polytypo 1.12.0 → 1.14.0

data/lib/polytypo/data/fixtures/sr.json

@@ -0,0 +1,183 @@
+{
+  "spec": "1.14.0",
+  "locale": "sr",
+  "cases": [
+    {
+      "id": "sr-quotes-primary",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "Рекао је: \"да\" и отишао.",
+      "out": "Рекао је: „да” и отишао.",
+      "note": "Klikovac 2024 on Pravopis 2010, pp. 343–344: the base form is „…” (U+201E, U+201D); the close is U+201D, not U+201C."
+    },
+    {
+      "id": "sr-quotes-primary-latin",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "Rekao je: \"da\" i otišao.",
+      "out": "Rekao je: „da” i otišao.",
+      "note": "No rule depends on the script: the Latin spelling gets the same pair."
+    },
+    {
+      "id": "sr-quotes-nested-secondary",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "Рекао је: \"песма 'Лепа' је\" и отишао.",
+      "out": "Рекао је: „песма ’Лепа’ је” и отишао.",
+      "note": "Klikovac 2024 on Pravopis 2010, p. 361: a quotation inside a quotation takes the single pair, written with U+2019 on both sides."
+    },
+    {
+      "id": "sr-quotes-nested-roundtrip",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "Рекао је: „песма ’Лепа’ је” и отишао.",
+      "out": "Рекао је: „песма ’Лепа’ је” и отишао.",
+      "note": "The same-glyph secondary pair is a fixed point."
+    },
+    {
+      "id": "sr-quotes-apostrophe-inside-secondary",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "Рекао је: \"песма 'McDonald's' је\" и отишао.",
+      "out": "Рекао је: „песма ’McDonald’s’ је” и отишао.",
+      "note": "An apostrophe inside the same-glyph secondary pair is kept and the pair still closes; the result is a fixed point."
+    },
+    {
+      "id": "sr-quotes-already-typeset-kept",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "„да” и „не”",
+      "out": "„да” и „не”",
+      "note": "Already typeset: left as it is."
+    },
+    {
+      "id": "sr-quotes-no-inner-space",
+      "rule": "quotes",
+      "mode": "text",
+      "in": "„ да ”",
+      "out": "„да”",
+      "note": "No space after the opening or before the closing mark."
+    },
+    {
+      "id": "sr-ellipsis-three-dots",
+      "rule": "ellipsis",
+      "mode": "text",
+      "in": "Чекај... шта?",
+      "out": "Чекај… шта?",
+      "note": "Pravopis 2010 (via Klikovac, p. 136): three dots, here the single ellipsis character."
+    },
+    {
+      "id": "sr-ellipsis-not-abbreviated",
+      "rule": "ellipsis",
+      "mode": "text",
+      "in": "Шта?.. да",
+      "out": "Шта?… да",
+      "note": "abbreviatedAfterTerminal is false (decision): `?..` is an ordinary ellipsis."
+    },
+    {
+      "id": "sr-dashes-parenthetical",
+      "rule": "dashes",
+      "mode": "text",
+      "in": "Идемо - ако пада - касније.",
+      "out": "Идемо – ако пада – касније.",
+      "note": "Klikovac p. 328: the separating dash is spaced; U+2013 is a decision."
+    },
+    {
+      "id": "sr-dashes-existing-em-dash",
+      "rule": "dashes",
+      "mode": "text",
+      "in": "Идемо — ако пада — касније.",
+      "out": "Идемо – ако пада – касније.",
+      "note": "An authored em dash becomes the locale's en dash."
+    },
+    {
+      "id": "sr-ranges-digits-tight",
+      "rule": "ranges",
+      "mode": "text",
+      "in": "рат 1941-1945 и стр. 145-162",
+      "out": "рат 1941⁠–⁠1945 и стр. 145⁠–⁠162",
+      "note": "ranges.md §3, Klikovac pp. 330–331: a dash between numbers is closed up («рат 1941–1945»). en-tight inserts U+2060 on both sides of the dash so a line cannot break there (ranges.md).",
+      "rules": {
+        "ranges": true
+      }
+    },
+    {
+      "id": "sr-ranges-already-tight",
+      "rule": "ranges",
+      "mode": "text",
+      "in": "рат 1941–1945",
+      "out": "рат 1941–1945",
+      "note": "ranges.md §3: already an en dash between numbers: left as it is.",
+      "rules": {
+        "ranges": true
+      }
+    },
+    {
+      "id": "sr-ranges-word-hyphen-untouched",
+      "rule": "ranges",
+      "mode": "text",
+      "in": "Нови Сад-Београд и српско-хрватски",
+      "out": "Нови Сад-Београд и српско-хрватски",
+      "note": "ranges.md §3: a hyphenated word is not a digit range.",
+      "rules": {
+        "ranges": true
+      }
+    },
+    {
+      "id": "sr-hyphen-noop",
+      "rule": "hyphen",
+      "mode": "text",
+      "in": "српско-хрватски речник и без-",
+      "out": "српско-хрватски речник и без-",
+      "note": "No closed list ships, so nothing changes."
+    },
+    {
+      "id": "sr-nbsp-units-inert",
+      "rule": "nbsp",
+      "mode": "text",
+      "in": "3 kg, 50 % и 99,90 €",
+      "out": "3 kg, 50 % и 99,90 €",
+      "note": "No verified rule: every nbsp list ships empty, so the plain spaces stay."
+    },
+    {
+      "id": "sr-nbsp-short-words-inert",
+      "rule": "nbsp",
+      "mode": "text",
+      "in": "у кући и с пријатељем",
+      "out": "у кући и с пријатељем",
+      "note": "afterShortWords ships empty."
+    },
+    {
+      "id": "sr-nbsp-initials-unbound",
+      "rule": "nbsp",
+      "mode": "text",
+      "in": "Б. Ћопић",
+      "out": "Б. Ћопић",
+      "note": "initialBinding is none."
+    },
+    {
+      "id": "sr-spaces-brackets",
+      "rule": "spaces",
+      "mode": "text",
+      "in": "Текст ( нормалан ) остаје  .",
+      "out": "Текст (нормалан) остаје.",
+      "note": "spaces.md: brackets tighten, no space before the final stop."
+    },
+    {
+      "id": "sr-symbols-copyright",
+      "rule": "symbols",
+      "mode": "text",
+      "in": "Copyright (c) 2026",
+      "out": "Copyright © 2026",
+      "note": "symbols rule, locale-neutral."
+    },
+    {
+      "id": "sr-apostrophe-foreign-name",
+      "rule": "apostrophe",
+      "mode": "text",
+      "in": "То је McDonald's",
+      "out": "То је McDonald’s",
+      "note": "apostrophe rule, locale-neutral."
+    }
+  ]
+}

data/lib/polytypo/data/fixtures/sv.json

@@ -1,5 +1,5 @@
 {
-  "spec": "1.12.0",
+  "spec": "1.14.0",
   "locale": "sv",
   "cases": [
     {

data/lib/polytypo/data/fixtures/tr.json

@@ -1,5 +1,5 @@
 {
-  "spec": "1.12.0",
+  "spec": "1.14.0",
   "locale": "tr",
   "cases": [
     {

data/lib/polytypo/data/fixtures/uk.json

@@ -1,5 +1,5 @@
 {
-  "spec": "1.12.0",
+  "spec": "1.14.0",
   "locale": "uk",
   "cases": [
     {

data/lib/polytypo/data/locales/registry.json

@@ -1,5 +1,5 @@
 {
-  "spec": "1.12.0",
+  "spec": "1.14.0",
   "$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.",
   "locales": [
     "en-US",
@@ -25,7 +25,8 @@
     "ro",
     "hu",
     "bg",
-    "hr"
+    "hr",
+    "sr"
   ],
   "aliases": {
     "en": "en-US",

data/lib/polytypo/data/locales/sr.json

@@ -0,0 +1,82 @@
+{
+  "locale": "sr",
+  "name": "Serbian",
+  "quotes": {
+    "primary": {
+      "open": "„",
+      "close": "”",
+      "innerSpace": "none"
+    },
+    "secondary": {
+      "open": "’",
+      "close": "’",
+      "innerSpace": "none"
+    },
+    "elisionIdioms": [],
+    "elisionClitics": {
+      "before": [],
+      "after": []
+    }
+  },
+  "dash": {
+    "parenthetical": "en-spaced",
+    "range": "en-tight"
+  },
+  "ellipsis": {
+    "abbreviatedAfterTerminal": false
+  },
+  "hyphen": {
+    "prefixes": [],
+    "suffixes": [],
+    "compounds": []
+  },
+  "nbsp": {
+    "beforePunctuation": [],
+    "narrowBeforePunctuation": [],
+    "afterShortWords": [],
+    "abbreviations": [],
+    "beforeUnits": [],
+    "beforeNumber": [],
+    "beforeWord": [],
+    "afterSymbols": [],
+    "initialBinding": "none"
+  },
+  "sources": [
+    {
+      "rule": "quotes",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, pp. 343–345 (table 5) and 360–363 (table 11)",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "Primary: U+201E open, U+201D close. p. 343, table 5: «Пижурица и др. 2010: „ ” (То је једини облик који је наведен у наслову.)»; p. 344: the 2010 Pravopis prescribes and uses only this form. Secondary: U+2019 on both sides (the same-glyph pair of fi and sv), p. 361, table 11: «Полунаводницима се издваја ужи навод у оквиру ширег навода» (2010:135); p. 363 fn. 30 notes the form ‘ ’ at 2010:135 as «вероватно грешком». innerSpace none for both. CONFLICT, recorded not resolved: Пешикан и др. 1993:286 also allowed „ “ and » «; the dictionaries РМС, РСАНУ and РСЈ use „ “ (p. 345); Klikovac and Milanović 2019 argue for „ “ or free choice. U+201D follows the current Pravopis, as hr does. Q-W: primary WIDE, secondary NARROW. The secondary U+2019/U+2019 is a DECISION (MEDIUM): table 11 attributes ’ ’ to 2010, but fn. 30 reports ‘ ’ (U+2018/U+2019) printed at 2010:135 and calls it probably a mistake without the note stating which form Klikovac holds correct; pinned as U+2019 both sides, the form table 11 gives. The issue text «„…“ / „…”» was an observation, not a norm. elisionIdioms and elisionClitics are empty: no rule found. Confidence MEDIUM-HIGH: the glyph form is named in words by Klikovac, not read from a text layer. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    },
+    {
+      "rule": "dashes",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, pp. 302, 320 (table 8), 328",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "dash.parenthetical = en-spaced. The spaces are normative: the separating dash is set as a spaced mark (Пешикан и др. 1993, p. 328; 2010:122 confirms the function and every example is set spaced). The WIDTH is a DECISION: p. 302 says the dash should be about twice as long as the hyphen or more, which does not choose between U+2013 and U+2014; the monograph sets U+2013. MEDIUM, as hr and bg. Klikovac's p. 328 cites «Пижурица и др. (1993: 126)», probably a misprint for 2010; noted, not corrected. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    },
+    {
+      "rule": "ranges",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, pp. 330–331",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "dash.range = en-tight. A dash between numbers is written closed up (Пешикан 1993:278, Пижурица 2010:126, Дешић 2015): «рат 1941–1945», «у XV–XVIII веку». ACCEPTED COSTS: a multi-word range member takes spaces in 2010:126–127 («рат 6. IV 1941 – 9. V 1945»), and the rule «од … до» (no dash after «од») is an algorithm, not data; locale data can express neither. The ranges rule is off by default. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    },
+    {
+      "rule": "ellipsis",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, p. 136 (table 1, fn. 5) and p. 164 (§12.1)",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "abbreviatedAfterTerminal = false is a DECISION, not a citation. 2010 gives three dots closed up (p. 136) and allows them after ? and ! (p. 164); the two-dot form after a terminal appears only as older practice (Мирко Николић 1973, p. 136 fn. 5) and is never prescribed. MEDIUM. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    },
+    {
+      "rule": "hyphen",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, (not consulted)",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "All three lists are empty: no source consulted gives a closed list of unbreakable hyphenated forms. The 2010 Pravopis chapters on word division were not consulted. Operator decision (no paid sources, pick and pin): ship empty. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    },
+    {
+      "rule": "nbsp",
+      "cite": "Д. Кликовац, Српска интерпункција, Филолошки факултет Универзитета у Београду, 2024 (серија «Монографије – језик»), ISBN 978-86-6153-752-3, analysing Пешикан/Јерковић/Пижурица, Правопис српскога језика, Матица српска, Нови Сад, 2010, (not consulted)",
+      "url": "https://www.fil.bg.ac.rs/uploads/files/Strane/Biblioteke/Izdavacka%20delatnost/Monografije%20-%20Jezik/SRSKA%20INTERPUNKCIJA%20elek%20izd.pdf",
+      "note": "Every list is empty and initialBinding is none: no source consulted gives a rule for a space between a number and a unit or percent sign, for one-letter words or initials at a line end. Nothing is derived from another language. The 2010 Pravopis chapters on numbers and word division were not consulted. Operator decision (no paid sources, pick and pin): ship empty. The Pravopis 2010 itself was not read (paid); every statement is the edition of 2010 as Klikovac reports it. The PDF has no text layer and was read as page images. Retrieved 2026-10-24."
+    }
+  ]
+}

data/lib/polytypo/data/rules/analyze.md

@@ -3,7 +3,8 @@
 **Not a rule.** No entry in `spec/rules/order.json`, no locale data of its own, no edits. This
 document specifies a **second public entry point** beside `transform`, which runs the identical
 pipeline and reports what it would do instead of doing it.
-**Spec version:** 1.3.0 (new in 1.3.0).
+**Spec version:** 1.13.0 (new in 1.3.0; `end` of a non-insertion defined in §2, and the matching
+obligation in §6, in 1.13.0).
 
 ---
 
@@ -23,7 +24,7 @@ transform(input, options) -> string
 analyze(input, options)   -> Change[]
 ```
 
-Same `options`, same validation, same errors, same purity (`ARCHITECTURE.md` §7: no I/O, no
+Same `options` (`protect` included, modes.md §3.9), same validation, same errors, same purity (`ARCHITECTURE.md` §7: no I/O, no
 clock, no globals, reentrant). Everything §2 of every rule document says about what the pipeline
 does applies unchanged — `analyze` **is** the pipeline; it merely keeps the edits instead of
 discarding them after applying.
@@ -49,8 +50,10 @@ Change {
   whose strings are UTF-16 must convert; a runtime whose strings are bytes must convert.
 - **Offsets are into `input` exactly as the caller passed it.** In `html`, `markdown` and
   `yaml` mode that means offsets into the **document**, not into the span the rules actually ran
-  over. This
-  is not a new obligation: [modes.md](modes.md) §4 already defines the output as "the input
+  over. A non-insertion's `end` is one past the offset of the last code point it replaced that
+  came from the input, so it can never run across a skipped region or a protected run that
+  follows (since 1.13.0; 1.12.0 scanned forward to the next code point with an origin). The
+  document-offset rule is not a new obligation: [modes.md](modes.md) §4 already defines the output as "the input
   source with a set of disjoint substring replacements applied **at recorded offsets**", and
   those are the offsets meant.
 - `start == end` is a pure insertion; `before` is then empty. `after` empty with `end > start`
@@ -137,6 +140,15 @@ naming:
 
 - **A3 against the whole fixture corpus.** For every canonical fixture case, `analyze` returns
   an empty list exactly when `in == out`. That is a strong test and it costs one loop.
+- **A range that crosses nothing it did not change (since 1.13.0).** A change's `start` and `end`
+  come from the code points it replaced, and `end` is not found by scanning forward from the
+  edit's end index to the next code point that has an origin: the boundary marker has none, so a
+  change ending beside a skipped element or a protected run (modes.md §3.9) would be carried
+  across it. Test a change whose last code point is a span's last, followed by a skipped region
+  longer than the change: no reported range may overlap a skipped region or a protected run, and
+  a non-insertion's `end` is one past the input offset of the last replaced code point that has
+  one. (`end - start` need not equal the length of `before`: `before` is what a later rule saw,
+  §5.) 1.12.0 reported `<p>x--<code>cc</code>y</p>` as `start` 4, `end` 21.
 - **A5 under the mode adapters.** A runtime that reports span-local offsets in `html`,
   `markdown` or `yaml` mode passes every text-mode test and is still wrong. Test a document
   whose first span does not start at offset 0. `yaml` is the cheapest of the three to get wrong