polytypo 1.23.1 → 1.25.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (38) hide show
  1. checksums.yaml +4 -4
  2. data/lib/polytypo/data/VERSION +1 -1
  3. data/lib/polytypo/data/fixtures/bg.json +1 -1
  4. data/lib/polytypo/data/fixtures/cs.json +1 -1
  5. data/lib/polytypo/data/fixtures/de-CH.json +1 -1
  6. data/lib/polytypo/data/fixtures/de-DE.json +1 -1
  7. data/lib/polytypo/data/fixtures/el.json +1 -1
  8. data/lib/polytypo/data/fixtures/en-GB.json +1 -1
  9. data/lib/polytypo/data/fixtures/en-US.json +1 -1
  10. data/lib/polytypo/data/fixtures/es.json +1 -1
  11. data/lib/polytypo/data/fixtures/et.json +266 -0
  12. data/lib/polytypo/data/fixtures/fi.json +1 -1
  13. data/lib/polytypo/data/fixtures/fr-CA.json +1 -1
  14. data/lib/polytypo/data/fixtures/fr.json +1 -1
  15. data/lib/polytypo/data/fixtures/hr.json +1 -1
  16. data/lib/polytypo/data/fixtures/hu.json +1 -1
  17. data/lib/polytypo/data/fixtures/it.json +1 -1
  18. data/lib/polytypo/data/fixtures/locale-resolution.json +13 -1
  19. data/lib/polytypo/data/fixtures/lt.json +1 -1
  20. data/lib/polytypo/data/fixtures/lv.json +290 -0
  21. data/lib/polytypo/data/fixtures/nl.json +1 -1
  22. data/lib/polytypo/data/fixtures/pl.json +1 -1
  23. data/lib/polytypo/data/fixtures/pt-BR.json +1 -1
  24. data/lib/polytypo/data/fixtures/pt-PT.json +1 -1
  25. data/lib/polytypo/data/fixtures/ro.json +1 -1
  26. data/lib/polytypo/data/fixtures/ru.json +1 -1
  27. data/lib/polytypo/data/fixtures/sk.json +1 -1
  28. data/lib/polytypo/data/fixtures/sl.json +1 -1
  29. data/lib/polytypo/data/fixtures/sr.json +1 -1
  30. data/lib/polytypo/data/fixtures/sv.json +1 -1
  31. data/lib/polytypo/data/fixtures/tr.json +1 -1
  32. data/lib/polytypo/data/fixtures/uk.json +1 -1
  33. data/lib/polytypo/data/locales/et.json +85 -0
  34. data/lib/polytypo/data/locales/lv.json +85 -0
  35. data/lib/polytypo/data/locales/registry.json +4 -2
  36. data/lib/polytypo/data/rules/order.json +1 -1
  37. data/lib/polytypo/version.rb +1 -1
  38. metadata +5 -1
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: 8f74fea845cc17b2d5a9a07e0193e5647ff6ee25cc4f0c0d221596ebca73990c
4
- data.tar.gz: 8a2b56d4f013163991580c5998716be7859e927552ff893484079dc946491c8b
3
+ metadata.gz: 0c2d7697d1f8c413a4f75e70258d8c4a03ea73e162b3ec1dfa3ff7f40a1e66d3
4
+ data.tar.gz: aff92a96164654408e9206b201f2c59edf10bc15047d06a57e42529f99504107
5
5
  SHA512:
6
- metadata.gz: bbbfceca0a5597e9e874c90a3585da7477e14f7fe88d77eb47835e1e949e91c5b643de074ba81925266dd9387a7aea20f04acc83e370ecfc0eb2c9dc0d0a2000
7
- data.tar.gz: 1563e1be46c7cf334d7f07913763c42689214c7d72d3cd9e31971c93523282ffbdff8effdc3fba7c77252c94e1a442ad8a46c4a1aff93938abf51ad5ff83ce45
6
+ metadata.gz: a52f59623f9f2b3540fe8961d7750b7bcac554643ed87f27a7e12d86e265788999fe9cbb49edee61030edb91a5bc4c7be6a8a65aa461d85c82dbaf408fec431b
7
+ data.tar.gz: 7cfb5ecbc1d6f7115f4ce516174a86b751704e26e8a3f2077515f4cf3f188b5f6daaf2d747b69c7e9b08a8f9f75e083a02b5f75746f632aef1d284afe753efc8
@@ -1 +1 @@
1
- 1.23.1
1
+ 1.25.0
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "bg",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "cs",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "de-CH",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "de-DE",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "el",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "en-GB",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "en-US",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "es",
4
4
  "cases": [
5
5
  {
@@ -0,0 +1,266 @@
1
+ {
2
+ "spec": "1.25.0",
3
+ "locale": "et",
4
+ "cases": [
5
+ {
6
+ "id": "et-quotes-primary",
7
+ "rule": "quotes",
8
+ "mode": "text",
9
+ "in": "Ta ütles: \"Mul on igav.\" ja läks.",
10
+ "out": "Ta ütles: „Mul on igav.“ ja läks.",
11
+ "note": "EKI, Eesti keele õigekirja põhireeglid R37: low-opening „ and 66-shaped closing “; the full stop of direct speech stays inside."
12
+ },
13
+ {
14
+ "id": "et-quotes-nested-same-pair",
15
+ "rule": "quotes",
16
+ "mode": "text",
17
+ "in": "Ta ütles: \"Õpetaja ütles: \"Ega hinne parane.\" Mis siis?\"",
18
+ "out": "Ta ütles: „Õpetaja ütles: „Ega hinne parane.“ Mis siis?“",
19
+ "note": "Põhireeglid R38 Märkus 1 and R51 Märkus 2: a quotation inside a quotation takes the same pair; the same pair at both levels is the pin."
20
+ },
21
+ {
22
+ "id": "et-quotes-nested-with-words",
23
+ "rule": "quotes",
24
+ "mode": "text",
25
+ "in": "Ta ütles: \"See on 'tsitaat' raamatust\" ja läks.",
26
+ "out": "Ta ütles: „See on „tsitaat“ raamatust“ ja läks.",
27
+ "note": "Põhireeglid R51 Märkus 2 permits a nested quotation; no read source uses single marks, so single straight inner quotes become the same pair (a pin)."
28
+ },
29
+ {
30
+ "id": "et-quotes-already-typeset",
31
+ "rule": "quotes",
32
+ "mode": "text",
33
+ "in": "Ta ütles „jah“ ja läks.",
34
+ "out": "Ta ütles „jah“ ja läks.",
35
+ "note": "Põhireeglid R37: text already carrying U+201E/U+201C is a fixed point."
36
+ },
37
+ {
38
+ "id": "et-quotes-old-closer-input",
39
+ "rule": "quotes",
40
+ "mode": "text",
41
+ "in": "Ta ütles „jah” ja läks.",
42
+ "out": "Ta ütles „jah“ ja läks.",
43
+ "note": "Raadik, Keelenõuanne soovitab 5, p. 227: ÕS 2013 abandoned the „…” form; a U+201D closer typed by the author is normalised to the pinned U+201C."
44
+ },
45
+ {
46
+ "id": "et-quotes-angle-input",
47
+ "rule": "quotes",
48
+ "mode": "text",
49
+ "in": "Ta ütles «jah» ja läks.",
50
+ "out": "Ta ütles „jah“ ja läks.",
51
+ "note": "« » is ranked second by EVS 8:2008 (Raadik, Keelenõuanne soovitab 5, p. 227); it is normalised to the pinned pair."
52
+ },
53
+ {
54
+ "id": "et-quotes-no-inner-space",
55
+ "rule": "quotes",
56
+ "mode": "text",
57
+ "in": "Ta ütles: \" jah \" ja läks.",
58
+ "out": "Ta ütles: „jah“ ja läks.",
59
+ "note": "Põhireeglid examples are closed: innerSpace is none."
60
+ },
61
+ {
62
+ "id": "et-quotes-nest-at-end-declined",
63
+ "rule": "quotes",
64
+ "mode": "text",
65
+ "in": "Ta ütles: \"See on 'tsitaat'\" ja läks.",
66
+ "out": "Ta ütles: \"See on ’tsitaat'\" ja läks.",
67
+ "note": "ACCEPTED COST of the identical pair: a nest whose inner mark touches the outer one is declined (quotes.md §6, §3.2 V1); the inner apostrophe-shaped mark is left to the apostrophe rule."
68
+ },
69
+ {
70
+ "id": "et-quotes-nest-at-start-declined",
71
+ "rule": "quotes",
72
+ "mode": "text",
73
+ "in": "Ta ütles: \"'Tsitaat' on\" ja läks.",
74
+ "out": "Ta ütles: \"'Tsitaat’ on\" ja läks.",
75
+ "note": "ACCEPTED COST of the identical pair: the nest opens against the outer opener, so it is declined (quotes.md §6, §3.2 V1)."
76
+ },
77
+ {
78
+ "id": "et-ellipsis-three-dots",
79
+ "rule": "ellipsis",
80
+ "mode": "text",
81
+ "in": "Oota... ja siis.",
82
+ "out": "Oota… ja siis.",
83
+ "note": "Three dots become the ellipsis character (Põhireeglid R56, mõttepunktid)."
84
+ },
85
+ {
86
+ "id": "et-ellipsis-after-question",
87
+ "rule": "ellipsis",
88
+ "mode": "text",
89
+ "in": "Järsku tuuakse veel midagi?...",
90
+ "out": "Järsku tuuakse veel midagi?…",
91
+ "note": "Raadik, Märgikobarad (Oma Keel 2/2022) p. 54: «huvitavatki?…» keeps the dots after the question mark, so abbreviatedAfterTerminal is false."
92
+ },
93
+ {
94
+ "id": "et-ellipsis-after-exclamation",
95
+ "rule": "ellipsis",
96
+ "mode": "text",
97
+ "in": "Hoidku selle eest!...",
98
+ "out": "Hoidku selle eest!…",
99
+ "note": "Raadik, Märgikobarad p. 55: «Küsi- ja hüüumärk säilivad ning lause lõpus võivad nad olla nii mõttepunktide ees kui ka järel»; p. 56 «kurat!…». No two-dot form after «!» is attested."
100
+ },
101
+ {
102
+ "id": "et-dashes-parenthetical",
103
+ "rule": "dashes",
104
+ "mode": "text",
105
+ "in": "Nüüd - tõsi küll - oleme siin.",
106
+ "out": "Nüüd – tõsi küll – oleme siin.",
107
+ "note": "Mäekivi, Raadik, Keelenõuanne soovitab 4, p. 194; Põhireeglid R53–54: en dash with a space on both sides."
108
+ },
109
+ {
110
+ "id": "et-dashes-em-input",
111
+ "rule": "dashes",
112
+ "mode": "text",
113
+ "in": "Nüüd — tõsi küll — oleme siin.",
114
+ "out": "Nüüd – tõsi küll – oleme siin.",
115
+ "note": "ESA vormistusnõuded: the en dash is the dash in a sentence; an em dash is normalised by this pin (no read source excludes U+2014)."
116
+ },
117
+ {
118
+ "id": "et-dashes-existing-en",
119
+ "rule": "dashes",
120
+ "mode": "text",
121
+ "in": "Nüüd – tõsi küll – oleme siin.",
122
+ "out": "Nüüd – tõsi küll – oleme siin.",
123
+ "note": "An en dash already in place is a fixed point."
124
+ },
125
+ {
126
+ "id": "et-ranges-digits-tight",
127
+ "rule": "ranges",
128
+ "mode": "text",
129
+ "in": "1986-1989 ja 30-35 aastat",
130
+ "out": "1986⁠–⁠1989 ja 30⁠–⁠35 aastat",
131
+ "note": "Mäekivi, Raadik 2008 p. 195: «1986–1989», en dash, no spaces; the en-tight form carries U+2060 either side (ranges.md §3).",
132
+ "rules": {
133
+ "ranges": true
134
+ }
135
+ },
136
+ {
137
+ "id": "et-ranges-already-tight",
138
+ "rule": "ranges",
139
+ "mode": "text",
140
+ "in": "2001–2006",
141
+ "out": "2001–2006",
142
+ "note": "An en dash already between digits is left as it is (ranges.md §3).",
143
+ "rules": {
144
+ "ranges": true
145
+ }
146
+ },
147
+ {
148
+ "id": "et-ranges-word-hyphen-untouched",
149
+ "rule": "ranges",
150
+ "mode": "text",
151
+ "in": "Lääne-Virumaa",
152
+ "out": "Lääne-Virumaa",
153
+ "note": "A hyphen between words is not a range (ranges.md §3).",
154
+ "rules": {
155
+ "ranges": true
156
+ }
157
+ },
158
+ {
159
+ "id": "et-hyphen-noop",
160
+ "rule": "hyphen",
161
+ "mode": "text",
162
+ "in": "Lääne-Virumaa ja e-post",
163
+ "out": "Lääne-Virumaa ja e-post",
164
+ "note": "Põhireeglid R59–61: no closed list of unbreakable forms, so the hyphen rule changes nothing."
165
+ },
166
+ {
167
+ "id": "et-nbsp-spaced-units",
168
+ "rule": "nbsp",
169
+ "mode": "text",
170
+ "in": "Maksab 15 € ja 3 km ja 10 kg ja 25 cm ja 7 ha ja –6 °C",
171
+ "out": "Maksab 15 € ja 3 km ja 10 kg ja 25 cm ja 7 ha ja –6 °C",
172
+ "note": "Põhireeglid R36: «30 g, 15 €, –6 °C»; Rehemaa p. 26: number and unit stay on one line."
173
+ },
174
+ {
175
+ "id": "et-nbsp-single-letter-unit-inert",
176
+ "rule": "nbsp",
177
+ "mode": "text",
178
+ "in": "5 m ja 3 g ja 7 a laps",
179
+ "out": "5 m ja 3 g ja 7 a laps",
180
+ "note": "Single-letter units are excluded by DECISION (they are also common words; nbsp.md §3.7); ACCEPTED COST: «4 m», «7 a» stay unbound."
181
+ },
182
+ {
183
+ "id": "et-nbsp-closed-units",
184
+ "rule": "nbsp",
185
+ "mode": "text",
186
+ "in": "9 % ja 0,4 ‰",
187
+ "out": "9% ja 0,4‰",
188
+ "note": "Põhireeglid R36 Märkus 2: «9%, 0,4‰» are written closed; closedUnits deletes the space."
189
+ },
190
+ {
191
+ "id": "et-nbsp-spaced-unit-inserted",
192
+ "rule": "nbsp",
193
+ "mode": "text",
194
+ "in": "Hind 15€ siin",
195
+ "out": "Hind 15 € siin",
196
+ "note": "nbsp.md §3.7: the unit is in spacedUnits, so the missing space is inserted as U+00A0."
197
+ },
198
+ {
199
+ "id": "et-nbsp-abbreviations-inert",
200
+ "rule": "nbsp",
201
+ "mode": "text",
202
+ "in": "See on k.a ja v.a ja nt ja jne",
203
+ "out": "See on k.a ja v.a ja nt ja jne",
204
+ "note": "Mäekivi, Raadik p. 193 and Põhireeglid R63 Märkus 1: Estonian abbreviations carry no internal space, so abbreviations is empty and nothing is bound."
205
+ },
206
+ {
207
+ "id": "et-nbsp-before-number-inert",
208
+ "rule": "nbsp",
209
+ "mode": "text",
210
+ "in": "määrus nr 144 ja vt lk 30",
211
+ "out": "määrus nr 144 ja vt lk 30",
212
+ "note": "The space is attested (Mäekivi, Raadik p. 192; Rehemaa p. 28) but no read source makes it non-breaking, so beforeNumber is empty."
213
+ },
214
+ {
215
+ "id": "et-nbsp-initials-inert",
216
+ "rule": "nbsp",
217
+ "mode": "text",
218
+ "in": "F. R. Kreutzwald ja A. Värnik",
219
+ "out": "F. R. Kreutzwald ja A. Värnik",
220
+ "note": "Mäekivi, Raadik p. 192 attests the space after an initial but not its non-breaking form, so initialBinding is none."
221
+ },
222
+ {
223
+ "id": "et-spaces-before-punctuation",
224
+ "rule": "spaces",
225
+ "mode": "text",
226
+ "in": "Mis siis ? Ei tea !",
227
+ "out": "Mis siis? Ei tea!",
228
+ "note": "Mäekivi, Raadik, Keelenõuanne soovitab 4, p. 191: «Koma, punkti, kooloni, semikooloni, küsimärgi ja hüüumärgi ette lauses tühikut ei panda»; nbsp.beforePunctuation is empty, so the space is removed, not bound."
229
+ },
230
+ {
231
+ "id": "et-spaces-brackets",
232
+ "rule": "spaces",
233
+ "mode": "text",
234
+ "in": "Ta ( vaikselt ) läks .",
235
+ "out": "Ta (vaikselt) läks.",
236
+ "note": "Spaces inside brackets and before a full stop are removed."
237
+ },
238
+ {
239
+ "id": "et-symbols-copyright",
240
+ "rule": "symbols",
241
+ "mode": "text",
242
+ "in": "(c) 2026",
243
+ "out": "© 2026",
244
+ "note": "(c) becomes the copyright sign."
245
+ },
246
+ {
247
+ "id": "et-apostrophe-case-ending",
248
+ "rule": "apostrophe",
249
+ "mode": "text",
250
+ "in": "Youtube'ist ja tõest'",
251
+ "out": "Youtube’ist ja tõest’",
252
+ "note": "Põhireeglid R57 and Märkus 4: a straight apostrophe becomes U+2019."
253
+ },
254
+ {
255
+ "id": "et-quote-punctuation-none",
256
+ "rule": "quote-punctuation",
257
+ "mode": "text",
258
+ "in": "alfa „beta“, gamma",
259
+ "out": "alfa „beta“, gamma",
260
+ "note": "Põhireeglid R37 vs R38; Raadik, «Kuidas vormistada tsitaati?», KS5 p. 218: placement depends on meaning, so punctuationPlacement is none and the rule leaves the text alone.",
261
+ "rules": {
262
+ "quote-punctuation": true
263
+ }
264
+ }
265
+ ]
266
+ }
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "fi",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "fr-CA",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "fr",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "hr",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "hu",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "it",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "cases": [
4
4
  {
5
5
  "id": "exact-en-us",
@@ -288,6 +288,18 @@
288
288
  "tag": "tr-TR",
289
289
  "resolves": "tr",
290
290
  "note": "section 3.4 step 3a, and the tag a Turkish caller actually sends. tr needs no registry alias precisely because the region strip reaches it — this case is what makes that claim testable rather than asserted."
291
+ },
292
+ {
293
+ "id": "strip-lv-lv",
294
+ "tag": "lv-LV",
295
+ "resolves": "lv",
296
+ "note": "section 3.4 step 3a: the region strip reaches lv, which needs no registry alias."
297
+ },
298
+ {
299
+ "id": "strip-et-ee",
300
+ "tag": "et-EE",
301
+ "resolves": "et",
302
+ "note": "section 3.4 step 3a: the region strip reaches et, which needs no registry alias."
291
303
  }
292
304
  ]
293
305
  }
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "lt",
4
4
  "cases": [
5
5
  {
@@ -0,0 +1,290 @@
1
+ {
2
+ "spec": "1.25.0",
3
+ "locale": "lv",
4
+ "cases": [
5
+ {
6
+ "id": "lv-quotes-primary",
7
+ "rule": "quotes",
8
+ "mode": "text",
9
+ "in": "Viņš teica: \"jā\" un aizgāja.",
10
+ "out": "Viņš teica: “jā” un aizgāja.",
11
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) 7.2.3, p. 87: round marks, opener high U+201C, closer U+201D, no inner space (pinned pair; LVEK 14.05.2014 also permits a low opener but names no closer)."
12
+ },
13
+ {
14
+ "id": "lv-quotes-nested-same-pair",
15
+ "rule": "quotes",
16
+ "mode": "text",
17
+ "in": "Viņš teica: \"Es zinu, ka \"Pīlādzis\" ir laba grāmata.\"",
18
+ "out": "Viņš teica: “Es zinu, ka “Pīlādzis” ir laba grāmata.”",
19
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) I pielikums p. 117: a quotation or title inside a quotation takes double marks («dubultpēdiņas»); the same pair at both levels is the pin, not the source."
20
+ },
21
+ {
22
+ "id": "lv-quotes-nested-with-words",
23
+ "rule": "quotes",
24
+ "mode": "text",
25
+ "in": "Viņš teica: \"Tā ir 'citāta' no grāmatas\" un aizgāja.",
26
+ "out": "Viņš teica: “Tā ir “citāta” no grāmatas” un aizgāja.",
27
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 117: no single-mark nesting level; single straight inner quotes become the same U+201C/U+201D pair."
28
+ },
29
+ {
30
+ "id": "lv-quotes-before-question-mark",
31
+ "rule": "quotes",
32
+ "mode": "text",
33
+ "in": "Vai tiešām? \"Jā!\" viņš teica.",
34
+ "out": "Vai tiešām? “Jā!” viņš teica.",
35
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 117: a closing mark follows the sentence mark flush; innerSpace is none."
36
+ },
37
+ {
38
+ "id": "lv-quotes-already-typeset",
39
+ "rule": "quotes",
40
+ "mode": "text",
41
+ "in": "Viņš teica “jā” un aizgāja.",
42
+ "out": "Viņš teica “jā” un aizgāja.",
43
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 87: text already carrying U+201C/U+201D is a fixed point."
44
+ },
45
+ {
46
+ "id": "lv-quotes-low-open-input",
47
+ "rule": "quotes",
48
+ "mode": "text",
49
+ "in": "Viņš teica „jā” un aizgāja.",
50
+ "out": "Viņš teica “jā” un aizgāja.",
51
+ "note": "A low-opening pair typed by the author is normalised to the pinned pair; LVEK 14.05.2014 permits either opener but names no closer, so one pair is pinned."
52
+ },
53
+ {
54
+ "id": "lv-quotes-angle-input",
55
+ "rule": "quotes",
56
+ "mode": "text",
57
+ "in": "Viņš teica «jā» un aizgāja.",
58
+ "out": "Viņš teica “jā” un aizgāja.",
59
+ "note": "Angle quotes are attested only through Blinkena (2009) p. 398 as quoted in VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) fn. 50; they are normalised to the pinned pair."
60
+ },
61
+ {
62
+ "id": "lv-quotes-no-inner-space",
63
+ "rule": "quotes",
64
+ "mode": "text",
65
+ "in": "Viņš teica: \" jā \" un aizgāja.",
66
+ "out": "Viņš teica: “jā” un aizgāja.",
67
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 117: «Starp pēdiņām un tajās iekļauto tekstu atstarpes neliek.»"
68
+ },
69
+ {
70
+ "id": "lv-quotes-nest-at-end-declined",
71
+ "rule": "quotes",
72
+ "mode": "text",
73
+ "in": "Viņš teica: \"Tā ir 'citāta'\" un aizgāja.",
74
+ "out": "Viņš teica: \"Tā ir ’citāta'\" un aizgāja.",
75
+ "note": "ACCEPTED COST of the identical pair: a nest whose inner mark touches the outer one is declined (quotes.md §6, §3.2 V1); the inner apostrophe-shaped mark is left to the apostrophe rule."
76
+ },
77
+ {
78
+ "id": "lv-quotes-nest-at-start-declined",
79
+ "rule": "quotes",
80
+ "mode": "text",
81
+ "in": "Viņš teica: \"'Pīlādzis' ir\" un aizgāja.",
82
+ "out": "Viņš teica: \"'Pīlādzis’ ir\" un aizgāja.",
83
+ "note": "ACCEPTED COST of the identical pair: the nest opens against the outer opener, so it is declined (quotes.md §6, §3.2 V1)."
84
+ },
85
+ {
86
+ "id": "lv-ellipsis-three-dots",
87
+ "rule": "ellipsis",
88
+ "mode": "text",
89
+ "in": "Pagaidi... un tad.",
90
+ "out": "Pagaidi… un tad.",
91
+ "note": "Three dots become the ellipsis character."
92
+ },
93
+ {
94
+ "id": "lv-ellipsis-after-question",
95
+ "rule": "ellipsis",
96
+ "mode": "text",
97
+ "in": "Ko gan viņš pateiks?...",
98
+ "out": "Ko gan viņš pateiks?…",
99
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) I pielikums p. 122: «?...» keeps three dots, so abbreviatedAfterTerminal is false and the dots become the ellipsis character after the mark."
100
+ },
101
+ {
102
+ "id": "lv-ellipsis-after-exclamation",
103
+ "rule": "ellipsis",
104
+ "mode": "text",
105
+ "in": "Neesi mierā!...",
106
+ "out": "Neesi mierā!…",
107
+ "note": "abbreviatedAfterTerminal is false: no two-dot form after «!» is attested."
108
+ },
109
+ {
110
+ "id": "lv-dashes-parenthetical",
111
+ "rule": "dashes",
112
+ "mode": "text",
113
+ "in": "Viņš teica - protams - ka nāks.",
114
+ "out": "Viņš teica – protams – ka nāks.",
115
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) pp. 86, 114: en dash with a space on both sides."
116
+ },
117
+ {
118
+ "id": "lv-dashes-em-input",
119
+ "rule": "dashes",
120
+ "mode": "text",
121
+ "in": "Tas — protams — labi.",
122
+ "out": "Tas – protams – labi.",
123
+ "note": "LVEK 10.09.2014 prot. Nr. 36 5. §: «latviešu valodā nelieto garo jeb dubulto domuzīmi (—)»; an em dash is normalised to the en dash."
124
+ },
125
+ {
126
+ "id": "lv-dashes-existing-en",
127
+ "rule": "dashes",
128
+ "mode": "text",
129
+ "in": "Dēls – meita atnāca.",
130
+ "out": "Dēls – meita atnāca.",
131
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 86: an en dash already in place is a fixed point."
132
+ },
133
+ {
134
+ "id": "lv-ranges-digits-tight",
135
+ "rule": "ranges",
136
+ "mode": "text",
137
+ "in": "1999-2000 un 30-35 gadi",
138
+ "out": "1999⁠–⁠2000 un 30⁠–⁠35 gadi",
139
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 86: «30–35 gadi», en dash, no spaces; the en-tight form carries U+2060 either side (ranges.md §3).",
140
+ "rules": {
141
+ "ranges": true
142
+ }
143
+ },
144
+ {
145
+ "id": "lv-ranges-already-tight",
146
+ "rule": "ranges",
147
+ "mode": "text",
148
+ "in": "2001–2006",
149
+ "out": "2001–2006",
150
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 114: «(1899–1915)» is a fixed point (ranges.md §3).",
151
+ "rules": {
152
+ "ranges": true
153
+ }
154
+ },
155
+ {
156
+ "id": "lv-ranges-word-hyphen-untouched",
157
+ "rule": "ranges",
158
+ "mode": "text",
159
+ "in": "jūnijs-augusts",
160
+ "out": "jūnijs-augusts",
161
+ "note": "A hyphen between words is not a range (ranges.md §2).",
162
+ "rules": {
163
+ "ranges": true
164
+ }
165
+ },
166
+ {
167
+ "id": "lv-hyphen-noop",
168
+ "rule": "hyphen",
169
+ "mode": "text",
170
+ "in": "Rīga-Jūrmala",
171
+ "out": "Rīga-Jūrmala",
172
+ "note": "The hyphen lists are empty: no closed list of unbreakable hyphenated forms is attested."
173
+ },
174
+ {
175
+ "id": "lv-nbsp-units",
176
+ "rule": "nbsp",
177
+ "mode": "text",
178
+ "in": "10 % un 21 € un 5 EUR un 3 km un -8 °C",
179
+ "out": "10 % un 21 € un 5 EUR un 3 km un -8 °C",
180
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89: «5 km, 25 t, 10 %, 21 €, –8 °C»; LVEK 12.03.2014 prot. Nr. 33 2. §: currency goes after the sum with a bound space."
181
+ },
182
+ {
183
+ "id": "lv-nbsp-single-letter-unit-inert",
184
+ "rule": "nbsp",
185
+ "mode": "text",
186
+ "in": "5 m un 3 g un 25 t",
187
+ "out": "5 m un 3 g un 25 t",
188
+ "note": "Single-letter m, g and t are not listed by DECISION, although «25 t» is attested on p. 89: they are also common words."
189
+ },
190
+ {
191
+ "id": "lv-nbsp-abbreviations",
192
+ "rule": "nbsp",
193
+ "mode": "text",
194
+ "in": "Tas ir, piemēram, u. c. un u. tml. un š. g. rudenī",
195
+ "out": "Tas ir, piemēram, u. c. un u. tml. un š. g. rudenī",
196
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89: «dažos saīsinājumos (u. c., u. tml., š. g.)»."
197
+ },
198
+ {
199
+ "id": "lv-nbsp-before-number",
200
+ "rule": "nbsp",
201
+ "mode": "text",
202
+ "in": "Nr. 20 un nr. 5",
203
+ "out": "Nr. 20 un nr. 5",
204
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89: «Nr. 20» binds; the only lowercase example of «nr.» stands after the number («155. nr.»), so it is not listed and stays as typed."
205
+ },
206
+ {
207
+ "id": "lv-nbsp-initial-binds",
208
+ "rule": "nbsp",
209
+ "mode": "text",
210
+ "in": "J. Kalniņš un A. Bērziņš",
211
+ "out": "J. Kalniņš un A. Bērziņš",
212
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89 «starp iniciāli un uzvārdu (J. Kalniņš)»; LVEK 10.09.2014 prot. Nr. 36 5. §: a bound space after an initial (initialBinding single)."
213
+ },
214
+ {
215
+ "id": "lv-nbsp-initials-chain",
216
+ "rule": "nbsp",
217
+ "mode": "text",
218
+ "in": "J. K. Bērziņš atnāca.",
219
+ "out": "J. K. Bērziņš atnāca.",
220
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89; LVEK prot. Nr. 36 5. §: the sources state the single-initial rule; binding within a run of initials is nbsp.md §3.9 behaviour."
221
+ },
222
+ {
223
+ "id": "lv-nbsp-initial-sentence-boundary",
224
+ "rule": "nbsp",
225
+ "mode": "text",
226
+ "in": "Es izvēlējos plānu B. Tad aizgāju.",
227
+ "out": "Es izvēlējos plānu B. Tad aizgāju.",
228
+ "note": "ACCEPTED COST of initialBinding = single: a lone capital before a full stop is read as an initial even at a sentence boundary (nbsp.md §3.9, §7.9a), the price of binding «J. // Kalniņš»."
229
+ },
230
+ {
231
+ "id": "lv-nbsp-short-words-inert",
232
+ "rule": "nbsp",
233
+ "mode": "text",
234
+ "in": "uz mājām un ar draugu",
235
+ "out": "uz mājām un ar draugu",
236
+ "note": "afterShortWords is empty: no source binds un, ar, uz or no."
237
+ },
238
+ {
239
+ "id": "lv-nbsp-bare-degree-closed",
240
+ "rule": "nbsp",
241
+ "mode": "text",
242
+ "in": "15°",
243
+ "out": "15°",
244
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 89 fn. 54: the bare degree sign is closed up."
245
+ },
246
+ {
247
+ "id": "lv-nbsp-spaced-unit-inserted",
248
+ "rule": "nbsp",
249
+ "mode": "text",
250
+ "in": "Value 50% here",
251
+ "out": "Value 50 % here",
252
+ "note": "spec 1.22.0, nbsp.md §3.7: the unit is in spacedUnits, so the missing space is inserted as U+00A0."
253
+ },
254
+ {
255
+ "id": "lv-spaces-brackets",
256
+ "rule": "spaces",
257
+ "mode": "text",
258
+ "in": "Viņš ( klusi ) aizgāja .",
259
+ "out": "Viņš (klusi) aizgāja.",
260
+ "note": "Spaces inside brackets and before a full stop are removed."
261
+ },
262
+ {
263
+ "id": "lv-symbols-copyright",
264
+ "rule": "symbols",
265
+ "mode": "text",
266
+ "in": "(c) 2026",
267
+ "out": "© 2026",
268
+ "note": "(c) becomes the copyright sign."
269
+ },
270
+ {
271
+ "id": "lv-apostrophe-foreign-name",
272
+ "rule": "apostrophe",
273
+ "mode": "text",
274
+ "in": "O'Brien atnāca.",
275
+ "out": "O’Brien atnāca.",
276
+ "note": "A straight apostrophe in a name becomes U+2019."
277
+ },
278
+ {
279
+ "id": "lv-quote-punctuation-none",
280
+ "rule": "quote-punctuation",
281
+ "mode": "text",
282
+ "in": "alfa “beta”, gamma",
283
+ "out": "alfa “beta”, gamma",
284
+ "note": "VVC, Tiesību aktu tulkošanas rokasgrāmata (2021) p. 117: placement depends on context, so punctuationPlacement is none and the rule leaves the text alone.",
285
+ "rules": {
286
+ "quote-punctuation": true
287
+ }
288
+ }
289
+ ]
290
+ }
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "nl",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "pl",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "pt-BR",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "pt-PT",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "ro",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "ru",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "sk",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "sl",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "sr",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "sv",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "tr",
4
4
  "cases": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "locale": "uk",
4
4
  "cases": [
5
5
  {
@@ -0,0 +1,85 @@
1
+ {
2
+ "locale": "et",
3
+ "name": "Estonian",
4
+ "quotes": {
5
+ "primary": {
6
+ "open": "„",
7
+ "close": "“",
8
+ "innerSpace": "none"
9
+ },
10
+ "secondary": {
11
+ "open": "„",
12
+ "close": "“",
13
+ "innerSpace": "none"
14
+ },
15
+ "punctuationPlacement": "none",
16
+ "elisionIdioms": [],
17
+ "elisionClitics": {
18
+ "before": [],
19
+ "after": []
20
+ }
21
+ },
22
+ "dash": {
23
+ "parenthetical": "en-spaced",
24
+ "range": "en-tight"
25
+ },
26
+ "ellipsis": {
27
+ "abbreviatedAfterTerminal": false
28
+ },
29
+ "hyphen": {
30
+ "prefixes": [],
31
+ "suffixes": [],
32
+ "compounds": []
33
+ },
34
+ "nbsp": {
35
+ "beforePunctuation": [],
36
+ "narrowBeforePunctuation": [],
37
+ "afterShortWords": [],
38
+ "abbreviations": [],
39
+ "beforeUnits": [],
40
+ "spacedUnits": ["€", "km", "kg", "cm", "ha", "°C"],
41
+ "closedUnits": ["%", "‰"],
42
+ "beforeNumber": [],
43
+ "beforeWord": [],
44
+ "afterSymbols": [],
45
+ "initialBinding": "none"
46
+ },
47
+ "sources": [
48
+ {
49
+ "rule": "quotes",
50
+ "cite": "Eesti Keele Instituut, Eesti keele õigekirja põhireeglid (2025; heaks kiidetud Emakeele Seltsi keeletoimkonnas 6.03.2024), REEGEL 37, 38 Märkus 1, 51 Märkus 2; M. Raadik, «Mis kujuga on jutumärgid?», Keelenõuanne soovitab 5 (2015), lk 227 (= Oma Keel 2014, nr 2): «Selle sihiga on 2013. a ÕS loobunud 2006. a valitud märkidest [„ ”] ning rakendanud kuju [„ “], millel on eesti keeles kõige tugevam traditsioon»; Põhireeglid R37: «Mari ütles: „Mul on igav.“»; R51 Märkus 2: «Ühtede jutumärkide sees võivad olla ka teised»",
51
+ "url": "https://eki.ee/wp-content/uploads/2025/08/Eesti-keele-oigekirja-pohireeglid.pdf",
52
+ "note": "PICK-AND-PIN of the code points: U+201E…U+201C (low-9 opener, 66-shaped closer). The closer U+201D belongs to the ÕS 2006 form that ÕS 2013 abandoned (Raadik, KS5 p. 227); the issue's „…” lead is therefore refuted. The code points were read from glyph shapes and the verbal description «„99 all ja 66 üleval“» (Emakeele Seltsi aastaraamatu vormistusnõuded, 20.06.2026, journal house style, corroboration only), not from a text layer of the Põhireeglid PDF (it has none); the text layers of KS5 pp. 213–218 and 226–227 and of Oma Keel 2/2022 carry U+201E…U+201C (the ÕS 2006 pair is U+201E…U+201D). Secondary equals primary: R51 Märkus 2 permits («võivad») a quotation inside a quotation and R38 Märkus 1 shows «„Õpetaja ütles: „Ega hinne … parandada.““»; no read source uses single marks for it, so this is a pin, not a stated requirement. ACCEPTED COST of the identical pair: a nest whose inner mark touches the outer one is declined (quotes.md §6, §3.2 V1), pinned by fixtures. « » is ranked second by EVS 8:2008 as reported by Raadik (not read) and is normalised to the pinned pair. innerSpace none: every Põhireeglid example is closed up; no source states the rule in words. Märgikobarad (Oma Keel 2/2022) p. 60, «Kõrvutiste jutumärkide vahele üldiselt tühikut ei panda», concerns two adjacent marks, which supports the nested pair standing flush. punctuationPlacement none: the full stop stands inside for direct speech (R37) and is dropped for a quotation inside a sentence (R38; M. Raadik, «Kuidas vormistada tsitaati?», KS5 p. 218), so placement depends on meaning. elisionIdioms and elisionClitics empty: apostrophe elision (R57 Märkus 4: tõest’, a’al, on’s) and case-ending apostrophes (Youtube’ist) are open classes. The Eesti keele käsiraamat was not reachable (both URLs returned 404) and is not cited. Retrieved 2026-11-14."
53
+ },
54
+ {
55
+ "rule": "dashes",
56
+ "cite": "H. Mäekivi, M. Raadik, «See tühine tühik», Keelenõuanne soovitab 4 (2008), lk 194: «Kõigil neil juhtudel kirjutatakse mõttekriips mõlemalt poolt lahku (kuid näiteks järgneva komaga kokku)»; Põhireeglid REEGEL 53–54",
57
+ "url": "https://keeleabi.eki.ee/pdf/191.pdf",
58
+ "note": "dash.parenthetical = en-spaced (U+0020 U+2013 U+0020). Spacing is stated by Mäekivi and Raadik; the width U+2013 is read from the REEGEL 53–54 examples («varuosad – kõik», «Nüüd – tõsi küll – oleme») and corroborated by the Emakeele Seltsi vormistusnõuded: «poolpikk, nn enn-kriips (–) … Sama poolpikka kriipsu kasutatakse ka lauses mõttekriipsuna». No read Estonian source states an exclusion of U+2014, so an em dash is normalised to the en dash by this pin, not by a citation. Retrieved 2026-11-14."
59
+ },
60
+ {
61
+ "rule": "ranges",
62
+ "cite": "H. Mäekivi, M. Raadik, «See tühine tühik», Keelenõuanne soovitab 4 (2008), lk 195: «Sõna kuni asendav mõttekriips enda ette ega taha üldjuhul tühikut ei nõua: … 75–100 mm», «1986–1989»; T. Rehemaa, Arvukirjutus (2. tr, 2020), lk 38, 53",
63
+ "url": "https://keeleabi.eki.ee/pdf/191.pdf",
64
+ "note": "dash.range = en-tight (U+2013, no spaces). A spaced form is permitted when a number itself contains a space («8000 – 10 000 eurot ~ 8000–10 000 eurot», Rehemaa p. 38); the closed form is pinned. The en dash stands for «kuni» only; a hyphen between consecutive ordinals means «ja» («1.-2. jaanuar», R60), and dotted flanks are not range candidates (an algorithm limit, not data). The ranges rule is off by default. Retrieved 2026-11-14."
65
+ },
66
+ {
67
+ "rule": "ellipsis",
68
+ "cite": "M. Raadik, «Märgikobarad», Oma Keel 2/2022, lk 54, 60: «Küll säilivad mõttepunktide kõrval küsi- ja hüüumärk … Järsku tuuakse veel midagi huvitavatki?…»; Põhireeglid REEGEL 56",
69
+ "url": "https://www.emakeeleselts.ee/wp-content/uploads/2022/12/raadik_OK_2022-2.pdf",
70
+ "note": "abbreviatedAfterTerminal = false: the dots stay after ? and ! and no read source reduces them to two. Estonian philological omission marks («katkestuspunktid», two or four dots; Mäekivi and Raadik 2008 p. 198) are a different mark; the ellipsis rule collapses a four-dot run to the ellipsis character, an accepted cost for such texts. Retrieved 2026-11-14."
71
+ },
72
+ {
73
+ "rule": "hyphen",
74
+ "cite": "Põhireeglid REEGEL 17 (poolitamine), 59–61 (sidekriips)",
75
+ "url": "https://eki.ee/wp-content/uploads/2025/08/Eesti-keele-oigekirja-pohireeglid.pdf",
76
+ "note": "All three lists are empty: hyphenation and the uses of the hyphen are described, but no closed list of unbreakable hyphenated forms is given. Retrieved 2026-11-14."
77
+ },
78
+ {
79
+ "rule": "nbsp",
80
+ "cite": "Põhireeglid REEGEL 36 ja Märkus 2: «eraldatakse lühend või tähis numbriga kirjutatud arvust tühikuga (nt 30 g, 15 €, –6 °C, 10 m² tuba, 7 a laps, § 5 ~ 5. §)», «Arvuga kokku kirjutatakse kraadimärk, … protsendi-, promillimärk (nt 4° sooja, … 9%, 0,4‰)»; T. Rehemaa, Arvukirjutus (2. tr, 2020), lk 26: «Arv ja lühend tuleb mahutada ühele reale. Poolitamatu tühiku saab Wordi programmis, kui valida Ctrl + Shift + tühikuklahv.», lk 27: «15 °C (mitte: „15° C“)»; Mäekivi, Raadik 2008 lk 191, 193, 204",
81
+ "url": "https://eki.ee/wp-content/uploads/2025/08/Eesti-keele-oigekirja-pohireeglid.pdf",
82
+ "note": "(1) beforePunctuation and narrowBeforePunctuation empty: a stated negative, «Koma, punkti, kooloni, semikooloni, küsimärgi ja hüüumärgi ette lauses tühikut ei panda» (Mäekivi, Raadik p. 191). (2) spacedUnits = € km kg cm ha °C: R36 (€, °C) and Rehemaa pp. 26, 30, 44, 46 (kg, ha, cm, km); the one-line rule keeps number and unit together (Rehemaa p. 26). Not listed though attested: °F, Hz, dl, px (by DECISION, no closed set is stated), and EUR (not attested). Single-letter units (m, g, l, t, s, a) are excluded because they are also common words and nbsp.md §3.7 forbids them; ACCEPTED COST: Mäekivi and Raadik's own «4 m» and «2005. a» (p. 204) stay unbound. (3) closedUnits = % and ‰ (R36 Märkus 2). ACCEPTED COST: Rehemaa p. 27 tolerates a space before % in specialised texts, which N5 will delete. A bare ° is in no list in any locale (nbsp.md §3.7). (4) abbreviations empty: Estonian lowercase abbreviations are written without internal spaces (k.a, s.o, v.a, e.m.a, R63 Märkus 1; Mäekivi, Raadik p. 193), so there is no space to bind; the issue's jne, nt, vt lead is refuted. Foreign-origin abbreviations do carry one («a. D.», «P. S.», p. 193) but no source makes it non-breaking, so they are not listed: ACCEPTED COST. (5) beforeNumber, afterSymbols and initialBinding left empty or none: the space is attested («määrus nr 144» (Mäekivi, Raadik p. 192), «vt lk 30–33» (Rehemaa p. 28), «§ 5 ~ 5. §» (R36; Mäekivi, Raadik p. 199), «A. Värnik, … F. R. Kreutzwald» (Mäekivi, Raadik p. 192)) but no read source says it is non-breaking, and unverified lists ship empty. (6) afterShortWords, beforeUnits and beforeWord empty: not addressed. Not expressible and so not covered: digit groups («10 000», Mäekivi, Raadik p. 204), «2005. a», «5. §». Retrieved 2026-11-14."
83
+ }
84
+ ]
85
+ }
@@ -0,0 +1,85 @@
1
+ {
2
+ "locale": "lv",
3
+ "name": "Latvian",
4
+ "quotes": {
5
+ "primary": {
6
+ "open": "“",
7
+ "close": "”",
8
+ "innerSpace": "none"
9
+ },
10
+ "secondary": {
11
+ "open": "“",
12
+ "close": "”",
13
+ "innerSpace": "none"
14
+ },
15
+ "punctuationPlacement": "none",
16
+ "elisionIdioms": [],
17
+ "elisionClitics": {
18
+ "before": [],
19
+ "after": []
20
+ }
21
+ },
22
+ "dash": {
23
+ "parenthetical": "en-spaced",
24
+ "range": "en-tight"
25
+ },
26
+ "ellipsis": {
27
+ "abbreviatedAfterTerminal": false
28
+ },
29
+ "hyphen": {
30
+ "prefixes": [],
31
+ "suffixes": [],
32
+ "compounds": []
33
+ },
34
+ "nbsp": {
35
+ "beforePunctuation": [],
36
+ "narrowBeforePunctuation": [],
37
+ "afterShortWords": [],
38
+ "abbreviations": ["u. c.", "u. tml.", "š. g."],
39
+ "beforeUnits": [],
40
+ "spacedUnits": ["%", "€", "EUR", "km", "°C"],
41
+ "closedUnits": [],
42
+ "beforeNumber": ["Nr."],
43
+ "beforeWord": [],
44
+ "afterSymbols": [],
45
+ "initialBinding": "single"
46
+ },
47
+ "sources": [
48
+ {
49
+ "rule": "quotes",
50
+ "cite": "Valsts valodas centrs, Latviešu valodas ekspertu komisija, «Svarīgākie lēmumi 2014. gadā», 14.05.2014, protokols Nr. 34 3. §: «no valodas prakses viedokļa ir pareizs atverošo pēdiņu novietojums gan apakšā („) gan augšā (“), tomēr vienā tekstā konsekventi lietojams viens variants»; Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021), 7.2.3 «Pēdiņas», p. 87: «centra tulkojumos lietojam apaļās pēdiņas (“”) … Centra tulkojumu praksē atvērējpēdiņas liekam augšā»; I pielikums p. 117: «Ja tiešajā runā vai citātā ir iekļauts cita runātāja teksts, cits citāts vai nosaukums (virsraksts), lietojamas dubultpēdiņas.»; «Starp pēdiņām un tajās iekļauto tekstu atstarpes neliek.»; «Ja pēdiņās ietvertā daļa ir teikuma (vai frāzes) beigās, pieturzīmi liek aiz pēdiņām»",
51
+ "url": "https://vvc.gov.lv/lv/media/149/download",
52
+ "note": "PICK-AND-PIN. LVEK accepts opening U+201E or U+201C in one text and names no closing mark, so a low-opening pair has no attested closer. The only complete pair attested is the VVC house style U+201C…U+201D (code points read from the PDF text layer of the handbook, HIGH); it is pinned, not claimed as the only legal Latvian form. « » is attested only through Blinkena (2009) p. 398, as quoted in handbook fn. 50; the book was not read. Secondary equals primary (p. 117: double marks for a quotation inside a quotation; single marks ‘’ only for cultivar names and glosses, p. 88). ACCEPTED COST of the identical pair: a nest whose inner mark touches the outer one is declined (quotes.md §6, §3.2 V1), pinned by fixtures. innerSpace none for both. punctuationPlacement none: placement depends on context (p. 117, p. 87). elisionIdioms and elisionClitics empty: apostrophe elision exists (p. 121) but no closed set. Handbook: https://vvc.gov.lv/lv/media/62/download (house style, SECONDARY). Retrieved 2026-11-14."
53
+ },
54
+ {
55
+ "rule": "dashes",
56
+ "cite": "Valsts valodas centrs, LVEK, «Svarīgākie lēmumi 2014. gadā», 10.09.2014, protokols Nr. 36 5. §: «latviešu valodā nelieto garo jeb dubulto domuzīmi (—)»; Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021), p. 86: «domuzīmi (–) (Alt + 0150)», «liekot to savrupinātās daļas abās pusēs»; p. 114: «Domuzīmi no pārējā teksta atdala ar atstarpēm abās pusēs.»",
57
+ "url": "https://vvc.gov.lv/lv/media/149/download",
58
+ "note": "dash.parenthetical = en-spaced (U+0020 U+2013 U+0020). The exclusion of U+2014 is NORMATIVE (LVEK); width U+2013 from Alt 0150 and the spacing from the VVC handbook (SECONDARY, HIGH). Retrieved 2026-11-14. Handbook: https://vvc.gov.lv/lv/media/62/download."
59
+ },
60
+ {
61
+ "rule": "ranges",
62
+ "cite": "Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021), p. 86: «Prievārda “līdz” vietā (ja saistāmie jēdzieni ir rakstīti ar cipariem) izmanto vienotājdomuzīmi, un to ar atstarpēm neatdala. 2005.–2007. gada plāns; viņa vecums varēja būt 30–35 gadi»; I pielikums p. 114: «(1899–1915)»",
63
+ "url": "https://vvc.gov.lv/lv/media/62/download",
64
+ "note": "dash.range = en-tight (U+2013, no spaces). SECONDARY (VVC house style), HIGH. Ordinal-dotted ranges such as «2005.–2007.» are not range candidates (the flank is U+002E): an algorithm limit, not data. The ranges rule is off by default. Retrieved 2026-11-14."
65
+ },
66
+ {
67
+ "rule": "ellipsis",
68
+ "cite": "Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021), I pielikums «Jautājuma zīme», p. 122: «kopā ar daudzpunkti, lai pavēstītu par lielāku pauzi, kas sekos (Ko gan viņš visu tur nepiedzīvos?...)»",
69
+ "url": "https://vvc.gov.lv/lv/media/62/download",
70
+ "note": "abbreviatedAfterTerminal = false: the handbook writes «?...» with three dots (read from a page image). One house-style example, MEDIUM; no source read reduces the dots after ? or ! (the handbook's «Divpunkte», pp. 88 and 118, is a different mark: two dots for an omission), so the three dots become the ellipsis character. Retrieved 2026-11-14."
71
+ },
72
+ {
73
+ "rule": "hyphen",
74
+ "cite": "Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021), p. 87 «Defise» and I pielikums p. 115 «Defise»",
75
+ "url": "https://vvc.gov.lv/lv/media/62/download",
76
+ "note": "All three lists are empty: uses of the hyphen are described, but no closed list of unbreakable hyphenated forms is given. Retrieved 2026-11-14."
77
+ },
78
+ {
79
+ "rule": "nbsp",
80
+ "cite": "Valsts valodas centrs, LVEK, «Svarīgākie lēmumi 2014. gadā», 12.03.2014, prot. Nr. 33 2. §: valūtu nosaukumus, kodus un grafiskos simbolus «norādīt aiz summas ar saistīto atstarpi, piemēram: … 1 234 EUR; 5678,90 $»; Valsts valodas centrs, Tiesību aktu tulkošanas rokasgrāmata (2021; https://vvc.gov.lv/lv/media/62/download), p. 89 «Atstarpes»: «5 km, 25 t, 10 %, 21 €, –8 °C»; «dažos saīsinājumos (u. c., u. tml., š. g.)»; «starp saīsinājumu Nr., nr. un skaitli (Nr. 20, 155. nr.)»; «starp iniciāli un uzvārdu (J. Kalniņš)»; pp. 107–124 «… un iepriekšējo burtu vai ciparu atstarpi neliek»; p. 78 §7.1.13 (currency)",
81
+ "url": "https://vvc.gov.lv/lv/media/149/download",
82
+ "note": "Currency after the sum is NORMATIVE (LVEK 12.03.2014 prot. Nr. 33 2. §, restated in the handbook p. 78 §7.1.13); the rest is VVC house style, called «vēlams» and bounded by a warning against overuse (p. 89), SECONDARY. Handbook: https://vvc.gov.lv/lv/media/62/download. (1) beforePunctuation and narrowBeforePunctuation empty: a stated negative, «… un iepriekšējo burtu vai ciparu atstarpi neliek» for . , : ; ? ! and the apostrophe (pp. 107–124). (2) spacedUnits: % and km and °C from p. 89 (and p. 88 «15 %»), € from pp. 78 and 89, EUR from LVEK («1 234 EUR») and p. 78. Not listed though attested, by DECISION: $ (LVEK «5678,90 $»), £, ¥, USD, GBP (p. 78) and °F (fn. 54); no closed set is stated. Single-letter t, m and g are excluded by DECISION although «25 t» is attested on p. 89: they are also common words. kg and cm are not listed because no read source spaces them. A bare ° is closed up (fn. 54) and not listed; closedUnits is empty. (3) abbreviations = u. c., u. tml., š. g. (p. 89). (4) beforeNumber = Nr. only (p. 89 «Nr. 20», p. 88, p. 119); the only lowercase example of «nr.» stands after the number («155. nr.»), which N5 cannot express, so it is not listed. (5) initialBinding = single: LVEK 10.09.2014 prot. Nr. 36 5. § («pēc datuma un iniciāļa norādes lietojama saistītā atstarpe») and p. 89 («starp iniciāli un uzvārdu (J. Kalniņš)») state the rule for one initial, so «chain» would leave the cited rule unmet. ACCEPTED COST, as in fr, cs and ro: a lone capital before a full stop is read as an initial even at a sentence boundary (nbsp.md §3.9, §7.9a). (6) afterShortWords, beforeUnits, beforeWord and afterSymbols empty: no source binds un, ar, uz or no, and § is written after its number («3. §», p. 85). Not expressible and so not covered: binding after a dotted ordinal («2018. gada», «4. stāvs», «155. nr.», p. 89), after a date (LVEK), and inside digit groups («10 000»). Retrieved 2026-11-14."
83
+ }
84
+ ]
85
+ }
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "$comment": "Locale resolution input. Algorithm is specified in spec/rules/locale-resolution.md and is identical in every runtime — never delegate it to a platform locale-negotiation library. \"spec\" here must track spec/VERSION exactly — it is not itself the global version source; scripts/validate-spec.mjs enforces the match.",
4
4
  "locales": [
5
5
  "en-US",
@@ -28,7 +28,9 @@
28
28
  "hr",
29
29
  "sr",
30
30
  "sl",
31
- "lt"
31
+ "lt",
32
+ "lv",
33
+ "et"
32
34
  ],
33
35
  "aliases": {
34
36
  "en": "en-US",
@@ -1,5 +1,5 @@
1
1
  {
2
- "spec": "1.23.1",
2
+ "spec": "1.25.0",
3
3
  "$comment": "Single source of truth for pipeline order. Rules run in ascending `order`. Disabling a rule removes it from the sequence and never reorders the rest. Rule ids are public API (see docs/ARCHITECTURE.md section 5). \"spec\" here must track spec/VERSION exactly — it is not itself the global version source; scripts/validate-spec.mjs enforces the match.",
4
4
  "rules": [
5
5
  {
@@ -1,5 +1,5 @@
1
1
  # frozen_string_literal: true
2
2
 
3
3
  module Polytypo
4
- VERSION = "1.23.1"
4
+ VERSION = "1.25.0"
5
5
  end
metadata CHANGED
@@ -1,7 +1,7 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: polytypo
3
3
  version: !ruby/object:Gem::Version
4
- version: 1.23.1
4
+ version: 1.25.0
5
5
  platform: ruby
6
6
  authors:
7
7
  - Iurii Rogulia
@@ -48,6 +48,7 @@ files:
48
48
  - lib/polytypo/data/fixtures/en-GB.json
49
49
  - lib/polytypo/data/fixtures/en-US.json
50
50
  - lib/polytypo/data/fixtures/es.json
51
+ - lib/polytypo/data/fixtures/et.json
51
52
  - lib/polytypo/data/fixtures/fi.json
52
53
  - lib/polytypo/data/fixtures/fr-CA.json
53
54
  - lib/polytypo/data/fixtures/fr.json
@@ -56,6 +57,7 @@ files:
56
57
  - lib/polytypo/data/fixtures/it.json
57
58
  - lib/polytypo/data/fixtures/locale-resolution.json
58
59
  - lib/polytypo/data/fixtures/lt.json
60
+ - lib/polytypo/data/fixtures/lv.json
59
61
  - lib/polytypo/data/fixtures/nl.json
60
62
  - lib/polytypo/data/fixtures/pl.json
61
63
  - lib/polytypo/data/fixtures/pt-BR.json
@@ -76,6 +78,7 @@ files:
76
78
  - lib/polytypo/data/locales/en-GB.json
77
79
  - lib/polytypo/data/locales/en-US.json
78
80
  - lib/polytypo/data/locales/es.json
81
+ - lib/polytypo/data/locales/et.json
79
82
  - lib/polytypo/data/locales/fi.json
80
83
  - lib/polytypo/data/locales/fr-CA.json
81
84
  - lib/polytypo/data/locales/fr.json
@@ -83,6 +86,7 @@ files:
83
86
  - lib/polytypo/data/locales/hu.json
84
87
  - lib/polytypo/data/locales/it.json
85
88
  - lib/polytypo/data/locales/lt.json
89
+ - lib/polytypo/data/locales/lv.json
86
90
  - lib/polytypo/data/locales/nl.json
87
91
  - lib/polytypo/data/locales/pl.json
88
92
  - lib/polytypo/data/locales/pt-BR.json