@glossarist/concept-browser 0.7.122 → 0.7.123

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@glossarist/concept-browser",
-  "version": "0.7.122",
+  "version": "0.7.123",
   "description": "Vue SPA for browsing Glossarist terminology datasets with cross-reference resolution, graph visualization, and multi-language support",
   "type": "module",
   "bin": {

package/scripts/validate-links.js

@@ -2,7 +2,7 @@
 
 // scripts/validate-links.ts
 import { readdirSync, readFileSync, statSync, existsSync } from "node:fs";
-import { join, resolve } from "node:path";
+import { join, dirname, resolve } from "node:path";
 import { cwd } from "node:process";
 var DATA_DIR = process.argv[2] ?? join(cwd(), "public", "data");
 function walk(dir, out = []) {
@@ -61,6 +61,101 @@ function checkFile(file, dataDir) {
       check(m["gl:ref"]?.["@id"] ?? m["gl:ref"], "gl:genericRelations.gl:hasMember[].gl:ref");
     }
   }
+  const datasetDir = file.includes("/concepts/") ? dirname(dirname(file)) : "";
+  const bibPath = datasetDir ? join(datasetDir, "bibliography.json") : "";
+  let bibliography = null;
+  if (bibPath && existsSync(bibPath)) {
+    try {
+      bibliography = JSON.parse(readFileSync(bibPath, "utf8"));
+    } catch {
+    }
+  }
+  const sourceIds = /* @__PURE__ */ new Set();
+  for (const s of json["gl:sources"] ?? []) {
+    if (s["gl:id"]) sourceIds.add(s["gl:id"]);
+  }
+  const textChunks = [];
+  for (const lc of Object.values(json["gl:localizedConcept"] ?? {})) {
+    const lcObj = lc;
+    for (const field of ["gl:definition", "gl:notes", "gl:examples", "gl:annotations"]) {
+      for (const entry of lcObj[field] ?? []) {
+        if (entry["gl:content"]) textChunks.push(entry["gl:content"]);
+      }
+    }
+  }
+  const fullText = textChunks.join("\n");
+  for (const m of fullText.matchAll(/\{\{([^{}]+?)\}\}/g)) {
+    const body = m[1].trim();
+    const relFile = file.replace(cwd() + "/", "");
+    const bibMatch = body.match(/^bib:(.+)$/i);
+    if (bibMatch) {
+      const rest = bibMatch[1];
+      const commaIdx = rest.indexOf(",");
+      const id = commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim();
+      if (sourceIds.has(id)) {
+        broken.push({
+          file: relFile,
+          line: 0,
+          field: "{{bib:...}}",
+          target: id,
+          reason: `"${id}" is a ConceptSource (concept), not a bibliography entry. Use {{cite:${id}}} instead. bib: is ONLY for pure bibliographic records (papers, external docs) that are NOT concepts in any dataset.`
+        });
+      } else if (bibliography) {
+        const entries = bibliography.bibliography ?? bibliography;
+        if (!entries[id] && !(Array.isArray(entries) && entries.some((e) => e.id === id))) {
+          broken.push({
+            file: relFile,
+            line: 0,
+            field: "{{bib:...}}",
+            target: id,
+            reason: `bibliography entry "${id}" not found in bibliography.json. If this is a concept, use {{cite:...}} or {{urn:...}} instead.`
+          });
+        }
+      } else {
+        broken.push({
+          file: relFile,
+          line: 0,
+          field: "{{bib:...}}",
+          target: id,
+          reason: `no bibliography.json for this dataset. If "${id}" is a concept, use {{cite:...}} or {{urn:...}} instead.`
+        });
+      }
+      continue;
+    }
+    const linkMatch = body.match(/^link:(.+)$/i);
+    if (linkMatch) {
+      const rest = linkMatch[1];
+      const commaIdx = rest.indexOf(",");
+      const url = commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim();
+      if (!/^https?:\/\//.test(url)) {
+        broken.push({ file: relFile, line: 0, field: "{{link:...}}", target: url, reason: "link must be http: or https: URL" });
+      }
+      continue;
+    }
+    const imageMatch = body.match(/^image:(.+)$/i);
+    if (imageMatch) {
+      const rest = imageMatch[1];
+      const commaIdx = rest.indexOf(",");
+      const src = commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim();
+      if (!/^https?:\/\//.test(src) && !src.startsWith("data:")) {
+        const imgPath = join(cwd(), "public", src.replace(/^\//, ""));
+        if (!existsSync(imgPath)) {
+          broken.push({ file: relFile, line: 0, field: "{{image:...}}", target: src, reason: `image file not found: public/${src.replace(/^\//, "")}` });
+        }
+      }
+      continue;
+    }
+    const citeMatch = body.match(/^cite:(.+)$/i);
+    if (citeMatch) {
+      const rest = citeMatch[1];
+      const commaIdx = rest.indexOf(",");
+      const id = commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim();
+      if (!sourceIds.has(id)) {
+        broken.push({ file: relFile, line: 0, field: "{{cite:...}}", target: id, reason: `source "${id}" not found in this concept's sources[]` });
+      }
+      continue;
+    }
+  }
   return broken;
 }
 function main() {

package/scripts/validate-links.ts

@@ -70,7 +70,8 @@ function checkFile(file: string, dataDir: string): BrokenLink[] {
     }
   }
 
-  // gl:related entries (concept-level)
+  // ── Concept-level cross-references ─────────────────────────────────────
+  // gl:related entries
   for (const r of json['gl:related'] ?? []) {
     check(r['gl:target'], 'gl:related[].gl:target');
     check(r['@id'], 'gl:related[].@id');
@@ -99,6 +100,111 @@ function checkFile(file: string, dataDir: string): BrokenLink[] {
     }
   }
 
+  // ── Inline mention validation ─────────────────────────────────────────
+  // Scan ALL text content (definitions, notes, examples, annotations) for
+  // {{kind:target}} mentions and validate each by kind. No mention is
+  // silently skipped — every xref must resolve.
+  const datasetDir = file.includes('/concepts/') ? dirname(dirname(file)) : '';
+  const bibPath = datasetDir ? join(datasetDir, 'bibliography.json') : '';
+  let bibliography: Record<string, any> | null = null;
+  if (bibPath && existsSync(bibPath)) {
+    try { bibliography = JSON.parse(readFileSync(bibPath, 'utf8')); } catch {}
+  }
+
+  const sourceIds = new Set<string>();
+  for (const s of json['gl:sources'] ?? []) {
+    if (s['gl:id']) sourceIds.add(s['gl:id']);
+  }
+
+  const textChunks: string[] = [];
+  for (const lc of Object.values(json['gl:localizedConcept'] ?? {})) {
+    const lcObj = lc as any;
+    for (const field of ['gl:definition', 'gl:notes', 'gl:examples', 'gl:annotations']) {
+      for (const entry of lcObj[field] ?? []) {
+        if (entry['gl:content']) textChunks.push(entry['gl:content']);
+      }
+    }
+  }
+  const fullText = textChunks.join('\n');
+
+  for (const m of fullText.matchAll(/\{\{([^{}]+?)\}\}/g)) {
+    const body = m[1].trim();
+    const relFile = file.replace(cwd() + '/', '');
+
+    // bib:id — must be a PURE bibliographic record, NOT a concept.
+    // If the id matches a source in sources[], the author used bib: for
+    // a concept citation — they should use {{cite:...}} instead.
+    // Per concept-model spec: IEV entries, ISO standards with concept IDs,
+    // anything that IS a concept in any dataset → must use cite:, not bib:.
+    const bibMatch = body.match(/^bib:(.+)$/i);
+    if (bibMatch) {
+      const rest = bibMatch[1];
+      const commaIdx = rest.indexOf(',');
+      const id = (commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim());
+      if (sourceIds.has(id)) {
+        broken.push({
+          file: relFile, line: 0, field: '{{bib:...}}', target: id,
+          reason: `"${id}" is a ConceptSource (concept), not a bibliography entry. Use {{cite:${id}}} instead. bib: is ONLY for pure bibliographic records (papers, external docs) that are NOT concepts in any dataset.`,
+        });
+      } else if (bibliography) {
+        const entries = bibliography.bibliography ?? bibliography;
+        if (!entries[id] && !(Array.isArray(entries) && entries.some((e: any) => e.id === id))) {
+          broken.push({
+            file: relFile, line: 0, field: '{{bib:...}}', target: id,
+            reason: `bibliography entry "${id}" not found in bibliography.json. If this is a concept, use {{cite:...}} or {{urn:...}} instead.`,
+          });
+        }
+      } else {
+        broken.push({
+          file: relFile, line: 0, field: '{{bib:...}}', target: id,
+          reason: `no bibliography.json for this dataset. If "${id}" is a concept, use {{cite:...}} or {{urn:...}} instead.`,
+        });
+      }
+      continue;
+    }
+
+    // link:url — must be a valid http/https URL
+    const linkMatch = body.match(/^link:(.+)$/i);
+    if (linkMatch) {
+      const rest = linkMatch[1];
+      const commaIdx = rest.indexOf(',');
+      const url = (commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim());
+      if (!/^https?:\/\//.test(url)) {
+        broken.push({ file: relFile, line: 0, field: '{{link:...}}', target: url, reason: 'link must be http: or https: URL' });
+      }
+      continue;
+    }
+
+    // image:src — if local path, must exist in public/
+    const imageMatch = body.match(/^image:(.+)$/i);
+    if (imageMatch) {
+      const rest = imageMatch[1];
+      const commaIdx = rest.indexOf(',');
+      const src = (commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim());
+      if (!/^https?:\/\//.test(src) && !src.startsWith('data:')) {
+        const imgPath = join(cwd(), 'public', src.replace(/^\//, ''));
+        if (!existsSync(imgPath)) {
+          broken.push({ file: relFile, line: 0, field: '{{image:...}}', target: src, reason: `image file not found: public/${src.replace(/^\//, '')}` });
+        }
+      }
+      continue;
+    }
+
+    // cite:id — source must exist in this concept's sources[]
+    const citeMatch = body.match(/^cite:(.+)$/i);
+    if (citeMatch) {
+      const rest = citeMatch[1];
+      const commaIdx = rest.indexOf(',');
+      const id = (commaIdx > 0 ? rest.slice(0, commaIdx).trim() : rest.trim());
+      if (!sourceIds.has(id)) {
+        broken.push({ file: relFile, line: 0, field: '{{cite:...}}', target: id, reason: `source "${id}" not found in this concept's sources[]` });
+      }
+      continue;
+    }
+
+    // numeric/designation/urn — checked via the concept-level cross-ref checks above
+  }
+
   return broken;
 }
 

package/src/__tests__/scripts/bib-misuse.test.ts

@@ -0,0 +1,76 @@
+import { describe, it, expect } from 'vitest';
+import { mkdirSync, writeFileSync, rmSync } from 'node:fs';
+import { join } from 'node:path';
+import { tmpdir } from 'node:os';
+import { checkFile } from '../../../scripts/validate-links';
+
+function fixture(conceptJson: object): string {
+  const dir = join(tmpdir(), `cb-vl-${Date.now()}-${Math.random().toString(36).slice(2)}`);
+  mkdirSync(join(dir, 'cie-2020', 'concepts'), { recursive: true });
+  const file = join(dir, 'cie-2020', 'concepts', '17-1.json');
+  writeFileSync(file, JSON.stringify(conceptJson));
+  return file;
+}
+
+describe('validate-links — bib: misuse detection (concept cited as bibliography)', () => {
+  it('flags {{bib:id}} when id matches a ConceptSource (should be cite:)', () => {
+    const file = fixture({
+      'gl:sources': [{ 'gl:id': '702-02-07' }],
+      'gl:localizedConcept': {
+        eng: {
+          'gl:definition': [{
+            'gl:content': 'See {{bib:702-02-07, IEV 702-02-07}}',
+          }],
+        },
+      },
+    });
+    const broken = checkFile(file, join(file, '..', '..', '..'));
+    const bibError = broken.find(b => b.field === '{{bib:...}}');
+    expect(bibError).toBeDefined();
+    expect(bibError!.target).toBe('702-02-07');
+    expect(bibError!.reason).toContain('Use {{cite:');
+    rmSync(join(file, '..', '..', '..'), { recursive: true });
+  });
+
+  it('passes {{bib:id}} when id is in bibliography.json (pure bib record)', () => {
+    const dir = join(tmpdir(), `cb-vl-${Date.now()}-${Math.random().toString(36).slice(2)}`);
+    mkdirSync(join(dir, 'cie-2020', 'concepts'), { recursive: true });
+    writeFileSync(join(dir, 'cie-2020', 'bibliography.json'), JSON.stringify({
+      bibliography: { 'iso704_doc': { reference: 'ISO 704:2022' } },
+    }));
+    const file = join(dir, 'cie-2020', 'concepts', '17-1.json');
+    writeFileSync(file, JSON.stringify({
+      'gl:localizedConcept': {
+        eng: { 'gl:definition': [{ 'gl:content': 'See {{bib:iso704_doc, ISO 704}}' }] },
+      },
+    }));
+    const broken = checkFile(file, dir);
+    expect(broken.find(b => b.field === '{{bib:...}}')).toBeUndefined();
+    rmSync(dir, { recursive: true });
+  });
+
+  it('flags {{bib:id}} when id is not in bibliography and not a source', () => {
+    const file = fixture({
+      'gl:localizedConcept': {
+        eng: { 'gl:definition': [{ 'gl:content': 'See {{bib:nonexistent_ref}}' }] },
+      },
+    });
+    const broken = checkFile(file, join(file, '..', '..', '..'));
+    const bibError = broken.find(b => b.field === '{{bib:...}}');
+    expect(bibError).toBeDefined();
+    expect(bibError!.reason).toContain('bibliography');
+    rmSync(join(file, '..', '..', '..'), { recursive: true });
+  });
+
+  it('passes {{cite:id}} when id matches a source', () => {
+    const file = fixture({
+      'gl:sources': [{ 'gl:id': 'iev-702-02-07' }],
+      'gl:localizedConcept': {
+        eng: { 'gl:definition': [{ 'gl:content': 'See {{cite:iev-702-02-07, IEV}}' }] },
+      },
+    });
+    const broken = checkFile(file, join(file, '..', '..', '..'));
+    expect(broken.find(b => b.field === '{{cite:...}}')).toBeUndefined();
+    rmSync(join(file, '..', '..', '..'), { recursive: true });
+  });
+});