@jjlmoya/utils-nature 1.22.0 → 1.24.0

package/src/tests/spanish_leakage.test.ts

@@ -1,175 +1,175 @@
-import { describe, expect, it } from 'vitest';
-import { ALL_ENTRIES } from '../entries';
-
-const STRUCTURAL_KEYS = new Set([
-  'id',
-  'slug',
-  'url',
-  'icon',
-  'image',
-  'imageUrl',
-  'keywords',
-  'category',
-  'tags',
-  'toolId',
-  'type',
-  'locale',
-  'language',
-  'direction',
-]);
-
-const TRANSLATABLE_KEYS = [
-  'title',
-  'description',
-  'faqTitle',
-  'faq',
-  'howTo',
-  'seo',
-  'schemas',
-] as const;
-
-const COPY_THRESHOLD = 0.9;
-
-const SPANISH_MARKERS = [
-  ['sangre', /\bsangre\b/gi],
-  ['molino', /\bmolino\b/gi],
-  ['grano', /\bgrano\b/gi],
-  ['paladar', /\bpaladar\b/gi],
-  ['cuchillas', /\bcuchillas\b/gi],
-  ['trozos', /\btrozos\b/gi],
-  ['frescura', /\bfrescura\b/gi],
-  ['ingresa', /\bingresa\b/gi],
-  ['selecciona', /\bselecciona\b/gi],
-  ['herramienta', /\bherramienta\b/gi],
-  ['según', /\bsegún\b/gi],
-  ['después', /\bdespués\b/gi],
-  ['puedes', /\bpuedes\b/gi],
-  ['debes', /\bdebes\b/gi],
-  ['tus', /\btus\b/gi],
-  ['a la vez', /\ba la vez\b/gi],
-  ['los datos', /\blos datos\b/gi],
-  ['las opciones', /\blas opciones\b/gi],
-  ['el resultado', /\bel resultado\b/gi],
-  ['método de extracción', /\bmétodo de extracción\b/gi],
-  ['uniformidad del molino', /\buniformidad del molino\b/gi],
-  ['café recién tostado', /\bcafé recién tostado\b/gi],
-] as const;
-
-type UnknownRecord = Record<string, unknown>;
-
-function normalize(value: string): string {
-  return value
-    .replace(/<[^>]*>/g, ' ')
-    .replace(/&nbsp;/gi, ' ')
-    .replace(/\s+/g, ' ')
-    .trim()
-    .toLocaleLowerCase('es');
-}
-
-function isTechnicalInvariant(text: string): boolean {
-  const ledUnitMatches = text.match(/\b\d+(?:-\d+)?\s*(?:w|lm)\b/g) ?? [];
-  return [
-    'data:image/svg+xml;base64',
-    'background-image: url',
-    '.layout-playground {',
-    'const samplerate =',
-  ].some((pattern) => text.includes(pattern)) ||
-    (text.includes('presets') && text.includes('hz')) ||
-    (text.includes('t_rectal') && text.includes('exp(-k * t)')) ||
-    (text.includes('led') && ledUnitMatches.length >= 3);
-}
-
-function collectString(value: string, output: string[]): void {
-  const text = normalize(value);
-  if (!isTechnicalInvariant(text) && text.length >= 20) output.push(text);
-}
-
-function collectObject(value: UnknownRecord, output: string[]): void {
-  Object.entries(value).forEach(([childKey, childValue]) => {
-    collectText(childValue, output, childKey);
-  });
-}
-
-function collectText(value: unknown, output: string[], key?: string): void {
-  if (key && STRUCTURAL_KEYS.has(key)) return;
-  if (typeof value === 'string') return collectString(value, output);
-  if (Array.isArray(value)) return value.forEach((item) => collectText(item, output));
-  if (value && typeof value === 'object') collectObject(value as UnknownRecord, output);
-}
-
-async function loadText(loader: unknown): Promise<string[]> {
-  if (typeof loader !== 'function') return [];
-  const module = await (loader as () => Promise<unknown>)();
-  const output: string[] = [];
-  if (module && typeof module === 'object') {
-    const record = module as UnknownRecord;
-    for (const key of TRANSLATABLE_KEYS) {
-      collectText(record[key], output, key);
-    }
-  }
-  return output;
-}
-
-function findSpanishMarkers(text: string[]): string[] {
-  const corpus = text.join(' ');
-  return SPANISH_MARKERS.flatMap(([label, pattern]) =>
-    pattern.test(corpus) ? [label] : [],
-  );
-}
-
-function similarity(left: string, right: string): number {
-  const leftTokens = left.split(/\s+/);
-  const rightCounts = new Map<string, number>();
-  right.split(/\s+/).forEach((token) => rightCounts.set(token, (rightCounts.get(token) ?? 0) + 1));
-  const matches = leftTokens.reduce((total, token) => {
-    const count = rightCounts.get(token) ?? 0;
-    if (count > 0) rightCounts.set(token, count - 1);
-    return total + (count > 0 ? 1 : 0);
-  }, 0);
-  return (2 * matches) / (leftTokens.length + right.split(/\s+/).length);
-}
-
-function findCopiedFragments(spanish: string[], translated: string[]): string[] {
-  return spanish
-    .filter((fragment) => fragment.length >= 80)
-    .filter((fragment) => translated.some((candidate) =>
-      candidate.length >= 80 &&
-      Math.min(fragment.length, candidate.length) / Math.max(fragment.length, candidate.length) >= COPY_THRESHOLD &&
-      similarity(fragment, candidate) >= COPY_THRESHOLD,
-    ))
-    .sort((a, b) => b.length - a.length)
-    .slice(0, 3);
-}
-
-describe('Locales must not contain copied Spanish content', () => {
-  for (const entry of ALL_ENTRIES) {
-    it(`${entry.id} has no untranslated Spanish blocks`, async () => {
-      const spanish = await loadText(entry.i18n.es);
-      const failures: string[] = [];
-
-      for (const [locale, loader] of Object.entries(entry.i18n)) {
-        if (locale === 'es') continue;
-
-        const translated = await loadText(loader);
-        const copiedFragments = findCopiedFragments(spanish, translated);
-        const markerHits = findSpanishMarkers(translated);
-
-        if (copiedFragments.length > 0 || markerHits.length >= 2) {
-          const details = [
-            copiedFragments.length > 0
-              ? `copied fragments: ${copiedFragments
-                  .map((fragment) => JSON.stringify(fragment.slice(0, 120)))
-                  .join(', ')}`
-              : '',
-            markerHits.length >= 2 ? `Spanish markers: ${markerHits.join(', ')}` : '',
-          ]
-            .filter(Boolean)
-            .join('; ');
-          failures.push(`${locale}: ${details}`);
-        }
-      }
-
-      expect(failures, failures.join('\n')).toEqual([]);
-    });
-  }
-});
+import { describe, expect, it } from 'vitest';
+import { ALL_ENTRIES } from '../entries';
+
+const STRUCTURAL_KEYS = new Set([
+  'id',
+  'slug',
+  'url',
+  'icon',
+  'image',
+  'imageUrl',
+  'keywords',
+  'category',
+  'tags',
+  'toolId',
+  'type',
+  'locale',
+  'language',
+  'direction',
+]);
+
+const TRANSLATABLE_KEYS = [
+  'title',
+  'description',
+  'faqTitle',
+  'faq',
+  'howTo',
+  'seo',
+  'schemas',
+] as const;
+
+const COPY_THRESHOLD = 0.9;
+
+const SPANISH_MARKERS = [
+  ['sangre', /\bsangre\b/gi],
+  ['molino', /\bmolino\b/gi],
+  ['grano', /\bgrano\b/gi],
+  ['paladar', /\bpaladar\b/gi],
+  ['cuchillas', /\bcuchillas\b/gi],
+  ['trozos', /\btrozos\b/gi],
+  ['frescura', /\bfrescura\b/gi],
+  ['ingresa', /\bingresa\b/gi],
+  ['selecciona', /\bselecciona\b/gi],
+  ['herramienta', /\bherramienta\b/gi],
+  ['según', /\bsegún\b/gi],
+  ['después', /\bdespués\b/gi],
+  ['puedes', /\bpuedes\b/gi],
+  ['debes', /\bdebes\b/gi],
+  ['tus', /\btus\b/gi],
+  ['a la vez', /\ba la vez\b/gi],
+  ['los datos', /\blos datos\b/gi],
+  ['las opciones', /\blas opciones\b/gi],
+  ['el resultado', /\bel resultado\b/gi],
+  ['método de extracción', /\bmétodo de extracción\b/gi],
+  ['uniformidad del molino', /\buniformidad del molino\b/gi],
+  ['café recién tostado', /\bcafé recién tostado\b/gi],
+] as const;
+
+type UnknownRecord = Record<string, unknown>;
+
+function normalize(value: string): string {
+  return value
+    .replace(/<[^>]*>/g, ' ')
+    .replace(/&nbsp;/gi, ' ')
+    .replace(/\s+/g, ' ')
+    .trim()
+    .toLocaleLowerCase('es');
+}
+
+function isTechnicalInvariant(text: string): boolean {
+  const ledUnitMatches = text.match(/\b\d+(?:-\d+)?\s*(?:w|lm)\b/g) ?? [];
+  return [
+    'data:image/svg+xml;base64',
+    'background-image: url',
+    '.layout-playground {',
+    'const samplerate =',
+  ].some((pattern) => text.includes(pattern)) ||
+    (text.includes('presets') && text.includes('hz')) ||
+    (text.includes('t_rectal') && text.includes('exp(-k * t)')) ||
+    (text.includes('led') && ledUnitMatches.length >= 3);
+}
+
+function collectString(value: string, output: string[]): void {
+  const text = normalize(value);
+  if (!isTechnicalInvariant(text) && text.length >= 20) output.push(text);
+}
+
+function collectObject(value: UnknownRecord, output: string[]): void {
+  Object.entries(value).forEach(([childKey, childValue]) => {
+    collectText(childValue, output, childKey);
+  });
+}
+
+function collectText(value: unknown, output: string[], key?: string): void {
+  if (key && STRUCTURAL_KEYS.has(key)) return;
+  if (typeof value === 'string') return collectString(value, output);
+  if (Array.isArray(value)) return value.forEach((item) => collectText(item, output));
+  if (value && typeof value === 'object') collectObject(value as UnknownRecord, output);
+}
+
+async function loadText(loader: unknown): Promise<string[]> {
+  if (typeof loader !== 'function') return [];
+  const module = await (loader as () => Promise<unknown>)();
+  const output: string[] = [];
+  if (module && typeof module === 'object') {
+    const record = module as UnknownRecord;
+    for (const key of TRANSLATABLE_KEYS) {
+      collectText(record[key], output, key);
+    }
+  }
+  return output;
+}
+
+function findSpanishMarkers(text: string[]): string[] {
+  const corpus = text.join(' ');
+  return SPANISH_MARKERS.flatMap(([label, pattern]) =>
+    pattern.test(corpus) ? [label] : [],
+  );
+}
+
+function similarity(left: string, right: string): number {
+  const leftTokens = left.split(/\s+/);
+  const rightCounts = new Map<string, number>();
+  right.split(/\s+/).forEach((token) => rightCounts.set(token, (rightCounts.get(token) ?? 0) + 1));
+  const matches = leftTokens.reduce((total, token) => {
+    const count = rightCounts.get(token) ?? 0;
+    if (count > 0) rightCounts.set(token, count - 1);
+    return total + (count > 0 ? 1 : 0);
+  }, 0);
+  return (2 * matches) / (leftTokens.length + right.split(/\s+/).length);
+}
+
+function findCopiedFragments(spanish: string[], translated: string[]): string[] {
+  return spanish
+    .filter((fragment) => fragment.length >= 80)
+    .filter((fragment) => translated.some((candidate) =>
+      candidate.length >= 80 &&
+      Math.min(fragment.length, candidate.length) / Math.max(fragment.length, candidate.length) >= COPY_THRESHOLD &&
+      similarity(fragment, candidate) >= COPY_THRESHOLD,
+    ))
+    .sort((a, b) => b.length - a.length)
+    .slice(0, 3);
+}
+
+describe('Locales must not contain copied Spanish content', () => {
+  for (const entry of ALL_ENTRIES) {
+    it(`${entry.id} has no untranslated Spanish blocks`, async () => {
+      const spanish = await loadText(entry.i18n.es);
+      const failures: string[] = [];
+
+      for (const [locale, loader] of Object.entries(entry.i18n)) {
+        if (locale === 'es') continue;
+
+        const translated = await loadText(loader);
+        const copiedFragments = findCopiedFragments(spanish, translated);
+        const markerHits = findSpanishMarkers(translated);
+
+        if (copiedFragments.length > 0 || markerHits.length >= 2) {
+          const details = [
+            copiedFragments.length > 0
+              ? `copied fragments: ${copiedFragments
+                  .map((fragment) => JSON.stringify(fragment.slice(0, 120)))
+                  .join(', ')}`
+              : '',
+            markerHits.length >= 2 ? `Spanish markers: ${markerHits.join(', ')}` : '',
+          ]
+            .filter(Boolean)
+            .join('; ');
+          failures.push(`${locale}: ${details}`);
+        }
+      }
+
+      expect(failures, failures.join('\n')).toEqual([]);
+    });
+  }
+});

package/src/tool/compostBinVolumeRatioCalculator/bibliography.astro

@@ -1,18 +1,6 @@
 ---
-import type { BibliographyEntry } from '../../types';
+import { Bibliography as SharedBibliography } from '@jjlmoya/utils-shared';
 import { bibliography } from './bibliography';
-
-interface Props {
-  locale?: string;
-}
-
-const { locale = 'en' } = Astro.props as Props;
-const entries = bibliography;
 ---
 
-<section class="compost-references" aria-labelledby="compost-references-title" data-locale={locale}>
-  <h2 id="compost-references-title">References</h2>
-  <ul>
-    {entries.map((entry: BibliographyEntry) => <li><a href={entry.url} target="_blank" rel="noreferrer">{entry.name}</a></li>)}
-  </ul>
-</section>
+<SharedBibliography links={bibliography} />

package/src/tool/compostBinVolumeRatioCalculator/seo.astro

@@ -1,23 +1,15 @@
 ---
-import type { SEOSection } from '../../types';
+import { SEORenderer } from '@jjlmoya/utils-shared';
 import { compostBinVolumeRatioCalculator } from './entry';
+import type { KnownLocale } from '../../types';
 
 interface Props {
-  locale?: string;
+  locale?: KnownLocale;
 }
 
-const { locale = 'en' } = Astro.props as Props;
-const loader = compostBinVolumeRatioCalculator.i18n[locale as keyof typeof compostBinVolumeRatioCalculator.i18n] ?? compostBinVolumeRatioCalculator.i18n.en;
+const { locale = 'es' } = Astro.props;
+const loader = compostBinVolumeRatioCalculator.i18n[locale as KnownLocale] ?? compostBinVolumeRatioCalculator.i18n.es ?? compostBinVolumeRatioCalculator.i18n.en;
 const content = loader ? await loader() : null;
-const sections = (content?.seo ?? []) as SEOSection[];
 ---
 
-<div class="compost-seo" data-locale={locale}>
-  {sections.map((section) => {
-    if (section.type === 'title') return <h2>{section.text}</h2>;
-    if (section.type === 'paragraph') return <p set:html={section.html} />;
-    if (section.type === 'list') return <ul>{section.items.map((item) => <li set:html={item} />)}</ul>;
-    if (section.type === 'tip') return <aside><h3>{section.title}</h3><div set:html={section.html} /></aside>;
-    return null;
-  })}
-</div>
+{content?.seo && content.seo.length > 0 && <SEORenderer content={{ locale, sections: content.seo }} />}