Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

zlib: Fix gzip member header/input buffer boundary issue #5883

Closed
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 7 additions & 9 deletions src/node_zlib.cc
Original file line number Diff line number Diff line change
Expand Up @@ -43,7 +43,6 @@ enum node_zlib_mode {

#define GZIP_HEADER_ID1 0x1f
#define GZIP_HEADER_ID2 0x8b
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

These defines are unused now right?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yep, but #5884 would start using them again, so I’d leave them in unless it gets decided that that should not get landed.

#define GZIP_MIN_HEADER_SIZE 10

void InitZlib(v8::Local<v8::Object> target);

Expand Down Expand Up @@ -257,17 +256,16 @@ class ZCtx : public AsyncWrap {
ctx->err_ = Z_NEED_DICT;
}
}
while (ctx->strm_.avail_in >= GZIP_MIN_HEADER_SIZE &&

while (ctx->strm_.avail_in > 0 &&
ctx->mode_ == GUNZIP &&
ctx->err_ == Z_STREAM_END) {
ctx->err_ == Z_STREAM_END &&
ctx->strm_.next_in[0] != 0x00) {
// Bytes remain in input buffer. Perhaps this is another compressed
// member in the same archive, or just trailing garbage.
// Check the header to find out.
if (ctx->strm_.next_in[0] != GZIP_HEADER_ID1 ||
ctx->strm_.next_in[1] != GZIP_HEADER_ID2) {
// Not a valid gzip member
break;
}
// Trailing zero bytes are okay, though, since they are frequently
// used for padding.

Reset(ctx);
ctx->err_ = inflate(&ctx->strm_, ctx->flush_);
}
Expand Down
30 changes: 28 additions & 2 deletions test/parallel/test-zlib-from-concatenated-gzip.js
Original file line number Diff line number Diff line change
Expand Up @@ -7,9 +7,12 @@ const zlib = require('zlib');
const path = require('path');
const fs = require('fs');

const abcEncoded = zlib.gzipSync('abc');
const defEncoded = zlib.gzipSync('def');

const data = Buffer.concat([
zlib.gzipSync('abc'),
zlib.gzipSync('def')
abcEncoded,
defEncoded
]);

assert.equal(zlib.gunzipSync(data).toString(), 'abcdef');
Expand Down Expand Up @@ -38,3 +41,26 @@ fs.createReadStream(pmmFileGz)
assert.deepStrictEqual(Buffer.concat(pmmResultBuffers), pmmExpected,
'result should match original random garbage');
}));

// test that the next gzip member can wrap around the input buffer boundary
[0, 1, 2, 3, 4, defEncoded.length].forEach((offset) => {
Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@kthelgason I think testing for 0 should take care of the case where data members fall exactly on input boundaries:

The first unzip.write() call writes abcEncoded + 0 bytes from defEncoded, and the second unzip.end() call writes all of defEncoded. Or did you have something else in mind?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

of course, you're absolutely right.

const resultBuffers = [];

const unzip = zlib.createGunzip()
.on('error', (err) => {
assert.ifError(err);
})
.on('data', (data) => resultBuffers.push(data))
.on('finish', common.mustCall(() => {
assert.strictEqual(Buffer.concat(resultBuffers).toString(), 'abcdef',
`result should match original input (offset = ${offset})`);
}));

// first write: write "abc" + the first bytes of "def"
unzip.write(Buffer.concat([
abcEncoded, defEncoded.slice(0, offset)
]));

// write remaining bytes of "def"
unzip.end(defEncoded.slice(offset));
});
5 changes: 2 additions & 3 deletions test/parallel/test-zlib-from-gzip-with-trailing-garbage.js
Original file line number Diff line number Diff line change
Expand Up @@ -42,9 +42,8 @@ data = Buffer.concat([
Buffer([0x1f, 0x8b, 0xff, 0xff])
]);

assert.equal(zlib.gunzipSync(data).toString(), 'abcdef');
assert.throws(() => zlib.gunzipSync(data));

zlib.gunzip(data, common.mustCall((err, result) => {
assert.ifError(err);
assert.equal(result, 'abcdef', 'result should match original string');
assert(err);
}));