[537] | 1 | ###########################################################################
|
---|
| 2 | #
|
---|
| 3 | # sorttools.pm --
|
---|
| 4 | # A component of the Greenstone digital library software
|
---|
| 5 | # from the New Zealand Digital Library Project at the
|
---|
| 6 | # University of Waikato, New Zealand.
|
---|
| 7 | #
|
---|
| 8 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
| 9 | #
|
---|
| 10 | # This program is free software; you can redistribute it and/or modify
|
---|
| 11 | # it under the terms of the GNU General Public License as published by
|
---|
| 12 | # the Free Software Foundation; either version 2 of the License, or
|
---|
| 13 | # (at your option) any later version.
|
---|
| 14 | #
|
---|
| 15 | # This program is distributed in the hope that it will be useful,
|
---|
| 16 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
| 17 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
| 18 | # GNU General Public License for more details.
|
---|
| 19 | #
|
---|
| 20 | # You should have received a copy of the GNU General Public License
|
---|
| 21 | # along with this program; if not, write to the Free Software
|
---|
| 22 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
| 23 | #
|
---|
| 24 | ###########################################################################
|
---|
| 25 |
|
---|
[383] | 26 | # various subroutines to format strings
|
---|
| 27 | # for sorting
|
---|
| 28 |
|
---|
| 29 | package sorttools;
|
---|
| 30 |
|
---|
| 31 |
|
---|
| 32 | # format an english string for sorting
|
---|
| 33 | # i.e. convert to lowercase, remove the, a or an
|
---|
| 34 | # from beginning of string etc.
|
---|
| 35 | sub format_string_english {
|
---|
| 36 | my ($stringref) = @_;
|
---|
| 37 |
|
---|
| 38 | $$stringref =~ tr/A-Z/a-z/;
|
---|
| 39 | $$stringref =~ s/[^a-z0-9 ]//g;
|
---|
| 40 | $$stringref =~ s/^\s*(the|a|an)\b//;
|
---|
| 41 | $$stringref =~ s/^\s+//;
|
---|
| 42 | }
|
---|
| 43 |
|
---|
| 44 | # format an english name for sorting
|
---|
| 45 | # i.e. convert to lowercase, put surname before
|
---|
| 46 | # first names etc.
|
---|
| 47 | sub format_string_name_english {
|
---|
| 48 | my ($stringref) = @_;
|
---|
| 49 |
|
---|
| 50 | $$stringref =~ tr/A-Z/a-z/;
|
---|
| 51 | $$stringref =~ s/[^a-z0-9 ]//g;
|
---|
| 52 | $$stringref =~ s/\s+/ /g;
|
---|
| 53 | $$stringref =~ s/^\s+//;
|
---|
| 54 | my @names = split / /, $$stringref;
|
---|
| 55 | my $surname = pop @names;
|
---|
| 56 | while (scalar @names && $surname =~ /^(jnr|snr)$/i) {
|
---|
| 57 | $surname = pop @names;
|
---|
| 58 | }
|
---|
| 59 | $$stringref = $surname . " " . $$stringref;
|
---|
| 60 | }
|
---|
| 61 |
|
---|
[392] | 62 | # takes arguments of day, month, year and converts to
|
---|
[467] | 63 | # date of form yyyymmdd. month may be full (e.g. "January"),
|
---|
| 64 | # abbreviated (e.g. "Jan"), or a number (1-12). Years like "86"
|
---|
| 65 | # will be assumed to be "1986".
|
---|
[392] | 66 | sub format_date {
|
---|
| 67 | my ($day, $month, $year) = @_;
|
---|
| 68 |
|
---|
| 69 | my %months = ('january' => '01', 'jan' => '01', 'february' => '02', 'feb' => '02',
|
---|
| 70 | 'march' => '03', 'mar' => '03', 'april' => '04', 'apr' => '04',
|
---|
| 71 | 'may' => '05', 'june' => '06', 'jun' => '06', 'july' => '07',
|
---|
| 72 | 'jul' => '07', 'august' => '08', 'aug' => '08', 'september' => '09',
|
---|
| 73 | 'sep' => '09', 'october' => '10', 'oct' => '10', 'november' => '11',
|
---|
| 74 | 'nov' => '11', 'december' => '12', 'dec' => '12');
|
---|
| 75 |
|
---|
| 76 | $month =~ tr/A-Z/a-z/;
|
---|
| 77 |
|
---|
| 78 | if ($day < 1) {
|
---|
| 79 | print STDERR "sorttools::format_date WARNING day $day out of range\n";
|
---|
| 80 | $day = "01";
|
---|
| 81 | } elsif ($day > 31) {
|
---|
| 82 | print STDERR "sorttools::format_date WARNING day $day out of range\n";
|
---|
| 83 | $day = "31";
|
---|
| 84 | }
|
---|
| 85 |
|
---|
| 86 | $day = "0$day" if (length($day) == 1);
|
---|
| 87 |
|
---|
[467] | 88 | if ($month =~ /^\d\d?$/) {
|
---|
| 89 | if ($month < 1) {
|
---|
| 90 | print STDERR "sorttools::format_date WARNING month $month out of range\n";
|
---|
| 91 | $month = "01";
|
---|
| 92 | } elsif ($month > 12) {
|
---|
| 93 | print STDERR "sorttools::format_date WARNING month $month out of range\n";
|
---|
| 94 | $month = "12";
|
---|
| 95 | }
|
---|
| 96 | if ($month =~ /^\d$/) {
|
---|
| 97 | $month = "0" . $month;
|
---|
| 98 | }
|
---|
| 99 | } elsif (!defined $months{$month}) {
|
---|
[392] | 100 | print STDERR "sorttools::format_date WARNING month $month out of range\n";
|
---|
| 101 | $month = "01";
|
---|
| 102 | } else {
|
---|
| 103 | $month = $months{$month};
|
---|
| 104 | }
|
---|
| 105 |
|
---|
| 106 | if ($year !~ /^\d\d\d\d$/) {
|
---|
| 107 | if ($year !~ /^\d\d$/) {
|
---|
| 108 | print STDERR "sorttools::format_date WARNING year $year assumed to be 19$year\n";
|
---|
| 109 | $year = "19" . $year;
|
---|
| 110 | } else {
|
---|
| 111 | print STDERR "sorttools::format_date WARNING year $year out of range - reset to 1900\n";
|
---|
| 112 | $year = "1900";
|
---|
| 113 | }
|
---|
| 114 | }
|
---|
| 115 |
|
---|
| 116 | return "$year$month$day";
|
---|
| 117 | }
|
---|
| 118 |
|
---|
| 119 |
|
---|
[383] | 120 | 1;
|
---|