92ef408ac5
A .docx editor is easy to write badly: read the file into HTML, let someone edit it, write a fresh document back, and hand them a file that lost its headers, its page size and half its formatting without ever saying so. Three things keep this one honest. The reader doesn't use mammoth's HTML. mammoth's converter is deliberately semantic, and HTML has nowhere to put a run's colour or a paragraph's line spacing, so it drops them — and those are controls this editor puts in the ribbon. Taking its parsed document model instead means what the ribbon offers is what the file can actually carry. Six properties mammoth's model didn't keep are added by build-time patches, each asserting its anchor so an upgrade that moves the code fails the build rather than shipping a lossy reader. The writer rebuilds the body but carries the rest of the package across: headers, footers, footnotes, endnotes, the document's own style catalogue, its theme and its page setup, with relationship ids and content types re-wired. Word features the editor can't model are still lost, so they are detected when the file opens and named in a banner before anyone edits. Tracked changes get their own gate. mammoth renders insertions as ordinary text and drops deletions, so saving would accept every pending revision without Word ever asking. Such a document opens read-only until the user says that is what they want. Verified over 66 real documents: 65 round-trip with an identical model and a structurally valid package, the one exception being a 7 MB WMF picture, which no browser can display and the writer cannot emit. Also driven end to end through a real Theseus over CDP — sidebar, ribbon, typing, save, reopen.
77 lines
3.7 KiB
JavaScript
77 lines
3.7 KiB
JavaScript
// Bundles the add-on's npm dependencies into
|
|
// bundled-addons/docx-editor/vendor/docx-vendor.js.
|
|
//
|
|
// npm run build (from addon-build/docx-editor/)
|
|
//
|
|
// mammoth is patched on the way through — see patches.mjs for why. The
|
|
// patched copy is materialised under .patched/ so the node round-trip tests
|
|
// exercise exactly the same reader the browser does.
|
|
import * as esbuild from "esbuild";
|
|
import { mkdirSync, writeFileSync, readFileSync, statSync, rmSync, cpSync } from "node:fs";
|
|
import { fileURLToPath } from "node:url";
|
|
import path from "node:path";
|
|
import { applyPatches, patches } from "./patches.mjs";
|
|
|
|
const here = path.dirname(fileURLToPath(import.meta.url));
|
|
const out = path.resolve(here, "../../bundled-addons/docx-editor/vendor");
|
|
const patchedRoot = path.join(here, ".patched");
|
|
|
|
// --- 1. patched mammoth ---------------------------------------------------
|
|
rmSync(patchedRoot, { recursive: true, force: true });
|
|
mkdirSync(patchedRoot, { recursive: true });
|
|
cpSync(path.join(here, "node_modules", "mammoth"), path.join(patchedRoot, "mammoth"), { recursive: true });
|
|
const touched = new Set();
|
|
for (const p of patches) {
|
|
const f = path.join(patchedRoot, "mammoth", p.file);
|
|
if (touched.has(f)) continue;
|
|
touched.add(f);
|
|
}
|
|
for (const f of touched) {
|
|
const rel = "mammoth/" + path.relative(path.join(patchedRoot, "mammoth"), f).split(path.sep).join("/");
|
|
const { code, hits } = applyPatches(rel, readFileSync(f, "utf8"));
|
|
writeFileSync(f, code);
|
|
console.log(`patched ${rel} (${hits} hunks)`);
|
|
}
|
|
|
|
// --- 2. bundle ------------------------------------------------------------
|
|
mkdirSync(out, { recursive: true });
|
|
const res = await esbuild.build({
|
|
entryPoints: [path.join(here, "vendor-entry.js")],
|
|
bundle: true,
|
|
minify: true,
|
|
format: "iife",
|
|
platform: "browser",
|
|
target: ["chrome120"],
|
|
legalComments: "none",
|
|
outfile: path.join(out, "docx-vendor.js"),
|
|
define: { "process.env.NODE_ENV": '"production"' },
|
|
alias: { mammoth: path.join(patchedRoot, "mammoth") },
|
|
logLevel: "info",
|
|
});
|
|
if (res.errors.length) process.exit(1);
|
|
|
|
// --- 3. licences ----------------------------------------------------------
|
|
// Vendored code without its licence text is the kind of thing that bites
|
|
// later; this file ships next to the bundle.
|
|
const pkgs = ["mammoth", "docx", "jszip", "underscore", "orderedmap", "w3c-keyname",
|
|
"rope-sequence", "prosemirror-state", "prosemirror-view", "prosemirror-model",
|
|
"prosemirror-schema-basic", "prosemirror-schema-list", "prosemirror-tables",
|
|
"prosemirror-history", "prosemirror-commands", "prosemirror-keymap",
|
|
"prosemirror-inputrules", "prosemirror-dropcursor", "prosemirror-gapcursor",
|
|
"prosemirror-transform"];
|
|
let notice = "Third-party code bundled into vendor/docx-vendor.js\n" +
|
|
"===================================================\n\n" +
|
|
"mammoth is shipped with small local patches (colour, paragraph spacing,\n" +
|
|
"numbering format); see addon-build/docx-editor/patches.mjs.\n\n";
|
|
for (const p of pkgs) {
|
|
let j; try { j = JSON.parse(readFileSync(path.join(here, "node_modules", p, "package.json"), "utf8")); }
|
|
catch { continue; }
|
|
let text = "";
|
|
for (const f of ["LICENSE", "LICENSE.md", "LICENSE.txt", "LICENCE", "license", "LICENSE-MIT"]) {
|
|
try { text = readFileSync(path.join(here, "node_modules", p, f), "utf8").trim(); break; } catch {}
|
|
}
|
|
notice += `--- ${p} ${j.version} — ${j.license || "see project"} ---\n` +
|
|
(text || `(no licence file in the package; see ${j.homepage || j.repository?.url || "the project homepage"})`) + "\n\n";
|
|
}
|
|
writeFileSync(path.join(out, "LICENSES.txt"), notice);
|
|
console.log(`vendor bundle: ${(statSync(path.join(out, "docx-vendor.js")).size / 1024).toFixed(0)} KB`);
|