP portal to make them importable by current portal. |
---|
| 24 | |
---|
| 25 | Usage: |
---|
| 26 | |
---|
| 27 | Change into this directory, set the options below (files are assumed |
---|
| 28 | to be in the same directory) and then run |
---|
| 29 | |
---|
| 30 | python fix_import_file.py |
---|
| 31 | |
---|
| 32 | Errors/warnings will be displayed on the shell, the output will be put |
---|
| 33 | into the specified output file. |
---|
| 34 | """ |
---|
| 35 | |
---|
| 36 | ## |
---|
| 37 | ## CONFIGURATION SECTION |
---|
| 38 | ## |
---|
| 39 | # file with input data |
---|
| 40 | INPUT_FILE = 'students_for_reimport.csv' |
---|
| 41 | |
---|
| 42 | # file written with modified output |
---|
| 43 | OUTPUT_FILE = 'out.csv' |
---|
| 44 | |
---|
| 45 | # keys are fieldnames in input file, values are methods of class |
---|
| 46 | # Converter (see below) |
---|
| 47 | OPTIONS = { |
---|
| 48 | 'sex': 'gender', |
---|
| 49 | 'birthday': 'date', |
---|
| 50 | 'request_date': 'datetime', |
---|
| 51 | } |
---|
| 52 | ## |
---|
| 53 | ## END OF CONFIG |
---|
| 54 | ## |
---|
| 55 | |
---|
| 56 | import csv |
---|
| 57 | import datetime |
---|
| 58 | import sys |
---|
| 59 | |
---|
| 60 | class Converters(): |
---|
| 61 | """Converters to turn old-style values into new ones. |
---|
| 62 | """ |
---|
| 63 | @classmethod |
---|
| 64 | def gender(self, value): |
---|
| 65 | """ 'True'/'False' --> 'female'/'male'/'' |
---|
| 66 | """ |
---|
| 67 | if value == 'True': |
---|
| 68 | value = 'female' |
---|
| 69 | elif value == 'False': |
---|
| 70 | value = 'male' |
---|
| 71 | else: |
---|
| 72 | value = '' |
---|
| 73 | return value |
---|
| 74 | |
---|
| 75 | @classmethod |
---|
| 76 | def date(self, value): |
---|
| 77 | """ 'yyyy/mm/dd' --> 'yyyy-mm-dd' |
---|
| 78 | """ |
---|
| 79 | if value == "None": |
---|
| 80 | value = "" |
---|
| 81 | elif value == "": |
---|
| 82 | value = "" |
---|
| 83 | else: |
---|
| 84 | value = value.replace('/', '-') |
---|
| 85 | return value |
---|
| 86 | |
---|
| 87 | @classmethod |
---|
| 88 | def datetime(self, value): |
---|
| 89 | """ 'yyyy/mm/dd' --> 'yyyy-mm-dd' |
---|
| 90 | """ |
---|
| 91 | #print "IN: ", value |
---|
| 92 | if value == "None": |
---|
| 93 | value = "" |
---|
| 94 | elif value == "": |
---|
| 95 | value = "" |
---|
| 96 | else: |
---|
| 97 | #value = datetime.datetime.strptime(value, '%Y/%m/%d') |
---|
| 98 | #value = datetime.datetime.strftime(value, '%Y-%m-%d') |
---|
| 99 | pass |
---|
| 100 | #print "OUT: ", value |
---|
| 101 | return value |
---|
| 102 | |
---|
| 103 | reader = csv.DictReader(open(INPUT_FILE, 'rb')) |
---|
| 104 | writer = None |
---|
| 105 | |
---|
| 106 | for num, row in enumerate(reader): |
---|
| 107 | if num == 0: |
---|
| 108 | writer = csv.DictWriter(open(OUTPUT_FILE, 'wb'), reader.fieldnames) |
---|
| 109 | print "FIELDS: " |
---|
| 110 | for x, y in enumerate(reader.fieldnames): |
---|
| 111 | print x, y |
---|
| 112 | header = dict([(x, x) for x in reader.fieldnames]) |
---|
| 113 | writer.writerow(header) |
---|
| 114 | for key, value in row.items(): |
---|
| 115 | if not key in OPTIONS.keys(): |
---|
| 116 | continue |
---|
| 117 | conv_name = OPTIONS[key] |
---|
| 118 | converter = getattr(Converters, conv_name, None) |
---|
| 119 | if converter is None: |
---|
| 120 | print "WARNING: cannot find converter %s" % conv_name |
---|
| 121 | continue |
---|
| 122 | row[key] = converter(row[key]) |
---|
| 123 | writer.writerow(row) |
---|
| 124 | |
---|
| 125 | print "Ouput written to %s" % OUTPUT_FILE |
---|