mirror of
https://github.com/andy-stack/vaultkeeper-ai.git
synced 2026-07-22 16:30:27 +00:00
fix: update security neutralisation for unpdf 1.6+ vendored duplicates
Adjust EXPECTED_NEW_FUNCTION_COUNT from 6 to 9 to account for unpdf >=1.6 vendoring PDF.js twice (main + worker/legacy), duplicating the 5 feature- detection probes and 2 PostScript JIT sites (Hogan pair unchanged). Extend WASM neutralisation to handle 4 total .wasm literals (qcms_bg.wasm ×2, jbig2.wasm, openjpeg.wasm) instead of just qcms_bg.wasm, moving from single- reference constant to EXPECTED_WASM_REFS map with per-filename counts and unknown-literal detection. Harden officeparser CJS shim regex to capture all minified identifiers (outer var, arrow param, getter params, function param) via backreferences rather than hardcoding param names, fixing breakage across esbuild releases (v7.1.0→v7.2.1 changed closure params r/t/e→n/e/t).
This commit is contained in:
Andrew Beal
parent
31812c80ad
commit
d500214e2e
1 changed files with 89 additions and 35 deletions
|
|
@ -55,7 +55,13 @@ function copyDir(src, dest) {
|
|||
// the Obsidian scanner's "Dynamic Code Execution" warning. The build FAILS if
|
||||
// the count drifts from EXPECTED_NEW_FUNCTION_COUNT, so a dependency change can
|
||||
// never silently reintroduce a live (or un-neutralised) dynamic-eval path.
|
||||
const EXPECTED_NEW_FUNCTION_COUNT = 6;
|
||||
// 9 = 5 PDF.js feature-detection probes (`new Function(""),!0}catch{return!1}`)
|
||||
// + 2 PDF.js PostScript JIT compilers (gated behind `isEvalSupported`, which
|
||||
// DocumentHelper passes as `false`) + 2 Hogan template compilers (reachable
|
||||
// only via Hogan.compile(), unused with diff2html's precompiled templates).
|
||||
// unpdf >=1.6 vendors PDF.js twice (main + worker/legacy), so the PDF.js probe
|
||||
// and JIT sites each appear in duplicate; the Hogan pair is unchanged.
|
||||
const EXPECTED_NEW_FUNCTION_COUNT = 9;
|
||||
const NEW_FUNCTION_STUB = "VKBlockedDynamicFn";
|
||||
function neutraliseDynamicEval(outfile) {
|
||||
if (!existsSync(outfile)) return;
|
||||
|
|
@ -88,48 +94,85 @@ function neutraliseDynamicEval(outfile) {
|
|||
);
|
||||
}
|
||||
|
||||
// Neutralises the lone `.wasm` filename literal in the bundled output.
|
||||
// Neutralises the `.wasm` filename literals in the bundled output.
|
||||
//
|
||||
// The reference comes from unpdf's vendored PDF.js: `${this.#x}qcms_bg.wasm`,
|
||||
// where qcms is PDF.js's WebAssembly colour-management module (ICC profile
|
||||
// transforms, used only when *rendering* colour-accurate images). Our PDF code
|
||||
// (Helpers/DocumentHelper.ts) only calls extractText for text, never renders,
|
||||
// so the qcms loader is never entered. The loader is also gated behind two
|
||||
// private-field checks and would resolve the file via a relative URL we do not
|
||||
// ship, so even if reached it falls back to PDF.js's JS path.
|
||||
// Every reference comes from unpdf's vendored PDF.js, which loads optional
|
||||
// WebAssembly modules only when *rendering/decoding image streams*:
|
||||
// - qcms_bg.wasm — colour management (ICC profile transforms)
|
||||
// - jbig2.wasm — JBIG2 decoder (scanned bilevel images)
|
||||
// - openjpeg.wasm — OpenJPEG decoder (JPEG2000 images)
|
||||
// Our PDF code (Helpers/DocumentHelper.ts) only calls extractText for text and
|
||||
// never renders, so none of these loaders is entered. Each is also gated behind
|
||||
// private-field checks and resolves its file via a relative URL we do not ship,
|
||||
// so even if reached it falls back to PDF.js's pure-JS path. (unpdf >=1.6
|
||||
// vendors PDF.js twice, so qcms_bg.wasm appears in duplicate.)
|
||||
//
|
||||
// The literal is therefore dead, but its `.wasm` filename trips Obsidian's
|
||||
// The literals are therefore dead, but their `.wasm` filenames trip Obsidian's
|
||||
// "unrecognised .wasm" scanner (native binary that can't be statically
|
||||
// reviewed). We rewrite the `qcms_bg.wasm` filename to a non-.wasm name so no
|
||||
// reviewed). We rewrite each `*.wasm` filename to a non-`.wasm` name so no
|
||||
// `.wasm` literal remains; the surrounding (unreachable) code stays intact. The
|
||||
// build FAILS if the count drifts from EXPECTED_WASM_REF_COUNT, so a dependency
|
||||
// change can never silently reintroduce an undocumented .wasm reference.
|
||||
const EXPECTED_WASM_REF_COUNT = 1;
|
||||
const WASM_REF_FROM = "qcms_bg.wasm";
|
||||
const WASM_REF_TO = "qcms_bg.disabled-wasm";
|
||||
// build FAILS if an unexpected `.wasm` literal appears (or an expected one
|
||||
// vanishes), so a dependency change can never silently reintroduce an
|
||||
// undocumented .wasm reference.
|
||||
//
|
||||
// Keyed by literal filename → number of occurrences expected in the bundle.
|
||||
const EXPECTED_WASM_REFS = {
|
||||
"qcms_bg.wasm": 2,
|
||||
"jbig2.wasm": 1,
|
||||
"openjpeg.wasm": 1,
|
||||
};
|
||||
const WASM_REF_SUFFIX_FROM = ".wasm";
|
||||
const WASM_REF_SUFFIX_TO = ".disabled-wasm";
|
||||
function neutraliseWasmRef(outfile) {
|
||||
if (!existsSync(outfile)) return;
|
||||
let contents = readFileSync(outfile, "utf-8");
|
||||
// Match the whole quoted/templated literal ending in `.wasm`, anchored on the
|
||||
// quote/backtick. The negated class `[^'"`]` cannot overlap the delimiters, so
|
||||
// this matches linearly — a `[\w.-]*\.wasm` form instead backtracks
|
||||
// catastrophically on the megabytes of base64 in the dev inline sourcemap and
|
||||
// hangs the watcher at 100% CPU.
|
||||
const wasmLiterals = contents.match(/['"`][^'"`]*\.wasm['"`]/g) || [];
|
||||
if (wasmLiterals.length !== EXPECTED_WASM_REF_COUNT) {
|
||||
// Extract just the `<name>.wasm` basename from each literal: drop the quotes,
|
||||
// and for template literals drop any `${...}` interpolation prefix (qcms refs
|
||||
// look like `` `${...}qcms_bg.wasm` ``). The basename is the trailing run of
|
||||
// filename characters ending in `.wasm`.
|
||||
const found = {};
|
||||
for (const lit of wasmLiterals) {
|
||||
const inner = lit.slice(1, -1);
|
||||
const name = (inner.match(/[\w.-]+\.wasm$/) || [inner])[0];
|
||||
found[name] = (found[name] || 0) + 1;
|
||||
}
|
||||
// Any literal we don't recognise → fail loudly (new undocumented .wasm).
|
||||
const unknown = Object.keys(found).filter((n) => !(n in EXPECTED_WASM_REFS));
|
||||
if (unknown.length) {
|
||||
throw new Error(
|
||||
`neutraliseWasmRef: expected ${EXPECTED_WASM_REF_COUNT} \`.wasm\` ` +
|
||||
`string literal(s) in ${outfile} but found ${wasmLiterals.length} ` +
|
||||
`(${wasmLiterals.join(", ")}). A dependency changed — re-audit which ` +
|
||||
`library introduced/removed the reference and confirm it is unreachable ` +
|
||||
`before updating EXPECTED_WASM_REF_COUNT.`,
|
||||
`neutraliseWasmRef: unexpected \`.wasm\` literal(s) in ${outfile}: ` +
|
||||
`${unknown.join(", ")}. A dependency introduced a new .wasm reference — ` +
|
||||
`re-audit that it is unreachable, then add it to EXPECTED_WASM_REFS.`,
|
||||
);
|
||||
}
|
||||
if (!contents.includes(WASM_REF_FROM)) {
|
||||
throw new Error(
|
||||
`neutraliseWasmRef: expected to find \`${WASM_REF_FROM}\` in ${outfile} ` +
|
||||
`but it was absent — the .wasm reference changed shape. Re-audit before shipping.`,
|
||||
// Each expected filename must be present with the expected count.
|
||||
for (const [name, count] of Object.entries(EXPECTED_WASM_REFS)) {
|
||||
if ((found[name] || 0) !== count) {
|
||||
throw new Error(
|
||||
`neutraliseWasmRef: expected ${count} \`${name}\` literal(s) in ` +
|
||||
`${outfile} but found ${found[name] || 0}. The .wasm references ` +
|
||||
`changed shape — re-audit before shipping, then update ` +
|
||||
`EXPECTED_WASM_REFS.`,
|
||||
);
|
||||
}
|
||||
}
|
||||
// Rewrite only the `.wasm` extension on the literals we've vetted, so no
|
||||
// `.wasm` string remains for the scanner to flag.
|
||||
for (const name of Object.keys(EXPECTED_WASM_REFS)) {
|
||||
contents = contents.split(name).join(
|
||||
name.slice(0, -WASM_REF_SUFFIX_FROM.length) + WASM_REF_SUFFIX_TO,
|
||||
);
|
||||
}
|
||||
contents = contents.split(WASM_REF_FROM).join(WASM_REF_TO);
|
||||
writeFileSync(outfile, contents);
|
||||
console.log(
|
||||
`🛡️ Neutralised \`.wasm\` reference (${WASM_REF_FROM} → ${WASM_REF_TO}) in ${outfile}`,
|
||||
`🛡️ Neutralised ${wasmLiterals.length} \`.wasm\` reference(s) ` +
|
||||
`(${Object.keys(EXPECTED_WASM_REFS).join(", ")}) in ${outfile}`,
|
||||
);
|
||||
}
|
||||
|
||||
|
|
@ -219,10 +262,12 @@ const cssMergerPlugin = {
|
|||
// the final output. On Obsidian mobile there is no "fs" module, so we replace that shim with a
|
||||
// stub that always throws, keeping the bundle self-contained.
|
||||
//
|
||||
// The shim's variable name is minified and churns between releases (Au in v5, Vm in v6, Gs in
|
||||
// v7.1.0), so the regex captures whatever identifier is used rather than hardcoding it. If the
|
||||
// shim shape changes such that nothing matches, the build FAILS loudly — a silent no-op here
|
||||
// previously let the stale v6 `Vm` regex match nothing on v7.
|
||||
// Every identifier in the shim is minified and churns between releases — not just the outer
|
||||
// variable name (Au in v5, Vm in v6, Gs in v7.1.0, dl in v7.2.1) but also the closure params
|
||||
// (r/t/e in v7.1.0 became n/e/t in v7.2.1). So the regex captures whatever identifiers appear
|
||||
// rather than hardcoding any of them. If the shim's *shape* changes such that nothing matches,
|
||||
// the build FAILS loudly — a silent no-op here previously let the stale v6 `Vm` regex match
|
||||
// nothing on v7, and the hardcoded v7.1.0 param names broke against v7.2.1.
|
||||
const officeParserPlugin = {
|
||||
name: "officeparser-cjs-shim",
|
||||
setup(build) {
|
||||
|
|
@ -232,9 +277,18 @@ const officeParserPlugin = {
|
|||
}));
|
||||
build.onLoad({ filter: /.*/, namespace: 'officeparser-shim' }, async (args) => {
|
||||
let contents = readFileSync(args.path, 'utf-8');
|
||||
// Capture the (minified) shim identifier in group 1 and reuse it in the replacement.
|
||||
const shimPattern =
|
||||
/var ([A-Za-z_$][\w$]*)=\(r=>typeof require<"u"\?require:typeof Proxy<"u"\?new Proxy\(r,\{get:\(t,e\)=>\(typeof require<"u"\?require:t\)\[e\]\}\):r\)\(function\(r\)\{if\(typeof require<"u"\)return require\.apply\(this,arguments\);throw Error\('Dynamic require of "'\+r\+'" is not supported'\)\}\)/;
|
||||
// Capture the (minified) outer var in group 1 and every other minified identifier in its
|
||||
// own group, reusing the captures via backreferences so we match regardless of which letters
|
||||
// esbuild chose this release. Group map: 1=outer var, 2=arrow param, 3/4=getter params,
|
||||
// 5=function param.
|
||||
const id = "[A-Za-z_$][\\w$]*";
|
||||
const shimPattern = new RegExp(
|
||||
`var (${id})=\\((${id})=>typeof require<"u"\\?require:typeof Proxy<"u"\\?` +
|
||||
`new Proxy\\(\\2,\\{get:\\((${id}),(${id})\\)=>\\(typeof require<"u"\\?require:\\3\\)` +
|
||||
`\\[\\4\\]\\}\\):\\2\\)\\(function\\((${id})\\)\\{if\\(typeof require<"u"\\)` +
|
||||
`return require\\.apply\\(this,arguments\\);` +
|
||||
`throw Error\\('Dynamic require of "'\\+\\5\\+'" is not supported'\\)\\}\\)`,
|
||||
);
|
||||
if (!shimPattern.test(contents)) {
|
||||
throw new Error(
|
||||
"officeparser-cjs-shim: dynamic-require shim not found in officeparser browser bundle. " +
|
||||
|
|
|
|||
Loading…
Reference in a new issue