1 | /* validation for the wikipedia=* tag - see tickets #8383, #14425, #18251 */
|
---|
2 |
|
---|
3 | /* If there is no language at all, this is broken. Also catches 'wikipedia' used as 'email', 'website', 'ele' [sic!] ... */
|
---|
4 | *[wikipedia][wikipedia !~ /(?i)^[-a-z]{2,12}:/] {
|
---|
5 | throwError: tr("wikipedia tag has no language given, use ''wikipedia''=''language:page title''");
|
---|
6 | assertMatch: "node wikipedia=Foobar";
|
---|
7 | assertNoMatch: "node wikipedia=en:Foobar";
|
---|
8 | assertNoMatch: "node wikipedia=en-GB:Foobar";
|
---|
9 | }
|
---|
10 |
|
---|
11 | /* Valid languages are extracted from <https://www.wikidata.org/w/api.php?action=sitematrix&formatversion=2>, which may change, so this is a warning only. */
|
---|
12 | *[wikipedia =~ /(?i)^[-a-z]{2,12}:/][wikipedia !~ /^https?:\/\//][wikipedia !~ /^(ab|ace|ady|af|ak|als|am|an|ang|ar|arc|arz|as|ast|atj|av|ay|az|azb|ba|ban|bar|bat-smg|bcl|be|be-x-old|bg|bh|bi|bjn|bm|bn|bo|bpy|br|bs|bug|bxr|ca|cbk-zam|cdo|ce|ceb|ch|chr|chy|ckb|co|cr|crh|cs|csb|cu|cv|cy|da|de|din|diq|dsb|dty|dv|dz|ee|el|eml|en|eo|es|et|eu|ext|fa|ff|fi|fiu-vro|fj|fo|fr|frp|frr|fur|fy|ga|gag|gan|gcr|gd|gl|glk|gn|gom|gor|got|gu|gv|ha|hak|haw|he|hi|hif|hr|hsb|ht|hu|hy|hyw|ia|id|ie|ig|ik|ilo|inh|io|is|it|iu|ja|jam|jbo|jv|ka|kaa|kab|kbd|kbp|kg|ki|kk|kl|km|kn|ko|koi|krc|ks|ksh|ku|kv|kw|ky|la|lad|lb|lbe|lez|lfn|lg|li|lij|lmo|ln|lo|lrc|lt|ltg|lv|mai|map-bms|mdf|mg|mhr|mi|min|mk|ml|mn|mnw|mr|mrj|ms|mt|mwl|my|myv|mzn|na|nah|nap|nds|nds-nl|ne|new|nl|nn|no|nov|nqo|nrm|nso|nv|ny|oc|olo|om|or|os|pa|pag|pam|pap|pcd|pdc|pfl|pi|pih|pl|pms|pnb|pnt|ps|pt|qu|rm|rmy|rn|ro|roa-rup|roa-tara|ru|rue|rw|sa|sah|sat|sc|scn|sco|sd|se|sg|sh|shn|si|simple|sk|sl|sm|sn|so|sq|sr|srn|ss|st|stq|su|sv|sw|szl|szy|ta|tcy|te|tet|tg|th|ti|tk|tl|tn|to|tpi|tr|ts|tt|tum|tw|ty|tyv|udm|ug|uk|ur|uz|ve|vec|vep|vi|vls|vo|wa|war|wo|wuu|xal|xh|xmf|yi|yo|za|zea|zh|zh-classical|zh-min-nan|zh-yue|zu):/] {
|
---|
13 | throwWarning: tr("wikipedia tag has an unknown language prefix");
|
---|
14 | assertMatch: "node wikipedia=X-Y-Z:Foobar";
|
---|
15 | assertNoMatch: "node wikipedia=en:Foobar";
|
---|
16 | }
|
---|
17 |
|
---|
18 | *[wikipedia =~ /^https?:\/\//],
|
---|
19 | *[wikipedia =~ /(?i)^[-a-z]{2,12}:https?:\/\//] {
|
---|
20 | throwWarning: tr("wikipedia tag format is deprecated");
|
---|
21 | suggestAlternative: tr("''wikipedia''=''language:page title''");
|
---|
22 | group: tr("deprecated tagging");
|
---|
23 | assertMatch: "node wikipedia=http://en.wikipedia.org/wiki/OpenStreetMap";
|
---|
24 | assertNoMatch: "node wikipedia=en:OpenStreetMap";
|
---|
25 | }
|
---|
26 |
|
---|
27 | *[wikipedia =~ /^be-x-old:/] {
|
---|
28 | throwWarning: tr("wikipedia ''{0}'' language is obsolete, use ''{1}'' instead", "be-x-old", "be-tarask");
|
---|
29 | fixAdd: concat("wikipedia=be-tarask:", get(regexp_match("^be-x-old:(.+)$", tag("wikipedia")),1));
|
---|
30 | assertMatch: "node wikipedia=be-x-old:foo";
|
---|
31 | assertNoMatch: "node wikipedia=abe-x-old:foo";
|
---|
32 | }
|
---|
33 |
|
---|
34 | *[wikipedia =~ /^cz:/] {
|
---|
35 | throwWarning: tr("wikipedia ''{0}'' language is invalid, use ''{1}'' instead", "cz", "cs");
|
---|
36 | fixAdd: concat("wikipedia=cs:", get(regexp_match("^cz:(.+)$", tag("wikipedia")),1));
|
---|
37 | assertMatch: "node wikipedia=cz:foo";
|
---|
38 | assertNoMatch: "node wikipedia=en:cz:foo";
|
---|
39 | }
|
---|
40 |
|
---|
41 | *[wikimedia_commons =~ /%[0-9A-F][0-9A-F]/] {
|
---|
42 | throwError: tr("{0} tag should not have URL-encoded values like ''%27''", "{0.key}");
|
---|
43 | fixAdd: concat("wikimedia_commons=", trim(replace(URL_decode(tag("wikimedia_commons")), "_", " ")));
|
---|
44 | assertMatch: "node wikimedia_commons=File:Foo%27s";
|
---|
45 | assertNoMatch: "node wikimedia_commons=File:Foo";
|
---|
46 | }
|
---|
47 |
|
---|
48 | *[wikipedia =~ /(?i)^[-a-z]{2,12}:.*%[0-9A-F][0-9A-F]/] {
|
---|
49 | throwError: tr("{0} tag should not have URL-encoded values like ''%27''", "{0.key}");
|
---|
50 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)(.*)$", tag("wikipedia")),1), trim(replace(URL_decode(get(regexp_match("(?i)^([-a-z]+:)(.+)$", tag("wikipedia")),2)), "_", " ")));
|
---|
51 | assertMatch: "node wikipedia=en:Foo%27s";
|
---|
52 | assertNoMatch: "node wikipedia=en:Foo";
|
---|
53 | }
|
---|
54 |
|
---|
55 | *[/^wikipedia:[-a-z]{2,12}$/][/^wikipedia:[-a-z]{2,12}$/ =~ /(?i).*%[0-9A-F][0-9A-F]/] {
|
---|
56 | throwError: tr("{0} tag should not have URL-encoded values like ''%27''", "{0.key}");
|
---|
57 | /* fixAdd: concat("{0.key}", "=", get(regexp_match("(?i)^([-a-z]+:)(.*)$", tag("{0.key}")),1), trim(replace(URL_decode(get(regexp_match("(?i)^([-a-z]+:)(.+)$", tag("{0.key}")),2)), "_", " "))); */
|
---|
58 | /* tag("{0.key}") is not yet supported */
|
---|
59 | assertMatch: "node wikipedia:de=Foo%27s";
|
---|
60 | assertNoMatch: "node wikipedia:de=Foo";
|
---|
61 | }
|
---|
62 |
|
---|
63 | *[wikipedia =~ /(?i)^[-a-z]{2,12}: /] {
|
---|
64 | throwWarning: tr("wikipedia title should not start with a space after language code");
|
---|
65 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)(.*)$", tag("wikipedia")),1), trim(get(regexp_match("(?i)^([-a-z]+:)(.*)$", tag("wikipedia")),2)));
|
---|
66 | assertMatch: "node wikipedia=en: foo";
|
---|
67 | assertNoMatch: "node wikipedia=en:foo";
|
---|
68 | }
|
---|
69 |
|
---|
70 | *[wikipedia =~ /(?i)^[-a-z]{2,12}:wiki\//] {
|
---|
71 | throwWarning: tr("wikipedia title should not have ''{0}'' prefix", "wiki/");
|
---|
72 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)wiki/(.*)$", tag("wikipedia")),1), trim(get(regexp_match("(?i)^([-a-z]+:)wiki/(.*)$", tag("wikipedia")),2)));
|
---|
73 | assertMatch: "node wikipedia=en:wiki/foo";
|
---|
74 | assertNoMatch: "node wikipedia=en:foo";
|
---|
75 | }
|
---|
76 |
|
---|
77 | /* All wikipedias except "jbo" automatically capitalize first letter of the page title.
|
---|
78 | To see the latest list, see <https://noc.wikimedia.org/conf/highlight.php?file=InitialiseSettings.php>
|
---|
79 | and look for 'wgCapitalLinks' setting. */
|
---|
80 | *[wikipedia =~ /^[-a-zA-Z]{2,12}:\p{Ll}/][wikipedia !~ /^jbo:/][wikipedia !~ /(?i)^[-a-z]{2,12}:https?:/] {
|
---|
81 | throwWarning: tr("wikipedia page title should have first letter capitalized");
|
---|
82 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)(.)(.*)$", tag("wikipedia")),1), upper(get(regexp_match("(?i)^([-a-z]+:)(.)(.*)$", tag("wikipedia")),2)), get(regexp_match("(?i)^([-a-z]+:)(.)(.*)$", tag("wikipedia")),3));
|
---|
83 | assertMatch: "node wikipedia=en:foo";
|
---|
84 | assertNoMatch: "node wikipedia=en:Foo";
|
---|
85 | assertMatch: "node wikipedia=ru:абв";
|
---|
86 | assertNoMatch: "node wikipedia=ru:�бв";
|
---|
87 | }
|
---|
88 |
|
---|
89 | *[wikipedia =~ /(?i)^[-a-z]{2,12}:.*_/][wikipedia !~ /(?i)^[-a-z]{2,12}:https?:/] {
|
---|
90 | throwWarning: tr("wikipedia page title should have spaces instead of underscores (''_''→'' '')");
|
---|
91 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)(.+)$", tag("wikipedia")),1), trim(replace(get(regexp_match("(?i)^([-a-z]+:)(.+)$", tag("wikipedia")),2), "_", " ")));
|
---|
92 | assertMatch: "node wikipedia=en:foo_bar";
|
---|
93 | assertNoMatch: "node wikipedia=en:foo bar";
|
---|
94 | }
|
---|
95 |
|
---|
96 | *[wikipedia ^= "da:da:"],
|
---|
97 | *[wikipedia ^= "da:dk:"],
|
---|
98 | *[wikipedia ^= "de:de:"],
|
---|
99 | *[wikipedia ^= "dk:dk:"],
|
---|
100 | *[wikipedia ^= "en:de:"],
|
---|
101 | *[wikipedia ^= "en:en:"],
|
---|
102 | *[wikipedia ^= "en:es:"],
|
---|
103 | *[wikipedia ^= "en:eu:"],
|
---|
104 | *[wikipedia ^= "en:fr:"],
|
---|
105 | *[wikipedia ^= "en:ja:"],
|
---|
106 | *[wikipedia ^= "en:pl:"],
|
---|
107 | *[wikipedia ^= "en:pt:"],
|
---|
108 | *[wikipedia ^= "en:zh:"],
|
---|
109 | *[wikipedia ^= "es:es:"],
|
---|
110 | *[wikipedia ^= "eu:eu:"],
|
---|
111 | *[wikipedia ^= "fr:fr:"],
|
---|
112 | *[wikipedia ^= "ja:ja:"],
|
---|
113 | *[wikipedia ^= "pl:en:"],
|
---|
114 | *[wikipedia ^= "pl:pl:"],
|
---|
115 | *[wikipedia ^= "pt:pt:"],
|
---|
116 | *[wikipedia ^= "ru:fr:"],
|
---|
117 | *[wikipedia ^= "ru:ru:"],
|
---|
118 | *[wikipedia ^= "zh:zh:"] {
|
---|
119 | throwWarning: tr("wikipedia language seems to be duplicated, e.g. en:en:Foo");
|
---|
120 | fixAdd: concat("wikipedia=", get(regexp_match("(?i)^([-a-z]+:)([-a-z]+:)(.*)$", tag("wikipedia")),2), trim(get(regexp_match("(?i)^([-a-z]+:)([-a-z]+:)(.*)$", tag("wikipedia")),3)));
|
---|
121 | assertMatch: "node wikipedia=en:en:Foo";
|
---|
122 | assertMatch: "node wikipedia=en:fr:Foo";
|
---|
123 | assertNoMatch: "node wikipedia=en:Bar";
|
---|
124 | }
|
---|
125 |
|
---|
126 | /* Detect invalid wikidata tags */
|
---|
127 | *[wikidata][wikidata !~ /^Q[1-9][0-9]{0,8}$/] {
|
---|
128 | throwError: tr("wikidata tag must be in Qnnnn format, where n is a digit");
|
---|
129 | assertMatch: "node wikidata=a";
|
---|
130 | assertMatch: "node wikidata=Q";
|
---|
131 | assertMatch: "node wikidata=Q0";
|
---|
132 | assertMatch: "node wikidata=Q0123";
|
---|
133 | assertNoMatch: "node wikidata=Q123";
|
---|
134 | assertNoMatch: "node wikidata=Q1";
|
---|
135 | }
|
---|
136 |
|
---|
137 | /* Wikipedia without wikidata */
|
---|
138 | *[wikipedia][!wikidata] {
|
---|
139 | throwOther: tr("wikipedia tag is set, but there is no wikidata tag. Wikipedia plugin might help with wikidata id lookups");
|
---|
140 | group: tr("missing tag");
|
---|
141 | assertMatch: "node wikipedia=a";
|
---|
142 | assertNoMatch: "node wikipedia=a wikidata=Q123";
|
---|
143 | assertNoMatch: "node wikidata=Q1";
|
---|
144 | assertNoMatch: "node foo=bar";
|
---|
145 | }
|
---|
146 |
|
---|
147 | /* Wikipedia:lang without wikipedia */
|
---|
148 | *[!wikipedia][/^wikipedia:/] {
|
---|
149 | throwWarning: tr("''{0}'' tag is set, but no ''{1}'' tag. Make sure to set ''wikipedia=language:value'' for the main article and optional ''wikipedia:language=value'' only for additional articles that are not just other language variants of the main article.", "{1.key}", "{0.key}");
|
---|
150 | assertMatch: "node wikipedia:en=a";
|
---|
151 | assertNoMatch: "node wikipedia=a wikipedia:en=b";
|
---|
152 | assertNoMatch: "node wikipedia=Foo";
|
---|
153 | }
|
---|