@heroiclands/package-build 23.13.0 → 24.0.0

package/engine/language-words.mjs

@@ -24,6 +24,7 @@
  *   each declared in `data.classModifiers`, and a declared modifier no word
  *   uses is a warning;
  * - a word's `field` is one of `data.fields`;
+ * - a word's `concept` names entries of the concept list, each once;
  * - a `borrowed` source names its `language`;
  * - no form stands twice in one language;
  * - a word's `gloss` states at least one sense;
@@ -61,6 +62,7 @@ import {
     pronunciationProblem,
 } from "./word-terms.mjs";
 import { WORD_CLASS_MODIFIERS, parseWordClass } from "./word-classes.mjs";
+import { conceptIdsIn, conceptProblems } from "./core-vocabulary.mjs";
 import { lineageChain, pieceGlosses } from "./word-rows.mjs";
 
 const MODIFIER_TOKENS = new Set(WORD_CLASS_MODIFIERS.map((item) => item.token));
@@ -192,6 +194,7 @@ export function checkLanguageWords(note, { index } = {}) {
     }
     const used = new Set();
     const forms = new Map();
+    const concepts = conceptIdsIn(index);
     listOf(data.words).forEach((word, i) => {
         if (!isMap(word)) return;
         const at = (key) => ["data", "words", i, key];
@@ -226,6 +229,15 @@ export function checkLanguageWords(note, { index } = {}) {
                 ),
             );
 
+        for (const problem of conceptProblems(word.concept, concepts))
+            findings.push(
+                finding(
+                    problem.entry === null ? at("concept") : [...at("concept"), problem.entry],
+                    "error",
+                    problem.message,
+                ),
+            );
+
         if (typeof word.form === "string") {
             const key = word.form.normalize("NFC");
             if (forms.has(key))

package/engine/logographic.mjs

@@ -0,0 +1,255 @@
+/*
+ * This file is part of the Song of Heroic Lands (SoHL) system for Foundry VTT.
+ * Copyright (c) 2024-2026 Tom Rodriguez ("Toasty") — <toasty@heroiclands.org>
+ *
+ * This work is licensed under the GNU General Public License v3.0 (GPLv3).
+ * You may copy, modify, and distribute it under the terms of that license.
+ *
+ * For full terms, see the LICENSE.md file in the project root or visit:
+ * https://www.gnu.org/licenses/gpl-3.0.html
+ *
+ * SPDX-License-Identifier: GPL-3.0-or-later
+ */
+
+/**
+ * **A logographic map writes meanings, so a word's writing comes from the
+ * lexicon and never from its sound.** A mapping note stating
+ * `data.map.logographic: true` writes each word of a run in this order:
+ *
+ * 1. **The word's own `native` entry** for the map, and for the run's face
+ *    where an entry names one.
+ * 2. **Otherwise its parts.** A word stating `built` is written by the
+ *    writings of its pieces joined in order: a piece is a word of the
+ *    lexicon, matched exactly and then with its joining hyphens taken off, or
+ *    an affix of `data.affixes` with its own `native`; a word piece resolves
+ *    by this same rule. A noun's plural and a verb's present forms are written
+ *    by the stem's writing followed by the ending's, which
+ *    `data.grammar.native` states by the ending's name: `plural`, or one of
+ *    the six persons. A hyphenated word no entry states is written by its
+ *    parts, each by this rule.
+ * 3. **Otherwise the word has no writing**, which a run reports at the word.
+ *
+ * A run's word is matched case folded, so one roman form may stand for more
+ * than one entry. Where those entries are written alike the word takes that
+ * writing; where they differ, or one has a writing and another none, the word
+ * is ambiguous, and the finding names every entry's gloss. A part matched to
+ * entries written differently makes the word ambiguous the same way.
+ *
+ * @module
+ */
+
+import { pluralOf } from "./dictionary-rows.mjs";
+import { verbForm, verbPattern } from "./verb-pattern.mjs";
+import { parseWordClass } from "./word-classes.mjs";
+import { foldWord } from "./word-forms.mjs";
+import { glossText } from "./word-terms.mjs";
+
+/** Whether a value is a plain map. */
+const isMap = (value) => value !== null && typeof value === "object" && !Array.isArray(value);
+
+/** A list value, or an empty one. */
+const listOf = (value) => (Array.isArray(value) ? value : []);
+
+/** A form with its joining hyphens taken off. */
+const bare = (form) => form.replace(/^[-–]+|[-–]+$/gu, "");
+
+/** The characters a hyphenated word is split at. */
+const JOINS = /[–-]/u;
+
+/**
+ * Where a word's writing comes from, or why it has none.
+ *
+ * @typedef {{text: string, source: "own"|"parts"}
+ *   | {missing: true, noLexicon?: true}
+ *   | {ambiguous: {form: string, glosses: string[]}}} Writing
+ */
+
+/**
+ * The text a list of native entries states for a map in a face: the entry for
+ * the face, else the entry for no face.
+ *
+ * @param {unknown} natives - A `native` list as authored.
+ * @param {string} map - The map's name.
+ * @param {string|undefined} font - The face.
+ * @returns {string|undefined} The text.
+ */
+export function nativeText(natives, map, font) {
+    const entries = listOf(natives).filter(
+        (entry) => isMap(entry) && entry.map === map && typeof entry.text === "string",
+    );
+    const own = font === undefined ? undefined : entries.find((entry) => entry.font === font);
+    return (own ?? entries.find((entry) => typeof entry.font !== "string"))?.text;
+}
+
+/**
+ * One writing for several candidates: the shared one where every candidate is
+ * written alike, else an ambiguity naming each candidate's gloss.
+ *
+ * @param {string} form - The form the candidates were matched by.
+ * @param {Array<{writing: Writing, gloss: string}>} candidates - At least one.
+ * @returns {Writing} The writing.
+ */
+function oneOf(form, candidates) {
+    const passed = candidates.find(({ writing }) => "ambiguous" in writing);
+    if (passed) return passed.writing;
+    const texts = new Set(
+        candidates.map(({ writing }) => ("text" in writing ? writing.text : undefined)),
+    );
+    if (texts.size === 1) return candidates[0].writing;
+    return { ambiguous: { form, glosses: candidates.map(({ gloss }) => gloss) } };
+}
+
+/**
+ * The writings of a language's words under one logographic map in one face.
+ *
+ * @param {Record<string, any>|undefined} data - The language note's `data`.
+ * @param {string} map - The map's name.
+ * @param {string|undefined} font - The face.
+ * @returns {{word: (word: string) => Writing, entry: (index: number) => Writing}}
+ *   The writing of a run's word, and of the `data.words` entry at an index.
+ */
+export function writingsOf(data, map, font) {
+    const words = listOf(data?.words);
+    const affixes = listOf(data?.affixes).filter(isMap);
+    const grammar = isMap(data?.grammar) ? data.grammar : undefined;
+    const verbs = verbPattern(grammar?.verbs);
+    const replaces = verbs?.replaces === undefined ? undefined : foldWord(verbs.replaces);
+    const endings = grammar?.native;
+    const indexed = words
+        .map((word, index) => ({ word, index }))
+        .filter(({ word }) => isMap(word) && typeof word.form === "string");
+    const glossOf = (entry) => glossText(entry.gloss) ?? String(entry.form ?? "");
+
+    /**
+     * @param {number} index - A `data.words` index.
+     * @param {Set<number>} visiting - The entries whose parts are being read.
+     * @returns {Writing}
+     */
+    const entry = (index, visiting) => {
+        const word = words[index];
+        const own = nativeText(word?.native, map, font);
+        if (own !== undefined) return { text: own, source: "own" };
+        const pieces = listOf(word?.built).filter((piece) => typeof piece === "string");
+        if (!pieces.length || visiting.has(index)) return { missing: true };
+        const inner = new Set(visiting).add(index);
+        let text = "";
+        for (const piece of pieces) {
+            const found = part(piece, inner);
+            if (!("text" in found)) return found;
+            text += found.text;
+        }
+        return { text, source: "parts" };
+    };
+
+    /** The candidates of a list of matched words and affixes. */
+    const candidatesOf = (matchedWords, matchedAffixes, visiting) => [
+        ...matchedWords.map(({ word, index }) => ({
+            writing: entry(index, visiting),
+            gloss: glossOf(word),
+        })),
+        ...matchedAffixes.map((affix) => {
+            const text = nativeText(affix.native, map, font);
+            return {
+                writing: /** @type {Writing} */ (
+                    text === undefined ? { missing: true } : { text, source: "own" }
+                ),
+                gloss: glossOf(affix),
+            };
+        }),
+    ];
+
+    /**
+     * A `built` piece: a word written exactly so, else an affix, else either
+     * with its joining hyphens taken off.
+     *
+     * @param {string} piece - The piece.
+     * @param {Set<number>} visiting - The entries whose parts are being read.
+     * @returns {Writing}
+     */
+    const part = (piece, visiting) => {
+        const affixForms = (affix) =>
+            [affix.form, affix.front].filter((form) => typeof form === "string");
+        const stages = [
+            [indexed.filter(({ word }) => word.form === piece), []],
+            [[], affixes.filter((affix) => affixForms(affix).includes(piece))],
+            [indexed.filter(({ word }) => bare(word.form) === bare(piece)), []],
+            [[], affixes.filter((affix) => affixForms(affix).some((f) => bare(f) === bare(piece)))],
+        ];
+        for (const [matchedWords, matchedAffixes] of stages) {
+            const candidates = candidatesOf(matchedWords, matchedAffixes, visiting);
+            if (candidates.length) return oneOf(piece, candidates);
+        }
+        return { missing: true };
+    };
+
+    /**
+     * The ending's writing a derived form takes after its stem's.
+     *
+     * @param {string} ending - A name of
+     *   {@link module:engine/language-terms.WRITTEN_ENDINGS}.
+     * @returns {string|undefined} The text.
+     */
+    const endingText = (ending) =>
+        nativeText(
+            listOf(endings).filter((one) => isMap(one) && one.ending === ending),
+            map,
+            font,
+        );
+
+    /**
+     * A word of a run, matched case folded.
+     *
+     * @param {string} written - The word as written.
+     * @returns {Writing}
+     */
+    const word = (written) => {
+        const folded = foldWord(written);
+        const none = new Set();
+        const stated = indexed.filter(({ word: one }) => foldWord(one.form) === folded);
+        if (stated.length) return oneOf(written, candidatesOf(stated, [], none));
+        const stripped = indexed.filter(({ word: one }) => bare(foldWord(one.form)) === folded);
+        if (stripped.length) return oneOf(written, candidatesOf(stripped, [], none));
+        const derived = [];
+        for (const { word: one, index } of indexed) {
+            const base = parseWordClass(one.class).base;
+            const plural = pluralOf(one, grammar);
+            const forms = [
+                ...(plural ? [{ form: plural, ending: "plural" }] : []),
+                ...(verbs && base === "v" ?
+                    verbs.endings.map(({ name, ending }) => ({
+                        form: verbForm(foldWord(one.form), foldWord(ending), replaces),
+                        ending: name,
+                    }))
+                :   []),
+            ];
+            for (const { form, ending } of forms) {
+                if (foldWord(form) !== folded) continue;
+                const stem = entry(index, none);
+                const tail = endingText(ending);
+                derived.push({
+                    writing: /** @type {Writing} */ (
+                        !("text" in stem) ? stem
+                        : tail === undefined ? { missing: true }
+                        : { text: stem.text + tail, source: "parts" }
+                    ),
+                    gloss: glossOf(one),
+                });
+            }
+        }
+        if (derived.length) return oneOf(written, derived);
+        if (JOINS.test(written)) {
+            let text = "";
+            for (const piece of written.split(JOINS)) {
+                if (!piece) continue;
+                const found = word(piece);
+                if ("missing" in found) return { missing: true };
+                if ("ambiguous" in found) return found;
+                text += found.text;
+            }
+            return { text, source: "parts" };
+        }
+        return { missing: true };
+    };
+
+    return { word, entry: (index) => entry(index, new Set()) };
+}