xref: /webtrees/app/Services/GedcomService.php (revision e873f434551745f888937263ff89e80db3b0f785)
1<?php
2
3/**
4 * webtrees: online genealogy
5 * Copyright (C) 2023 webtrees development team
6 * This program is free software: you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation, either version 3 of the License, or
9 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
14 * You should have received a copy of the GNU General Public License
15 * along with this program. If not, see <https://www.gnu.org/licenses/>.
16 */
17
18declare(strict_types=1);
19
20namespace Fisharebest\Webtrees\Services;
21
22use Fisharebest\Webtrees\Gedcom;
23
24/**
25 * Utilities for manipulating GEDCOM data.
26 */
27class GedcomService
28{
29    // Some applications, such as FTM, use GEDCOM tag names instead of the tags.
30    private const array TAG_NAMES = [
31        'ABBREVIATION'      => 'ABBR',
32        'ADDRESS'           => 'ADDR',
33        'ADDRESS1'          => 'ADR1',
34        'ADDRESS2'          => 'ADR2',
35        'ADDRESS3'          => 'ADR3',
36        'ADOPTION'          => 'ADOP',
37        'AGENCY'            => 'AGNC',
38        'ALIAS'             => 'ALIA',
39        'ANCESTORS'         => 'ANCE',
40        'ANCES_INTEREST'    => 'ANCI',
41        'ANULMENT'          => 'ANUL',
42        'ASSOCIATES'        => 'ASSO',
43        'AUTHOR'            => 'AUTH',
44        'BAPTISM-LDS'       => 'BAPL',
45        'BAPTISM'           => 'BAPM',
46        'BAR_MITZVAH'       => 'BARM',
47        'BAS_MITZVAH'       => 'BASM',
48        'BIRTH'             => 'BIRT',
49        'BLESSING'          => 'BLES',
50        'BURIAL'            => 'BURI',
51        'CALL_NUMBER'       => 'CALN',
52        'CASTE'             => 'CAST',
53        'CAUSE'             => 'CAUS',
54        'CENSUS'            => 'CENS',
55        'CHANGE'            => 'CHAN',
56        'CHARACTER'         => 'CHAR',
57        'CHILD'             => 'CHIL',
58        'CHRISTENING'       => 'CHR',
59        'ADULT_CHRISTENING' => 'CHRA',
60        'CONCATENATION'     => 'CONC',
61        'CONFIRMATION'      => 'CONF',
62        'CONFIRMATION-LDS'  => 'CONL',
63        'CONTINUED'         => 'CONT',
64        'COPYRIGHT'         => 'COPY',
65        'CORPORTATE'        => 'CORP',
66        'CREMATION'         => 'CREM',
67        'COUNTRY'           => 'CTRY',
68        'DEATH'             => 'DEAT',
69        'DESCENDANTS'       => 'DESC',
70        'DESCENDANTS_INT'   => 'DESI',
71        'DESTINATION'       => 'DEST',
72        'DIVORCE'           => 'DIV',
73        'DIVORCE_FILED'     => 'DIVF',
74        'PHY_DESCRIPTION'   => 'DSCR',
75        'EDUCATION'         => 'EDUC',
76        'EMIGRATION'        => 'EMIG',
77        'ENDOWMENT'         => 'ENDL',
78        'ENGAGEMENT'        => 'ENGA',
79        'EVENT'             => 'EVEN',
80        'FAMILY'            => 'FAM',
81        'FAMILY_CHILD'      => 'FAMC',
82        'FAMILY_FILE'       => 'FAMF',
83        'FAMILY_SPOUSE'     => 'FAMS',
84        'FACIMILIE'         => 'FAX',
85        'FIRST_COMMUNION'   => 'FCOM',
86        'FORMAT'            => 'FORM',
87        'PHONETIC'          => 'FONE',
88        'GEDCOM'            => 'GEDC',
89        'GIVEN_NAME'        => 'GIVN',
90        'GRADUATION'        => 'GRAD',
91        'HEADER'            => 'HEAD',
92        'HUSBAND'           => 'HUSB',
93        'IDENT_NUMBER'      => 'IDNO',
94        'IMMIGRATION'       => 'IMMI',
95        'INDIVIDUAL'        => 'INDI',
96        'LANGUAGE'          => 'LANG',
97        'LATITUDE'          => 'LATI',
98        'LONGITUDE'         => 'LONG',
99        'MARRIAGE_BANN'     => 'MARB',
100        'MARR_CONTRACT'     => 'MARC',
101        'MARR_LICENSE'      => 'MARL',
102        'MARRIAGE'          => 'MARR',
103        'MEDIA'             => 'MEDI',
104        'NATIONALITY'       => 'NATI',
105        'NATURALIZATION'    => 'NATU',
106        'CHILDREN_COUNT'    => 'NCHI',
107        'NICKNAME'          => 'NICK',
108        'MARRIAGE_COUNT'    => 'NMR',
109        'NAME_PREFIX'       => 'NPFX',
110        'NAME_SUFFIX'       => 'NSFX',
111        'OBJECT'            => 'OBJE',
112        'OCCUPATION'        => 'OCCU',
113        'ORDINANCE'         => 'ORDI',
114        'ORDINATION'        => 'ORDN',
115        'PEDIGREE'          => 'PEDI',
116        'PHONE'             => 'PHON',
117        'PLACE'             => 'PLAC',
118        'POSTAL_CODE'       => 'POST',
119        'PROBATE'           => 'PROB',
120        'PROPERTY'          => 'PROP',
121        'PUBLICATION'       => 'PUBL',
122        'QUALITY_OF_DATA'   => 'QUAY',
123        'REFERENCE'         => 'REFN',
124        'RELATIONSHIP'      => 'RELA',
125        'RELIGION'          => 'RELI',
126        'REPOSITORY'        => 'REPO',
127        'RESIDENCE'         => 'RESI',
128        'RESTRICTION'       => 'RESN',
129        'RETIREMENT'        => 'RETI',
130        'REC_FILE_NUMBER'   => 'RFN',
131        'REC_ID_NUMBER'     => 'RIN',
132        'ROMANIZED'         => 'ROMN',
133        'SEALING_CHILD'     => 'SLGC',
134        'SEALING_SPOUSE'    => 'SLGS',
135        'SOURCE'            => 'SOUR',
136        'SURN_PREFIX'       => 'SPFX',
137        'SOC_SEC_NUMBER'    => 'SSN',
138        'STATE'             => 'STAE',
139        'STATUS'            => 'STAT',
140        'SUBMITTER'         => 'SUBM',
141        'SUBMISSION'        => 'SUBN',
142        'SURNAME'           => 'SURN',
143        'TEMPLE'            => 'TEMP',
144        'TITLE'             => 'TITL',
145        'TRAILER'           => 'TRLR',
146        'VERSION'           => 'VERS',
147        'WEB'               => 'WWW',
148        '_DEATH_OF_SPOUSE'  => 'DETS',
149        '_DEGREE'           => '_DEG',
150        '_MEDICAL'          => '_MCL',
151        '_MILITARY_SERVICE' => '_MILT',
152    ];
153
154    // Custom GEDCOM tags used by other applications, with direct synonyms
155    private const array TAG_SYNONYMS = [
156        // Convert PhpGedView tag to webtrees
157        '_PGVU'     => '_WT_USER',
158        '_PGV_OBJS' => '_WT_OBJE_SORT',
159    ];
160
161    /**
162     * Convert a GEDCOM tag to a canonical form.
163     *
164     * @param string $tag
165     *
166     * @return string
167     */
168    public function canonicalTag(string $tag): string
169    {
170        $tag = strtoupper($tag);
171
172        $tag = self::TAG_NAMES[$tag] ?? self::TAG_SYNONYMS[$tag] ?? $tag;
173
174        return $tag;
175    }
176
177    public function readLatitude(string $text): float|null
178    {
179        return $this->readDegrees($text, Gedcom::LATITUDE_NORTH, Gedcom::LATITUDE_SOUTH);
180    }
181
182    public function readLongitude(string $text): float|null
183    {
184        return $this->readDegrees($text, Gedcom::LONGITUDE_EAST, Gedcom::LONGITUDE_WEST);
185    }
186
187    private function readDegrees(string $text, string $positive, string $negative): float|null
188    {
189        $text       = trim($text);
190        $hemisphere = substr($text, 0, 1);
191        $degrees    = substr($text, 1);
192
193        // Match a valid GEDCOM format
194        if (is_numeric($degrees)) {
195            $hemisphere = strtoupper($hemisphere);
196            $degrees    = (float) $degrees;
197
198            if ($hemisphere === $positive) {
199                return $degrees;
200            }
201
202            if ($hemisphere === $negative) {
203                return -$degrees;
204            }
205        }
206
207        // Just a number?
208        if (is_numeric($text)) {
209            return (float) $text;
210        }
211
212        // Can't match anything.
213        return null;
214    }
215}
216