@geml/geml 1.8.1 → 1.8.3

package/codemap/render-all.mjs

@@ -1,90 +1,90 @@
-#!/usr/bin/env node
-// geml codemap render — render every codemap document to a sibling .html.
-//
-//   geml codemap render [codemap-dir]
-//
-// The output folder then works with NO server: open index.html straight from
-// disk (file://). Module click-through opens each container page inside the
-// graph area (nested frame), so the whole map is browsable offline — this is
-// the "copy the folder to someone" mode. For a live view that never goes
-// stale, use `geml codemap serve` instead.
-import { readdirSync, readFileSync, writeFileSync, realpathSync, unlinkSync } from "node:fs";
-import { join, basename, sep, resolve as resolvePath } from "node:path";
-import { parse, renderHtml } from "../dist/geml.js";
-
-if (process.argv[2] === "--help" || process.argv[2] === "-h") {
-  console.error("usage: geml codemap render [codemap-dir]   (dir defaults to ./.geml-code-graph)");
-  process.exit(2);
-}
-const dir = process.argv[2] || ".geml-code-graph";
-
-// One shared cache for the whole batch: every page's graph slice crosses the
-// same neighbour documents, and a fresh parse per page turns N pages into
-// O(N x working set) — hours at repo scale. A one-shot process has no
-// staleness to worry about, so cache unconditionally (the whole codemap's
-// text + parsed docs live in memory for the duration of the run).
-const texts = new Map();  // rel -> text | null
-const parsed = new Map(); // text -> Document
-// `rel` is document-controlled — a `src=` composed by the renderer — and the
-// output of this command is a published artifact, so an unconfined read here
-// writes any .geml on the filesystem into a file that then gets served. Gate on
-// the realpath, the same shape `serve` and the CLI resolver use.
-let realDir = null;
-try { realDir = realpathSync(resolvePath(dir)); } catch { realDir = null; }
-const loadDoc = (rel) => {
-  if (!texts.has(rel)) {
-    let text = null;
-    if (realDir !== null) {
-      try {
-        const real = realpathSync(join(dir, rel));
-        if (real === realDir || real.startsWith(realDir + sep)) text = readFileSync(real, "utf8");
-      } catch { text = null; }
-    }
-    texts.set(rel, text);
-  }
-  return texts.get(rel);
-};
-const parseDoc = (s) => {
-  let d = parsed.get(s);
-  if (!d) { d = parse(s); parsed.set(s, d); }
-  return d;
-};
-
-let n = 0;
-const failed = [];
-let files;
-try {
-  files = readdirSync(dir);
-} catch {
-  console.error(`error: cannot read directory ${dir}`);
-  process.exit(1);
-}
-for (const f of files) {
-  if (!f.endsWith(".geml")) continue;
-  try {
-    const text = loadDoc(f);
-    if (text === null) throw new Error("unreadable");
-    const doc = parseDoc(text);
-    const html = renderHtml(doc, { source: basename(f), loadDoc, parseDoc });
-    writeFileSync(join(dir, f.replace(/\.geml$/, ".html")), html);
-    n++;
-  } catch (e) {
-    failed.push(f);
-    console.error(`render: ${f}: ${e.message}`);
-  }
-}
-// Each tool prunes what it owns: build removes the documents it no longer
-// produces, and this removes the pages whose document is gone. An orphan page
-// is worse than a stale one — it is unreachable from index.html yet still
-// served, so a copied folder ships a page describing deleted code with no way
-// to notice. Only a `<base>.html` whose `<base>.geml` is absent qualifies, so
-// nothing that has a document behind it is ever touched.
-const prunedPages = [];
-for (const f of files) {
-  if (!f.endsWith(".html")) continue;
-  if (files.includes(f.replace(/\.html$/, ".geml"))) continue;
-  try { unlinkSync(join(dir, f)); prunedPages.push(f); } catch { /* already gone */ }
-}
-console.error(`rendered ${n} page(s) -> ${dir}${failed.length ? `; FAILED: ${failed.join(", ")}` : ""}`);
-if (prunedPages.length) console.error(`  pruned ${prunedPages.length} orphan page(s): ${prunedPages.join(", ")}`);
-process.exit(failed.length ? 1 : 0);
+#!/usr/bin/env node
+// geml codemap render — render every codemap document to a sibling .html.
+//
+//   geml codemap render [codemap-dir]
+//
+// The output folder then works with NO server: open index.html straight from
+// disk (file://). Module click-through opens each container page inside the
+// graph area (nested frame), so the whole map is browsable offline — this is
+// the "copy the folder to someone" mode. For a live view that never goes
+// stale, use `geml codemap serve` instead.
+import { readdirSync, readFileSync, writeFileSync, realpathSync, unlinkSync } from "node:fs";
+import { join, basename, sep, resolve as resolvePath } from "node:path";
+import { parse, renderHtml } from "../dist/geml.js";
+
+if (process.argv[2] === "--help" || process.argv[2] === "-h") {
+  console.error("usage: geml codemap render [codemap-dir]   (dir defaults to ./.geml-code-graph)");
+  process.exit(2);
+}
+const dir = process.argv[2] || ".geml-code-graph";
+
+// One shared cache for the whole batch: every page's graph slice crosses the
+// same neighbour documents, and a fresh parse per page turns N pages into
+// O(N x working set) — hours at repo scale. A one-shot process has no
+// staleness to worry about, so cache unconditionally (the whole codemap's
+// text + parsed docs live in memory for the duration of the run).
+const texts = new Map();  // rel -> text | null
+const parsed = new Map(); // text -> Document
+// `rel` is document-controlled — a `src=` composed by the renderer — and the
+// output of this command is a published artifact, so an unconfined read here
+// writes any .geml on the filesystem into a file that then gets served. Gate on
+// the realpath, the same shape `serve` and the CLI resolver use.
+let realDir = null;
+try { realDir = realpathSync(resolvePath(dir)); } catch { realDir = null; }
+const loadDoc = (rel) => {
+  if (!texts.has(rel)) {
+    let text = null;
+    if (realDir !== null) {
+      try {
+        const real = realpathSync(join(dir, rel));
+        if (real === realDir || real.startsWith(realDir + sep)) text = readFileSync(real, "utf8");
+      } catch { text = null; }
+    }
+    texts.set(rel, text);
+  }
+  return texts.get(rel);
+};
+const parseDoc = (s) => {
+  let d = parsed.get(s);
+  if (!d) { d = parse(s); parsed.set(s, d); }
+  return d;
+};
+
+let n = 0;
+const failed = [];
+let files;
+try {
+  files = readdirSync(dir);
+} catch {
+  console.error(`error: cannot read directory ${dir}`);
+  process.exit(1);
+}
+for (const f of files) {
+  if (!f.endsWith(".geml")) continue;
+  try {
+    const text = loadDoc(f);
+    if (text === null) throw new Error("unreadable");
+    const doc = parseDoc(text);
+    const html = renderHtml(doc, { source: basename(f), loadDoc, parseDoc });
+    writeFileSync(join(dir, f.replace(/\.geml$/, ".html")), html);
+    n++;
+  } catch (e) {
+    failed.push(f);
+    console.error(`render: ${f}: ${e.message}`);
+  }
+}
+// Each tool prunes what it owns: build removes the documents it no longer
+// produces, and this removes the pages whose document is gone. An orphan page
+// is worse than a stale one — it is unreachable from index.html yet still
+// served, so a copied folder ships a page describing deleted code with no way
+// to notice. Only a `<base>.html` whose `<base>.geml` is absent qualifies, so
+// nothing that has a document behind it is ever touched.
+const prunedPages = [];
+for (const f of files) {
+  if (!f.endsWith(".html")) continue;
+  if (files.includes(f.replace(/\.html$/, ".geml"))) continue;
+  try { unlinkSync(join(dir, f)); prunedPages.push(f); } catch { /* already gone */ }
+}
+console.error(`rendered ${n} page(s) -> ${dir}${failed.length ? `; FAILED: ${failed.join(", ")}` : ""}`);
+if (prunedPages.length) console.error(`  pruned ${prunedPages.length} orphan page(s): ${prunedPages.join(", ")}`);
+process.exit(failed.length ? 1 : 0);