polytypo 1.21.0 → 1.23.0
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.
- checksums.yaml +4 -4
- data/lib/polytypo/data/VERSION +1 -1
- data/lib/polytypo/data/fixtures/bg.json +12 -4
- data/lib/polytypo/data/fixtures/cs.json +9 -1
- data/lib/polytypo/data/fixtures/de-CH.json +12 -4
- data/lib/polytypo/data/fixtures/de-DE.json +113 -13
- data/lib/polytypo/data/fixtures/el.json +28 -4
- data/lib/polytypo/data/fixtures/en-GB.json +23 -7
- data/lib/polytypo/data/fixtures/en-US.json +694 -162
- data/lib/polytypo/data/fixtures/es.json +1 -1
- data/lib/polytypo/data/fixtures/fi.json +10 -2
- data/lib/polytypo/data/fixtures/fr-CA.json +9 -1
- data/lib/polytypo/data/fixtures/fr.json +46 -13
- data/lib/polytypo/data/fixtures/hr.json +9 -1
- data/lib/polytypo/data/fixtures/hu.json +28 -4
- data/lib/polytypo/data/fixtures/it.json +1 -1
- data/lib/polytypo/data/fixtures/locale-resolution.json +1 -1
- data/lib/polytypo/data/fixtures/lt.json +9 -1
- data/lib/polytypo/data/fixtures/nl.json +17 -1
- data/lib/polytypo/data/fixtures/pl.json +17 -1
- data/lib/polytypo/data/fixtures/pt-BR.json +9 -1
- data/lib/polytypo/data/fixtures/pt-PT.json +9 -1
- data/lib/polytypo/data/fixtures/ro.json +2 -2
- data/lib/polytypo/data/fixtures/ru.json +38 -6
- data/lib/polytypo/data/fixtures/sk.json +9 -1
- data/lib/polytypo/data/fixtures/sl.json +9 -1
- data/lib/polytypo/data/fixtures/sr.json +12 -4
- data/lib/polytypo/data/fixtures/sv.json +12 -4
- data/lib/polytypo/data/fixtures/tr.json +9 -1
- data/lib/polytypo/data/fixtures/uk.json +1 -1
- data/lib/polytypo/data/locales/bg.json +9 -1
- data/lib/polytypo/data/locales/cs.json +2 -0
- data/lib/polytypo/data/locales/de-CH.json +9 -1
- data/lib/polytypo/data/locales/de-DE.json +9 -1
- data/lib/polytypo/data/locales/el.json +9 -1
- data/lib/polytypo/data/locales/en-GB.json +9 -4
- data/lib/polytypo/data/locales/en-US.json +11 -11
- data/lib/polytypo/data/locales/es.json +2 -0
- data/lib/polytypo/data/locales/fi.json +8 -3
- data/lib/polytypo/data/locales/fr-CA.json +9 -1
- data/lib/polytypo/data/locales/fr.json +3 -1
- data/lib/polytypo/data/locales/hr.json +3 -1
- data/lib/polytypo/data/locales/hu.json +9 -1
- data/lib/polytypo/data/locales/it.json +2 -0
- data/lib/polytypo/data/locales/lt.json +12 -5
- data/lib/polytypo/data/locales/nl.json +10 -2
- data/lib/polytypo/data/locales/pl.json +10 -2
- data/lib/polytypo/data/locales/pt-BR.json +4 -2
- data/lib/polytypo/data/locales/pt-PT.json +5 -3
- data/lib/polytypo/data/locales/registry.json +1 -1
- data/lib/polytypo/data/locales/ro.json +2 -0
- data/lib/polytypo/data/locales/ru.json +9 -17
- data/lib/polytypo/data/locales/sk.json +7 -5
- data/lib/polytypo/data/locales/sl.json +3 -1
- data/lib/polytypo/data/locales/sr.json +9 -2
- data/lib/polytypo/data/locales/sv.json +8 -6
- data/lib/polytypo/data/locales/tr.json +4 -2
- data/lib/polytypo/data/locales/uk.json +2 -0
- data/lib/polytypo/data/rules/modes.md +22 -5
- data/lib/polytypo/data/rules/nbsp.md +271 -74
- data/lib/polytypo/data/rules/order.json +1 -1
- data/lib/polytypo/data/rules/pipeline-idempotency.md +54 -7
- data/lib/polytypo/data/rules/quote-punctuation.md +14 -7
- data/lib/polytypo/data/rules/quotes.md +14 -5
- data/lib/polytypo/data/rules/ranges.md +7 -1
- data/lib/polytypo/data/rules/spaces.md +4 -2
- data/lib/polytypo/data/rules/symbols.md +2 -1
- data/lib/polytypo/data/schema/locale.schema.json +15 -1
- data/lib/polytypo/engine/rules/nbsp.rb +161 -11
- data/lib/polytypo/modes/html.rb +10 -0
- data/lib/polytypo/version.rb +1 -1
- metadata +1 -1
|
@@ -54,23 +54,9 @@
|
|
|
54
54
|
"со"
|
|
55
55
|
],
|
|
56
56
|
"abbreviations": ["и т. д.", "и т. п.", "т. е.", "и др."],
|
|
57
|
-
"beforeUnits": [
|
|
58
|
-
|
|
59
|
-
|
|
60
|
-
"₽",
|
|
61
|
-
"°C",
|
|
62
|
-
"км",
|
|
63
|
-
"см",
|
|
64
|
-
"мм",
|
|
65
|
-
"кг",
|
|
66
|
-
"г.",
|
|
67
|
-
"гг.",
|
|
68
|
-
"в.",
|
|
69
|
-
"вв.",
|
|
70
|
-
"тыс.",
|
|
71
|
-
"млн",
|
|
72
|
-
"млрд"
|
|
73
|
-
],
|
|
57
|
+
"beforeUnits": ["‰", "₽", "г.", "гг.", "в.", "вв.", "тыс.", "млн", "млрд"],
|
|
58
|
+
"spacedUnits": ["°C", "км", "см", "мм", "кг"],
|
|
59
|
+
"closedUnits": ["%"],
|
|
74
60
|
"beforeNumber": [],
|
|
75
61
|
"beforeWord": ["ул.", "пл."],
|
|
76
62
|
"afterSymbols": ["№", "§"],
|
|
@@ -112,6 +98,12 @@
|
|
|
112
98
|
"cite": "Правила русской орфографии и пунктуации. Полный академический справочник / под ред. В. В. Лопатина. М., 2006: § 135 — пишутся через дефис местоименные слова с начальной частью (приставкой) кое- (кой-) и с конечными частями (постфиксами) -либо, -нибудь, -то; § 141 п. 2 — сложные предлоги из-за, из-под (а также диалектные по-за, по-над); § 141 п. 3 и § 143 — частицы -де, -ка, -те, -то, -с и -таки (всё-таки, опять-таки, прямо-таки, так-таки)",
|
|
113
99
|
"url": "https://orthographia.ru/orf.php?paragraph=pp135.php",
|
|
114
100
|
"note": "Тексты §§ 135, 141, 143 и 219 прочитаны по онлайн-изданию справочника. Эти параграфы нормативны для СОСТАВА форм, но не для запрета переноса, и это важно не смешивать: § 219 («не подлежат переносу») называет только аббревиатуры из прописных букв, графические сокращения (б-ка, ж.-д.) и наращения (20-й); слов с дефисом там нет, а правило повторять дефис в начале перенесённой части (военно-/-морской) справочник даёт как факультативное. Значит, связывание дефиса неразрывным U+2011 — решение проекта (docs/PLAN.md §3.3), а не орфографическая норма; его практическое обоснование то же, что у факультативного правила: при переносе по дефису теряется различие слитного и дефисного написания. Диалектные и просторечные по-за, по-над, для-ради, за-ради не внесены как непродуктивные. Класс графических сокращений из § 219 (б-ка, ж.-д., р/сч) нормативно неразрывен, но перечислить его литеральным списком нельзя, поэтому он здесь не отражён. Из § 143 взята также частица -ка; -де, -с, -те опущены как архаичные и редкие."
|
|
101
|
+
},
|
|
102
|
+
{
|
|
103
|
+
"rule": "nbsp",
|
|
104
|
+
"cite": "ГОСТ 8.417-2024 «ГСИ. Единицы величин», п. 8.3: «Между последней цифрой числа и обозначением единицы оставляют пробел»; «20 °C» (неправильно «20°C»); «80 %» в тексте стандарта; Бюро Горбунова, «Знак процента» (8.8.2021): объективного стандарта нет, за пробел — Мильчин, ГОСТ, стиль РИА, против — справочники Вигдорчика и Гиленсона, советские энциклопедии, редакторы федеральных ведомств и большинство СМИ; Грамота.ру, справка № 265643: «рекомендуется отбивать пробелом»",
|
|
105
|
+
"url": "https://bureau.ru/soviet/20210808/",
|
|
106
|
+
"note": "°C, км, см, мм, кг — spacedUnits по ГОСТ 8.417-2024 (п. 8.3; текст прочитан на стороннем зеркале через пересказ, уверенность средняя-высокая). % — closedUnits: ГОСТ относится к метрологическим и техническим документам, для обычного текста источники расходятся, а где допустимы обе формы, locale выбирает слитную (решение оператора 11.11.2026), чтобы текст выглядел единообразно. Варианты написания «80 %» по ГОСТ в этом режиме станут «80%». ‰ и ₽ остаются в beforeUnits: для них нет подтверждённой формы. Голый ° не в списках: угол и широта пишутся слитно. Проверено 11.11.2026."
|
|
115
107
|
}
|
|
116
108
|
]
|
|
117
109
|
}
|
|
@@ -36,7 +36,9 @@
|
|
|
36
36
|
"narrowBeforePunctuation": [],
|
|
37
37
|
"afterShortWords": [],
|
|
38
38
|
"abbreviations": ["t. j.", "a pod.", "a. s."],
|
|
39
|
-
"beforeUnits": ["
|
|
39
|
+
"beforeUnits": ["kg", "km", "km/s", "m/s", "min"],
|
|
40
|
+
"spacedUnits": ["%", "‰"],
|
|
41
|
+
"closedUnits": [],
|
|
40
42
|
"beforeNumber": ["č.", "ods.", "s."],
|
|
41
43
|
"beforeWord": [],
|
|
42
44
|
"afterSymbols": ["§"],
|
|
@@ -57,7 +59,7 @@
|
|
|
57
59
|
},
|
|
58
60
|
{
|
|
59
61
|
"rule": "ranges",
|
|
60
|
-
"cite": "Jazykovedný ústav Ľudovíta Štúra SAV, Pravidlá slovenského pravopisu (3., upravené a doplnené vydanie, Veda 2000, tak ako je zverejnené na https://www.juls.savba.sk/ediela/psp2000/psp.pdf; JÚĽŠ na https://www.juls.savba.sk/psp_2013.html označuje vydanie z roku 2013 za nezmenené), kap. 2.8 bod 16a: «na naznačenie rozpätia… a) časového a miestneho rozpätia a rozpätia v počte, pričom nahrádza výraz až, napr. 1980 – 1990, január – február, strana 5 – 12»; 2.12 bod 3: «I. Newton (1643 – 1727)»; STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://
|
|
62
|
+
"cite": "Jazykovedný ústav Ľudovíta Štúra SAV, Pravidlá slovenského pravopisu (3., upravené a doplnené vydanie, Veda 2000, tak ako je zverejnené na https://www.juls.savba.sk/ediela/psp2000/psp.pdf; JÚĽŠ na https://www.juls.savba.sk/psp_2013.html označuje vydanie z roku 2013 za nezmenené), kap. 2.8 bod 16a: «na naznačenie rozpätia… a) časového a miestneho rozpätia a rozpätia v počte, pričom nahrádza výraz až, napr. 1980 – 1990, január – február, strana 5 – 12»; 2.12 bod 3: «I. Newton (1643 – 1727)»; STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://gturzovka.edupage.org/files/STN_01_6910_Pravidla_pisania_a_upravy_pisomnosti.pdf), s. 7: medzi chybami uvádza chýbajúcu medzeru pred pomlčkou a za ňou («1995 – 2000… pomlčku nemožno písať bez medzier»)",
|
|
61
63
|
"url": "https://www.juls.savba.sk/ediela/psp2000/psp.pdf",
|
|
62
64
|
"note": "Odôvodňuje dash.range = «en-spaced», a to je presne bod, kde analógia s češtinou zlyháva: ÚJČ sadzí rozsah tesne («en-tight»), PSP s medzerami na oboch stranách. Druhý zdroj (výklad STN) len potvrdzuje, nie je normatívny. Pravidlo ranges je predvolene vypnuté (spec 0.5.0). Overené 19. 10. 2026."
|
|
63
65
|
},
|
|
@@ -81,14 +83,14 @@
|
|
|
81
83
|
},
|
|
82
84
|
{
|
|
83
85
|
"rule": "nbsp",
|
|
84
|
-
"cite": "Jazykovedný ústav Ľudovíta Štúra SAV, Pravidlá slovenského pravopisu (3., upravené a doplnené vydanie, Veda 2000, tak ako je zverejnené na https://www.juls.savba.sk/ediela/psp2000/psp.pdf; JÚĽŠ na https://www.juls.savba.sk/psp_2013.html označuje vydanie z roku 2013 za nezmenené), kap. 2.1 bod 4: «…a pod. (a podobne), t. j. (to jest)…»; 2.1 «Bodka sa nepíše» bod 5: «v bode č. 2 (v bode číslo dva), uznesenie č. 5…, § 7 ods. 2 (paragraf sedem odsek dva)…, na s. 5 (na strane päť)»; 2.4.1 bod 7c: «Váhostav, a. s., Žilina patrí medzi najväčšie stavebné podniky u nás.»; STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://
|
|
86
|
+
"cite": "Jazykovedný ústav Ľudovíta Štúra SAV, Pravidlá slovenského pravopisu (3., upravené a doplnené vydanie, Veda 2000, tak ako je zverejnené na https://www.juls.savba.sk/ediela/psp2000/psp.pdf; JÚĽŠ na https://www.juls.savba.sk/psp_2013.html označuje vydanie z roku 2013 za nezmenené), kap. 2.1 bod 4: «…a pod. (a podobne), t. j. (to jest)…»; 2.1 «Bodka sa nepíše» bod 5: «v bode č. 2 (v bode číslo dva), uznesenie č. 5…, § 7 ods. 2 (paragraf sedem odsek dva)…, na s. 5 (na strane päť)»; 2.4.1 bod 7c: «Váhostav, a. s., Žilina patrí medzi najväčšie stavebné podniky u nás.»; STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://gturzovka.edupage.org/files/STN_01_6910_Pravidla_pisania_a_upravy_pisomnosti.pdf), 3.1.1: «Značka paragrafu sa píše len v spojení s číslom… s medzerou z ľavej i pravej strany.»",
|
|
85
87
|
"url": "https://www.juls.savba.sk/ediela/psp2000/psp.pdf",
|
|
86
88
|
"note": "(1) abbreviations: «t. j.», «a pod.», «a. s.» — tvar skratky s medzerou je v PSP doložený. «s. r. o.» je vynechané: opiera sa len o výklad STN, ktorý ho pripisuje Obchodnému zákonníku; text zákona sa nečítal. (2) beforeNumber: «č.», «ods.», «s.» — doložené tvary z PSP, rovnaký základ ako «Nr.» a «S.» v de-DE; «s.» je zároveň PSP skratka pre stredný rod, čo je neškodné, lebo N9 sa spúšťa len pred číslicou. (3) afterSymbols = [«§»]. (4) afterShortWords je PRÁZDNE: kapitola o delení riadkov (V) má uzavretý zoznam «Nerozdeľujeme» a pravidlo o jednopísmenných predložkách neobsahuje; či ho obsahuje platená norma STN 01 6910, nie je zistené. ROZHODNUTIE, nie citácia: pri nedoložení sa nepridáva nič (opak voči cs, kde zákaz ÚJČ doložil). (5) initialBinding = «none» a beforeWord prázdne: žiadny zdroj nehovorí, že iniciála a priezvisko či «tzv.» a slovo za ním sa nesmú rozdeliť (PSP 2.1 Poznámka uvádza len tvary skratiek mien («A. Bernolák», «J. M. Hurban», «M. K.»)); rozhodnutie, nie citácia. Overené 19. 10. 2026."
|
|
87
89
|
},
|
|
88
90
|
{
|
|
89
91
|
"rule": "nbsp",
|
|
90
|
-
"cite": "STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://
|
|
91
|
-
"url": "https://
|
|
92
|
+
"cite": "STN 01 6910 Pravidlá písania a úpravy písomností – výklad normy (STU Bratislava, 6. 8. 2010; výklad normy, nie jej text ani JÚĽŠ; https://gturzovka.edupage.org/files/STN_01_6910_Pravidla_pisania_a_upravy_pisomnosti.pdf), 3.3: «Značky jednotiek sa píšu vždy v rovnakom riadku za číselnou hodnotou, od ktorej sa oddeľujú medzerou» (15 kg, 10 m/s, 45 min); 3.6.3: «Značky percenta (%), promile (‰) a priemeru (Ø) sa používajú len v spojení s číslom, od ktorého sa oddeľujú medzerou» (20 % z ceny); Jazykovedný ústav Ľudovíta Štúra SAV, Pravidlá slovenského pravopisu (3., upravené a doplnené vydanie, Veda 2000, tak ako je zverejnené na https://www.juls.savba.sk/ediela/psp2000/psp.pdf; JÚĽŠ na https://www.juls.savba.sk/psp_2013.html označuje vydanie z roku 2013 za nezmenené), kap. 2.13 bod 2: «spotreba 6 l/100 km», «120 000 km/s»; BIPM, The International System of Units (SI), 9th ed. (2019), concise summary (https://www.bipm.org/documents/20126/41483022/SI-Brochure-9-concise-EN.pdf), «The language of science: using the SI to express the values of quantities»: «A single space is always left between the number and the unit»",
|
|
93
|
+
"url": "https://gturzovka.edupage.org/files/STN_01_6910_Pravidla_pisania_a_upravy_pisomnosti.pdf",
|
|
92
94
|
"note": "POZOR, nižšia vrstva: tu je normatívny len BIPM pre totožnosť značiek SI; slovenskú konvenciu dokladá výklad STN (univerzitný, nie JÚĽŠ) a príklady PSP. Rozdelenie rolí ako v cs, en-US, fi, sv. Jednopísmenové značky vynechané (s, m, h, l, W): «s», «v», «z», «k» sú predložky, rovnaká kolízia ako v cs. «°C» vynechané: výklad STN sám tlačí «-23˚C» bez medzery, proti vlastnej vete aj proti BIPM, takže hodnota je nepotvrdená. Overené 19. 10. 2026."
|
|
93
95
|
}
|
|
94
96
|
]
|
|
@@ -36,7 +36,9 @@
|
|
|
36
36
|
"narrowBeforePunctuation": [],
|
|
37
37
|
"afterShortWords": [],
|
|
38
38
|
"abbreviations": ["t. i."],
|
|
39
|
-
"beforeUnits": ["
|
|
39
|
+
"beforeUnits": ["€", "km"],
|
|
40
|
+
"spacedUnits": ["%", "‰", "°C", "kg"],
|
|
41
|
+
"closedUnits": [],
|
|
40
42
|
"beforeNumber": [],
|
|
41
43
|
"beforeWord": [],
|
|
42
44
|
"afterSymbols": ["§"],
|
|
@@ -36,7 +36,9 @@
|
|
|
36
36
|
"narrowBeforePunctuation": [],
|
|
37
37
|
"afterShortWords": [],
|
|
38
38
|
"abbreviations": [],
|
|
39
|
-
"beforeUnits": [
|
|
39
|
+
"beforeUnits": [],
|
|
40
|
+
"spacedUnits": ["kg", "cm", "mm", "°C"],
|
|
41
|
+
"closedUnits": ["%"],
|
|
40
42
|
"beforeNumber": [],
|
|
41
43
|
"beforeWord": [],
|
|
42
44
|
"afterSymbols": [],
|
|
@@ -77,7 +79,12 @@
|
|
|
77
79
|
"rule": "nbsp",
|
|
78
80
|
"cite": "Direkcija za mere i dragocene metale (DMDM), «Pravila upotrebe i pisanja mernih jedinica i fizičkih veličina»: «Oznake mernih jedinica pišu se iza numeričkih vrednosti u izrazima za veličinu, ostavljajući razmak između numeričke vrednosti i oznake merne jedinice.»; «U novi red se ne sme preneti samo oznaka merne jedinice.»; primeri označeni kao ispravni: «2 kg, 3 kg i 4 kg», «80 mm • 25 mm • 50 mm», «lAl = 30 cm», «25 °C ± 5 °C». Uredba o zakonskim mernim jedinicama i načinu njihove upotrebe («Sl. glasnik RS», br. 132/2021), član 5: «Način pisanja mernih jedinica mora biti u skladu sa standardom SRPS EN ISO 80000.»",
|
|
79
81
|
"url": "https://www.dmdm.rs/sr/o-nama/merne-jedinice/pravila",
|
|
80
|
-
"note": "beforeUnits = kg, cm, mm, °C (U+00B0 U+0043): each printed with a space after the number by the national metrology authority, which also states outright that the unit symbol must not be carried alone to a new line. A metrology source, not a language authority, so MEDIUM — the same footing as the form evidence of hr, sk and bg. Single-letter units are excluded as in hr, sk and bg. km, km/h, ha, min and the Cyrillic forms (кг, км) are not attested with a space and are not added. '%' is
|
|
82
|
+
"note": "beforeUnits = kg, cm, mm, °C (U+00B0 U+0043): each printed with a space after the number by the national metrology authority, which also states outright that the unit symbol must not be carried alone to a new line. A metrology source, not a language authority, so MEDIUM — the same footing as the form evidence of hr, sk and bg. Single-letter units are excluded as in hr, sk and bg. km, km/h, ha, min and the Cyrillic forms (кг, км) are not attested with a space and are not added. '%' is not in spacedUnits: BIPM SI Brochure 9th ed. §5.4.7 sets a space before it, but Serbian practice in print is closed up (Брборић and Новокмет, Књижевност и језик 64/3–4, 2017, «96%», «73,5%») and no Serbian norm was read, so it is in closedUnits (entry below). Corroborated, language-neutrally, by BIPM SI Brochure 9th ed. §5.4.3 (https://www.bipm.org/documents/20126/41483022/SI-Brochure-9.pdf). Currency, afterSymbols, abbreviations, afterShortWords and initialBinding stay empty or none because no free source states a rule: Klikovac, Српска интерпункција, pp. 57–64 lists % and ° among signs of Пижурица и др. 2010 without any spacing rule; the Pravopis 2010 (paid) was not read. Operator decision (no paid sources, pick and pin). Retrieved 2026-11-06."
|
|
83
|
+
},
|
|
84
|
+
{
|
|
85
|
+
"rule": "nbsp",
|
|
86
|
+
"cite": "DMDM, «Pravila upotrebe i pisanja mernih jedinica»; Брборић и Новокмет, Књижевност и језик 64/3–4 (2017): «96%», «73,5%»",
|
|
87
|
+
"note": "% in closedUnits: Serbian printed practice sets it closed up and no Serbian norm was read that requires a space (operator decision, pick and pin, 2026-11-11). kg, cm, mm, °C in spacedUnits on the metrology authority's rule. Retrieved 2026-11-11."
|
|
81
88
|
}
|
|
82
89
|
]
|
|
83
90
|
}
|
|
@@ -37,14 +37,8 @@
|
|
|
37
37
|
"afterShortWords": [],
|
|
38
38
|
"abbreviations": [],
|
|
39
39
|
"beforeUnits": [
|
|
40
|
-
"%",
|
|
41
|
-
"‰",
|
|
42
|
-
"kr",
|
|
43
|
-
"€",
|
|
44
|
-
"kg",
|
|
45
40
|
"g",
|
|
46
41
|
"mg",
|
|
47
|
-
"km",
|
|
48
42
|
"m",
|
|
49
43
|
"cm",
|
|
50
44
|
"mm",
|
|
@@ -60,6 +54,8 @@
|
|
|
60
54
|
"h",
|
|
61
55
|
"°C"
|
|
62
56
|
],
|
|
57
|
+
"spacedUnits": ["%", "‰", "kr", "€", "km", "kg"],
|
|
58
|
+
"closedUnits": [],
|
|
63
59
|
"beforeNumber": [],
|
|
64
60
|
"beforeWord": [],
|
|
65
61
|
"afterSymbols": ["§"],
|
|
@@ -124,6 +120,12 @@
|
|
|
124
120
|
"cite": "Språkrådet, Snabba skrivregler, §4.4 (sammansättningar med förkortningar) and kap. 7 (avstavning)",
|
|
125
121
|
"url": "https://www.diva-portal.org/smash/get/diva2:1829117/FULLTEXT04.pdf",
|
|
126
122
|
"note": "`prefixes`, `suffixes` and `compounds` are empty. The guide’s hyphen material states when to insert a bindestreck in compounds (e-post, EU-länderna, FN-arbete, satellit-tv-program, lan-spel) and treats avstavning as a separate topic; it never forbids a line break at an existing hyphen. Swedish compounding is productive, so there is no closed inventory to list — the field exists for the Russian кое- / -таки / из-под case. No U+2011 requirement found in any Språkrådet material consulted."
|
|
123
|
+
},
|
|
124
|
+
{
|
|
125
|
+
"rule": "nbsp",
|
|
126
|
+
"cite": "Myndigheternas skrivregler (Ds 2009:38), §10.15 «Procent och promille»: «Mellanslag sätts före tecknet.»; Snabba skrivregler §5.1.1 och §1.16: «10 km, 11 kg», «pris 7,50 kr», «3 € (tre euro)»",
|
|
127
|
+
"url": "https://www.regeringen.se/contentassets/cf18bfb62e0c4282ab3e8632080c2fb4/myndigheternas-skrivregler-ds-200938",
|
|
128
|
+
"note": "%, ‰, kr, €, km, kg in spacedUnits. Other units stay in beforeUnits: the sources list their symbols without a number. °C and bare ° were not found in either document. Retrieved 2026-11-11."
|
|
127
129
|
}
|
|
128
130
|
]
|
|
129
131
|
}
|
|
@@ -36,7 +36,9 @@
|
|
|
36
36
|
"narrowBeforePunctuation": [],
|
|
37
37
|
"afterShortWords": [],
|
|
38
38
|
"abbreviations": ["Kur. Bşk.", "Nö. Sb."],
|
|
39
|
-
"beforeUnits": [
|
|
39
|
+
"beforeUnits": [],
|
|
40
|
+
"spacedUnits": ["mm", "cm", "km", "kg", "mg", "hl", "m²", "cm²", "°C", "ton"],
|
|
41
|
+
"closedUnits": [],
|
|
40
42
|
"beforeNumber": [],
|
|
41
43
|
"beforeWord": [],
|
|
42
44
|
"afterSymbols": [],
|
|
@@ -101,7 +103,7 @@
|
|
|
101
103
|
"rule": "nbsp",
|
|
102
104
|
"cite": "Türk Dil Kurumu, Sıkça Sorulan Sorular, «°C \"santigrat derece\" işaretiyle sayı arasında boşluk bırakılır mı?»: «Sayı ve ölçü birimi kısaltmaları aralarında boşluk bırakılır. Tıpkı 5 m, 20 kg, 350 ton örneklerinde olduğu gibi, °C \"santigrat derece\" işaretiyle sayı arasında boşluk bırakılır: 15 °C gibi.»; Türk Dil Kurumu, Yazım Kuralları, «Kısaltmalar» 2: «Ölçü birimlerinin uluslararası kısaltmaları kullanılır: m (metre), mm (milimetre), cm (santimetre), km (kilometre), g (gram), kg (kilogram), l (litre), hl (hektolitre), mg (miligram), m² (metrekare), cm² (santimetrekare) vb.»",
|
|
103
105
|
"url": "https://tdk.gov.tr/icerik/sikca-sorulan-sorular/c-derece-santigrat-isaretiyle-sayi-arasinda-bosluk-birakilir-mi/",
|
|
104
|
-
"note": "beforeUnits. Rol paylaşımı başka locale'lerdeki gibi iki kuruma değil tek kuruma düşüyor: hangi dizgilerin uluslararası birim simgesi olduğunu «Kısaltmalar» 2, aralarında boşluk bırakıldığını SSS söylüyor. KAYNAK RÜTBESİ açıkça belirtiliyor: Sıkça Sorulan Sorular, Yazım Kılavuzu'nun altındadır ve boşluk hükmünün tek dayanağı odur. Tek harfli simgeler (m, g, l) fr.json ve pl.json uygulamasına uyularak atlandı. «ton» ayrıca açıklanıyor: uluslararası bir simge değil bir kelimedir ve yalnızca SSS'nin düzyazı örneğinden («350 ton») gelir — listede tutulmasının nedeni N5'in sağ sınır testinin onu «tonluk» içinde eşleştirmemesi ve hiçbir şey uydurulmamış olmasıdır. «cm» ile «cm²» birlikte listelenebilir, çünkü aynı konumda EN UZUN EŞLEŞME KAZANIR (nbsp.md §3.7) — «15 cm²» girdisinde «cm» değil «cm²» eşleşir. «%» BİLEREK YOK: TDK yüzde işaretini sayıdan ÖNCE ve bitişik yazar (ayrı kayıt). N5
|
|
106
|
+
"note": "beforeUnits. Rol paylaşımı başka locale'lerdeki gibi iki kuruma değil tek kuruma düşüyor: hangi dizgilerin uluslararası birim simgesi olduğunu «Kısaltmalar» 2, aralarında boşluk bırakıldığını SSS söylüyor. KAYNAK RÜTBESİ açıkça belirtiliyor: Sıkça Sorulan Sorular, Yazım Kılavuzu'nun altındadır ve boşluk hükmünün tek dayanağı odur. Tek harfli simgeler (m, g, l) fr.json ve pl.json uygulamasına uyularak atlandı. «ton» ayrıca açıklanıyor: uluslararası bir simge değil bir kelimedir ve yalnızca SSS'nin düzyazı örneğinden («350 ton») gelir — listede tutulmasının nedeni N5'in sağ sınır testinin onu «tonluk» içinde eşleştirmemesi ve hiçbir şey uydurulmamış olmasıdır. «cm» ile «cm²» birlikte listelenebilir, çünkü aynı konumda EN UZUN EŞLEŞME KAZANIR (nbsp.md §3.7) — «15 cm²» girdisinde «cm» değil «cm²» eşleşir. «%» BİLEREK YOK: TDK yüzde işaretini sayıdan ÖNCE ve bitişik yazar (ayrı kayıt). «20kg» N5 tarafından «20 kg» yapılır (spacedUnits). Erişim: 23.09.2026."
|
|
105
107
|
},
|
|
106
108
|
{
|
|
107
109
|
"rule": "nbsp",
|
|
@@ -19,7 +19,7 @@ step 9 with §3.8.5's content-line condition, added in 1.10.0, §3.8.4's carriag
|
|
|
19
19
|
continuation line with the bails of §3.8.6 that use it, added in 1.11.0, and §3.8.4's node lines,
|
|
20
20
|
declined keys, node markers and own-line region, with the changes to steps 1, 3, 4, 5, 6, 7 and 8
|
|
21
21
|
that go with them, added in 1.11.0, and §3.9 with the amendments it carries to §3.1, §3.5 step 2,
|
|
22
|
-
§4, §5, §7 item 7 and §8, added in 1.13.0, and §3.4's delimiter-neighbour rule with §3.7.3's autolink-literal tail, terminator guard and reference labels, and the amendments they carry to §3.4's deletion paragraph, §5 items 2 and 3, its fifth-part note and testing obligation, and §6, added in 1.17.2, and §3.4's escape-and-reference rule with the amendments it carries to §3.6, §3.7.3, §5 item 2's fifth-part note, §5 item 3 and the testing obligation, added in 1.17.4, and §3.4's bracket-shape rule, added in 1.17.5).
|
|
22
|
+
§4, §5, §7 item 7 and §8, added in 1.13.0, and §3.4's delimiter-neighbour rule with §3.7.3's autolink-literal tail, terminator guard and reference labels, and the amendments they carry to §3.4's deletion paragraph, §5 items 2 and 3, its fifth-part note and testing obligation, and §6, added in 1.17.2, and §3.4's escape-and-reference rule with the amendments it carries to §3.6, §3.7.3, §5 item 2's fifth-part note, §5 item 3 and the testing obligation, added in 1.17.4, and §3.4's bracket-shape rule, added in 1.17.5, and §3.6's whitespace-only text node rule, added in 1.23.0).
|
|
23
23
|
|
|
24
24
|
---
|
|
25
25
|
|
|
@@ -125,7 +125,7 @@ possibility of collision with author content and makes the classification table
|
|
|
125
125
|
|
|
126
126
|
The test is on the raw source bytes of the gap, so it is decidable without asking the parser
|
|
127
127
|
anything and is identical in five runtimes. `a<em>x</em>b` gives −1; `foo\nbar`, `foo\n\nbar`
|
|
128
|
-
and `<p>a</p>\n<p>b</p>` all give −2.
|
|
128
|
+
and `<p>a</p>\n<p>b</p>` all give −2 (in the last the text node `\n` is not a span, §3.6).
|
|
129
129
|
|
|
130
130
|
**Why a line boundary is a `BREAK` and not just another opaque marker.** Every rule already has
|
|
131
131
|
correct, specified, fixture-covered behaviour at a line terminator, because `text` mode has real
|
|
@@ -165,7 +165,7 @@ simply a member of `BREAK` and of nothing else, so it needs no table: every rule
|
|
|
165
165
|
| `DOTLIKE`, `STRIP-BEFORE`, `TERMINAL` | **no** |
|
|
166
166
|
| `STRAIGHT`, `SQ`, `DQ` | **no** |
|
|
167
167
|
| `OPEN-BRACKET`, `CLOSE-BRACKET` | **no** |
|
|
168
|
-
| any literal matching list (`abbreviations`, `beforeUnits`, `hyphen.*`, the trademark table) | **no** — the marker never matches a literal |
|
|
168
|
+
| any literal matching list (`abbreviations`, `beforeUnits`, `spacedUnits`, `closedUnits`, `hyphen.*`, the trademark table) | **no** — the marker never matches a literal |
|
|
169
169
|
| `OPENISH` **and** `CLOSEISH` (`quotes`, `apostrophe`) | **yes, both** |
|
|
170
170
|
| `CLOSEISH` (`nbsp`) | **yes** |
|
|
171
171
|
| `OPENISH` (`nbsp`), `OPENQUOTE` and `CLOSEDELIM` (`apostrophe`) | **no** |
|
|
@@ -272,7 +272,7 @@ special-casing**:
|
|
|
272
272
|
| ----------------------------- | ---------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
|
|
273
273
|
| `<em>foo </em>- bar` | `foo ⟦- bar` | `dashes`: the code point left of the dash is the marker, not a space, so `lsp = 0` while `rsp = 1` — the symmetry guard declines. Miss, not damage |
|
|
274
274
|
| `He said <code>x</code> "hi"` | `He said ⟦ "hi"` | `spaces`: two runs of length 1 separated by the marker, not one run of 2. No collapse |
|
|
275
|
-
| `5 <em>km</em>` | `5 ⟦km` | `nbsp` N5
|
|
275
|
+
| `5 <em>km</em>` | `5 ⟦km` | `nbsp` N5 needs `cp[a-1]` to be a separator or a `DIGIT` (§3.7 step 3); it is the marker, which is neither. Declines, so nothing is inserted or deleted |
|
|
276
276
|
| `из<em>-под</em>` | `из⟦-под` | `hyphen`: no listed form matches across the marker. Declines |
|
|
277
277
|
| `(c<em>)</em>` | `(c⟦)` | `symbols`: the literal `(c)` does not match. Declines |
|
|
278
278
|
|
|
@@ -539,7 +539,24 @@ Skipped, exhaustively and exactly:
|
|
|
539
539
|
conversions on both sides of it for nothing — `Tom & Jerry's "book"` would lose the pairing of
|
|
540
540
|
its quotation marks. Well-formedness, not the presence of an ampersand, is what makes a
|
|
541
541
|
skipped region;
|
|
542
|
-
- comments, CDATA, doctype, processing instructions, and any prologue
|
|
542
|
+
- comments, CDATA, doctype, processing instructions, and any prologue;
|
|
543
|
+
- every **whitespace-only text run that contains a line break** (since 1.23.0): a maximal run of
|
|
544
|
+
source text between two pieces of markup (a tag, comment, CDATA section, doctype or processing
|
|
545
|
+
instruction) or a document edge, whose **raw source**, as written and before the split around
|
|
546
|
+
character references, consists only of U+0009, U+000A, U+000C, U+000D and U+0020, at least one
|
|
547
|
+
of them U+000A, U+000C or U+000D. A character reference is never whitespace for this test, so
|
|
548
|
+
`\n \n` and `\n&\n` are processed as before. Such a run is not a span, so it lies in the
|
|
549
|
+
gap between the spans on either side and §3.2 chooses the marker from its raw source: the
|
|
550
|
+
pretty-printed `<p>a</p>\n<p>b</p>` has one −2 line marker between `a` and `b`, as the same text
|
|
551
|
+
has in `text` mode, instead of two −1 markers around a literal line feed. At the start or end
|
|
552
|
+
of a text unit such a run leaves no marker at all (§3.5 step 2), so `<p>Call it "ready".</p>`
|
|
553
|
+
followed by a line feed ends the unit as it does without one. The same holds inside a
|
|
554
|
+
`markdown` HTML block, which §3.7.3 hands to this list. A whitespace-only text node with **no** line break
|
|
555
|
+
(`<b>a</b> <i>b</i>`), and any text node holding a non-whitespace code point, is processed
|
|
556
|
+
as before. The accepted cost: no rule reaches the node's own characters any more, and a
|
|
557
|
+
line of its own that holds only ` - ` (`<p>a</p>`, a line feed, `<p> - </p>`, a line feed,
|
|
558
|
+
`<p>b</p>`) is now left alone, exactly as `text` mode leaves it between two line feeds,
|
|
559
|
+
where before 1.23.0 it became a dash.
|
|
543
560
|
|
|
544
561
|
**Element names compare case-insensitively**, per HTML: `<CODE>`, `<Code>` and `<code>` are all
|
|
545
562
|
skipped. (In MDX, JSX names compare case-sensitively — §3.7.3.)
|