inkhaven 1.5.8

Inkhaven — TUI literary work editor for Typst books
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249

//! INNER-THEOLOGIAN-1 (IT-P1) — per-language fast-track vocabulary, five
//! languages. Three tradition-neutral lists: lethal/severe **violence**,
//! moral **consequence/acknowledgment**, and **sacred/ritual** vocabulary. All
//! lowercased; matching is case-insensitive. Single tokens match on word
//! boundaries; multi-word entries (e.g. "could not forget") match as substrings.
//!
//! No tradition is weighted more heavily than any other within a language's
//! sacred list — all eleven contribute terms. Patterned on CHAR-1's action-verb
//! lists and DIALOG-1's said-bookism lists (linear scan, not binary search:
//! Unicode-correct sorting of Cyrillic/accented literals is error-prone and the
//! lists are short).

use crate::prose::ProseLanguage;

pub(crate) struct VocabLists {
    pub violence: &'static [&'static str],
    pub consequence: &'static [&'static str],
    pub sacred: &'static [&'static str],
    /// Comic / dismissive context markers — Signal 3 fires when a sacred term
    /// shares a paragraph with one of these (the most cautious detector).
    pub levity: &'static [&'static str],
}

/// The three lists for a language. `Other` falls back to English.
pub(crate) fn lists_for(lang: &ProseLanguage) -> &'static VocabLists {
    match lang {
        ProseLanguage::En => &EN,
        ProseLanguage::Ru => &RU,
        ProseLanguage::De => &DE,
        ProseLanguage::Fr => &FR,
        ProseLanguage::Es => &ES,
        ProseLanguage::Other(_) => &EN,
    }
}

/// Return the first list entry present in `text_lc` (which the caller has
/// already lowercased). A single-token entry matches only on a whole-word
/// boundary; a multi-word entry matches as a substring.
pub(crate) fn scan_list(text_lc: &str, list: &[&'static str]) -> Option<&'static str> {
    for &term in list {
        if term.contains(' ') {
            if text_lc.contains(term) {
                return Some(term);
            }
        } else if contains_word(text_lc, term) {
            return Some(term);
        }
    }
    None
}

/// Whole-word, Unicode-aware containment: `word` appears in `text_lc` delimited
/// by non-alphanumeric boundaries (so "killed" doesn't match "skilled").
fn contains_word(text_lc: &str, word: &str) -> bool {
    text_lc
        .split(|c: char| !c.is_alphanumeric())
        .any(|tok| tok == word)
}

// ── English ───────────────────────────────────────────────────────────────────

static EN: VocabLists = VocabLists {
    violence: &[
        "killed", "murdered", "slain", "executed", "died", "shot", "stabbed", "struck",
        "beaten", "tortured", "destroyed", "massacred", "drowned", "hanged", "burned",
        "starved", "violated", "crushed", "wounded", "severed", "impaled", "strangled",
        "sacrificed", "perished", "slaughtered",
    ],
    consequence: &[
        "grief", "guilt", "mourned", "wept", "anguish", "remorse", "regret", "responsible",
        "reckoned", "consequence", "cost", "burden", "debt", "owed", "atoned", "forgave",
        "forgiven", "acknowledged", "bore", "paid", "answered", "haunted",
        "could not forget", "could not escape",
    ],
    sacred: &[
        "sacred", "holy", "blessed", "divine", "prayer", "ritual", "ceremony", "sacrifice",
        "grace", "sin", "soul", "spirit", "god", "goddess", "temple", "altar", "covenant",
        "revelation", "scripture", "prophet", "imam", "rabbi", "priest", "monk", "baptism",
        "eucharist", "mosque", "synagogue", "enlightenment", "nirvana", "karma", "dharma",
        "moksha", "ahimsa", "gnosis", "pleroma", "demiurge", "theosis", "kenosis",
        "atonement", "redemption", "salvation", "damnation", "purgatory", "paradise",
        "resurrection", "incarnation", "consecrated", "anointed", "eternal", "numinous",
    ],
    levity: &[
        "laughed", "laughing", "joke", "joked", "grinned", "chuckled", "snorted", "smirked",
        "giggled", "mocked", "mockingly", "ridiculous", "absurd", "sarcastic", "sarcastically",
        "ironic", "winked", "teasing", "jest", "silly",
    ],
};

// ── Russian ───────────────────────────────────────────────────────────────────

static RU: VocabLists = VocabLists {
    violence: &[
        "убил", "убила", "убит", "убита", "погиб", "погибла", "умер", "умерла", "расстрелян",
        "зарезал", "задушил", "сжёг", "утопил", "повесил", "разрушил", "уничтожил", "истязал",
        "замучил", "погубил", "растерзал", "пал", "пронзил", "истребил",
    ],
    consequence: &[
        "горе", "вина", "скорбь", "плакал", "плакала", "сожаление", "раскаяние",
        "ответственность", "расплата", "долг", "тяжесть", "бремя", "искупление", "простил",
        "простила", "признал", "признала", "понёс", "понесла", "заплатил",
    ],
    sacred: &[
        "святой", "священный", "молитва", "душа", "бог", "богиня", "храм", "алтарь", "обряд",
        "ритуал", "жертва", "благодать", "грех", "карма", "дхарма", "нирвана", "просветление",
        "сансара", "ахимса", "гнозис", "теозис", "кенозис", "соборность", "покаяние", "мицва",
        "тора", "шаббат", "намаз", "джихад", "спасение", "исповедь", "причастие", "воскресение",
        "пророчество", "откровение", "вечность", "икона", "благословение", "завет", "искупление",
    ],
    levity: &[
        "смеялся", "смеялась", "шутка", "пошутил", "ухмыльнулся", "хихикнул", "усмехнулся",
        "насмешливо", "нелепо", "абсурд", "иронично", "подмигнул", "дразнил", "смешно",
    ],
};

// ── German ────────────────────────────────────────────────────────────────────

static DE: VocabLists = VocabLists {
    violence: &[
        "tötete", "erschoss", "erschlug", "erdrosselte", "ertränkte", "verbrannte",
        "hinrichtete", "starb", "zerstörte", "vernichtete", "marterte", "folterte",
        "schlachtete", "ermordete", "durchbohrte", "opferte", "verhungerte", "ertrank",
        "erhängte", "verstümmelte",
    ],
    consequence: &[
        "trauer", "schuld", "reue", "weinte", "klagte", "verantwortung", "konsequenz", "last",
        "bürde", "sühne", "vergab", "gestand", "trug", "zahlte", "büßte", "anerkannte",
    ],
    sacred: &[
        "heilig", "sakral", "gebet", "seele", "gott", "göttin", "tempel", "altar", "ritus",
        "opfer", "gnade", "sünde", "karma", "dharma", "nirwana", "erleuchtung", "gnosis",
        "theosis", "kenosis", "thora", "namaz", "dschihad", "erlösung", "buße", "beichte",
        "abendmahl", "auferstehung", "prophezeiung", "offenbarung", "ewigkeit", "ikone",
        "weihe", "segen", "fluch", "bund", "sühne",
    ],
    levity: &[
        "lachte", "witz", "scherzte", "grinste", "kicherte", "schmunzelte", "spöttisch",
        "lächerlich", "absurd", "ironisch", "zwinkerte", "neckte", "albern",
    ],
};

// ── French ────────────────────────────────────────────────────────────────────

static FR: VocabLists = VocabLists {
    violence: &[
        "tua", "abattit", "exécuta", "étrangla", "noya", "brûla", "mourut", "détruisit",
        "anéantit", "massacra", "tortura", "poignarda", "pendit", "fusilla", "sacrifia",
        "périt", "affama", "mutila",
    ],
    consequence: &[
        "deuil", "culpabilité", "remords", "pleura", "chagrin", "regret", "responsabilité",
        "conséquence", "poids", "fardeau", "dette", "expiation", "pardonna", "reconnut",
        "porta", "paya", "ne pouvait oublier", "le hantait",
    ],
    sacred: &[
        "sacré", "saint", "prière", "âme", "dieu", "déesse", "temple", "autel", "rite",
        "sacrifice", "grâce", "péché", "karma", "dharma", "nirvana", "éveil", "gnose", "théose",
        "kénose", "tikoun", "torah", "namaz", "djihad", "salut", "pénitence", "confession",
        "eucharistie", "résurrection", "prophétie", "révélation", "éternité", "icône",
        "consécration", "bénédiction", "malédiction", "alliance", "expiation",
    ],
    levity: &[
        "rit", "rire", "blague", "plaisanta", "sourit", "ricana", "gloussa", "moqueur",
        "ridicule", "absurde", "ironique", "cligna", "taquina", "drôle",
    ],
};

// ── Spanish ───────────────────────────────────────────────────────────────────

static ES: VocabLists = VocabLists {
    violence: &[
        "mató", "asesinó", "ejecutó", "estranguló", "ahogó", "quemó", "murió", "destruyó",
        "masacró", "torturó", "apuñaló", "colgó", "fusiló", "sacrificó", "pereció", "mutiló",
        "feneció", "extinguió",
    ],
    consequence: &[
        "duelo", "culpa", "remordimiento", "lloró", "pena", "arrepentimiento", "responsabilidad",
        "consecuencia", "peso", "carga", "deuda", "expiación", "perdonó", "reconoció", "cargó",
        "pagó", "no podía olvidar", "le perseguía",
    ],
    sacred: &[
        "sagrado", "santo", "oración", "alma", "dios", "diosa", "templo", "altar", "rito",
        "sacrificio", "gracia", "pecado", "karma", "dharma", "nirvana", "iluminación", "gnosis",
        "teosis", "kénosis", "tikún", "torá", "namaz", "yihad", "salvación", "penitencia",
        "confesión", "eucaristía", "resurrección", "profecía", "revelación", "eternidad", "icono",
        "consagración", "bendición", "maldición", "alianza", "expiación",
    ],
    levity: &[
        "rió", "risa", "broma", "bromeó", "sonrió", "se burló", "burlonamente", "ridículo",
        "absurdo", "irónico", "guiñó", "bromeando", "gracioso",
    ],
};

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn all_languages_have_three_nonempty_lists() {
        for lang in [
            ProseLanguage::En,
            ProseLanguage::Ru,
            ProseLanguage::De,
            ProseLanguage::Fr,
            ProseLanguage::Es,
        ] {
            let l = lists_for(&lang);
            assert!(l.violence.len() >= 15, "violence too short for {lang:?}");
            assert!(l.consequence.len() >= 14, "consequence too short for {lang:?}");
            assert!(l.sacred.len() >= 30, "sacred too short for {lang:?}");
            assert!(l.levity.len() >= 12, "levity too short for {lang:?}");
            for &w in l.violence.iter().chain(l.consequence).chain(l.sacred).chain(l.levity) {
                assert_eq!(w, w.to_lowercase(), "`{w}` must be lowercase");
            }
        }
    }

    #[test]
    fn other_language_falls_back_to_english() {
        let l = lists_for(&ProseLanguage::Other("pl".into()));
        assert!(scan_list("the soldier killed the man", l.violence).is_some());
    }

    #[test]
    fn whole_word_match_not_substring() {
        let l = lists_for(&ProseLanguage::En);
        // "killed" matches; "skilled" must not.
        assert_eq!(scan_list("he killed the guard", l.violence), Some("killed"));
        assert_eq!(scan_list("a skilled archer", l.violence), None);
    }

    #[test]
    fn multiword_consequence_phrase_matches_as_substring() {
        let l = lists_for(&ProseLanguage::En);
        assert_eq!(scan_list("she could not forget it", l.consequence), Some("could not forget"));
        assert!(scan_list("he felt grief", l.consequence).is_some());
    }

    #[test]
    fn cyrillic_and_accented_match() {
        let ru = lists_for(&ProseLanguage::Ru);
        assert_eq!(scan_list("солдат убил пленного", ru.violence), Some("убил"));
        let es = lists_for(&ProseLanguage::Es);
        assert!(scan_list("el soldado mató al hombre", es.violence).is_some());
        let sacred_fr = lists_for(&ProseLanguage::Fr);
        assert!(scan_list("une prière silencieuse", sacred_fr.sacred).is_some());
    }
}