Calling Http2ServerResponse.end multiple times should never
cause the code to throw an error, subsequent calls should
instead return false. Fix behaviour to match http1.
Fixes: https://github.com/nodejs/node/issues/15385
PR-URL: https://github.com/nodejs/node/pull/15414
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
The test parallel/test-dgram-multicast-set-interface.js was
calling common.skip() on hosts that do not support IPv6. However,
by this point, there were several outstanding common.mustCall()
invocations. The process.exit() in common.skip() triggered
those common.mustCall()s as errors.
Fixes: https://github.com/nodejs/node/issues/15419
PR-URL: https://github.com/nodejs/node/pull/15421
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/15398
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Showing the stack trace in a error message obfuscates the actual
message and should not be visible therefore.
PR-URL: https://github.com/nodejs/node/pull/15025
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
The main optimizations are
- Removed visibleKeys
- Removed proxy cache
- Removed Object.assign
- No key concatenating anymore
- No key recalculating anymore
- Improved indentation logic
- Improved string escape logic
- Added many fast paths
- Optimized code branches a lot
- Optimized (boxed) primitive handling
- Inline code if possible
- Only check extra keys if necessary
- Guard against unnecessary more expensive calls
This also fixes a bug with special array number keys as e.g. "00".
Besides that there were lots of smaller optimizations, the
code got a bit cleaned up and a few more tests got in.
PR-URL: https://github.com/nodejs/node/pull/14881
Fixes: https://github.com/nodejs/node/issues/15288
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
`test-benchmark-buffer` has been observed to timeout on CI on SmartOS.
Move the test to `sequential` so it is not competing with other tests
for resources.
PR-URL: https://github.com/nodejs/node/pull/15373
Fixes: https://github.com/nodejs/node/issues/15372
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Remove napi_is_construct_call and introduce napi_get_new_target.
PR-URL: https://github.com/nodejs/node/pull/14698
Reviewed-By: Jason Ginchereau <jasongin@microsoft.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Kyle Farnung <kfarnung@microsoft.com>
- Add napi_async_context opaque pointer type.
(If needed, we could later add APIs for getting the async IDs
out of this context.)
- Add napi_async_init() and napi_async_destroy() APIs.
- Add async_context parameter to napi_make_callback().
- Add code and checks to test_make_callback to validate async context
APIs by checking async hooks are called with correct context.
- Update API documentation.
PR-URL: https://github.com/nodejs/node/pull/15189
Fixes: https://github.com/nodejs/node/issues/13254
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
As per discussion in abi-stable-node:
https://github.com/nodejs/abi-stable-node/issues/256,
take a refactor to napi_addon_register_func such that
the result from the register function is assigned to
the module exports property. By making this change,
native module can be agnostic about which type of
module the environment supports.
PR-URL: https://github.com/nodejs/node/pull/15088
Reviewed-By: Gabriel Schulhof <gabriel.schulhof@intel.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Hitesh Kanwathirtha <digitalinfinity@gmail.com>
Enable combining N-API async work with async-hooks.
PR-URL: https://github.com/nodejs/node/pull/14697
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Jason Ginchereau <jasongin@microsoft.com>
Reviewed-By: Michael Dawson <mhdawson@ibm.com>
- Merge the two almost-but-not-quite identical `MakeCallback()`
implementations
- Provide a public `CallbackScope` class for embedders in order
to enable `MakeCallback()`-like behaviour without tying that
to calling a JS function
PR-URL: https://github.com/nodejs/node/pull/14697
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
`node_internals.h` already includes the most common headers,
so double includes can be avoided in a lot of cases. Also don’t include
`node_internals.h` from `node.h` implicitly anymore, as that is mostly
unnecessary.
PR-URL: https://github.com/nodejs/node/pull/14697
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Add wrapper for uv's uv_udp_set_multicast_interface which provides the
sender side mechanism to explicitly select an interface. The
equivalent receiver side mechanism is the optional 2nd argument of
addMembership().
PR-URL: https://github.com/nodejs/node/pull/7855
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
This commit verifies that the child_process fork() method does
not honor the shell option.
Refs: https://github.com/nodejs/node/pull/15299
PR-URL: https://github.com/nodejs/node/pull/15352
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
The tests in `test/debugger` all fail since the removal of the
pre-inspector debugger (if they weren't already failing). They do not
run in CI (probably because they were never reliable). Remove them and
associated fixtures.
PR-URL: https://github.com/nodejs/node/pull/15139
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
`test/known_issues/test-stdout-buffer-flush-on-exit.js` is invalid. The
behavior seen currently (with the test failing) is in accordance with
documentation which indicates that calling `process.exit()` may mean
scheduled asynchronous I/O does not happen. The documentation also
indicates that `process.stdout` is asynchronous when it is a POSIX pipe.
PR-URL: https://github.com/nodejs/node/pull/15320
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
* Use cleaner `process.stdin.write('.exit')` to exit the process rather
than `proc.kill()`.
* Move test to sequential. It uses the default port 9229. It will fail
if another inspector test (or test using port 0) is already using that
port. So it needs to be run sequentially rather than in parallel with
other tests. (We haven't seen many failures with it yet because there
aren't a lot of other inspector tests in parallel at this time.)
PR-URL: https://github.com/nodejs/node/pull/15141
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Previously, this test would contain a DNS query that timed out
after 60 seconds, thus occupying one of the parallel test slots
for that period.
Fix that by creating a new channel for that request, and cancelling
it immediately.
PR-URL: https://github.com/nodejs/node/pull/15319
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
`tls.parseCertString()` exposed by accident. Now move this function to
`internal/tls` and mark the original one as deprecated.
PR-URL: https://github.com/nodejs/node/pull/14249
Refs: https://github.com/nodejs/node/issues/14193
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Add tests that cover errors for wrong arguments, as well as
tests for error codes from nghttp2. Fix pushStream to emit
NGHTTP2_ERR_STREAM_ID_NOT_AVAILABLE on session rather than
stream.
PR-URL: https://github.com/nodejs/node/pull/15281
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
This change introduces an AliasedBuffer class and updates asytnc-wrap
and http2 to use this class.
A common technique to optimize performance is to create a native buffer
and then map that native buffer to user space via JS array. The runtime
can efficiently write to the native buffer without having to route
though JS, and the values being written are accessible from user space.
While efficient, this technique allows modifications to user
space memory w/out going through JS type system APIs, effectively
bypassing any monitoring the JS VM has in place to track program state
modifications. The result is that monitors have an incorrect view
of prorgram state.
The AliasedBuffer class provides a future placeholder where this
technique can be used, but writes can still be observed. To achieve
this, the node-chakra-core fork will add in appropriate tracking logic
in the AliasedBuffer's SetValue() method. Going forward, this class can
evolve to support more sophisticated mechanisms if necessary.
PR-URL: https://github.com/nodejs/node/pull/15077
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
* setRecvBufferSize(int) and setSendBufferSize(int)
* added docs for send/receive buffer sizes
* Added options support to set buffer sizes in
dgram.createSocket().
PR-URL: https://github.com/nodejs/node/pull/13623
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
The binding testing napi_wrap() creates references to primitives passed
into the binding in its second parameter. This is unnecessary and not
at all the point of the test. Additionally, creating persistent
references to primitive values may not be supported by all VMs, since
primitives are best persisted in their native form.
Instead, the point of the test is to make sure that the finalize
callback gets called when it should get called, that it gets called
with the correct pointer, and that it does not get called when it
should not get called. Creating persistent references is not necessary
for verifying this.
PR-URL: https://github.com/nodejs/node/pull/15289
Reviewed-By: Jason Ginchereau <jasongin@microsoft.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Re: https://github.com/nodejs/node-chakracore/issues/380
Unbox all primitives and compare them as well instead of
only comparing boxed strings.
PR-URL: https://github.com/nodejs/node/pull/15050
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
This commit adds special handling of Error instances when passed
as the message argument to assert functions. With this commit,
if an Error is passed as the message, then that Error is thrown
instead of an AssertionError.
PR-URL: https://github.com/nodejs/node/pull/15304
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Mostly shared/duplicated logic between all benchmark test files, so
creating a new common module to store it.
PR-URL: https://github.com/nodejs/node/pull/15004
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
PR #11705 switched Node away from using using OpenSSL's legacy EVP_Sign*
and EVP_Verify* APIs. Instead, it computes a hash normally via
EVP_Digest* and then uses EVP_PKEY_sign and EVP_PKEY_verify to verify
the hash directly. This change corrects two problems:
1. The documentation still recommends the signature algorithm EVP_MD
names of OpenSSL's legacy APIs. OpenSSL has since moved away from
thosee, which is why ECDSA was strangely inconsistent. (This is why
"ecdsa-with-SHA256" was missing.)
2. Node_SignFinal copied some code from EVP_SignFinal's internals. This
is problematic for OpenSSL 1.1.0 and is missing a critical check
that prevents pkey->pkey.ptr from being cast to the wrong type.
To resolve this, remove the non-EVP_PKEY_sign codepath. This codepath is
no longer necessary. PR #11705's verify half was already assuming all
EVP_PKEYs supported EVP_PKEY_sign and EVP_PKEY_verify. Also, in the
documentation, point users towards using hash function names which are
more consisent. This avoids an ECDSA special-case and some strangeness
around RSA-PSS ("RSA-SHA256" is the OpenSSL name of the
sha256WithRSAEncryption OID which is not used for RSA-PSS).
PR-URL: https://github.com/nodejs/node/pull/15024
Reviewed-By: Shigeki Ohtsu <ohtsu@ohtsu.org>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
test-tls-fast-writing can fail on a heavily-loaded system due to an
arbitrary 1-second timeout. Remove the arbitrary timeout.
PR-URL: https://github.com/nodejs/node/pull/15138
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
On a slow/busy machine, `test-http-agent-maxsockets` can fail if the
test takes longer than 5 seconds because that is the default value for
`server.keepAliveTimeout`. Disable `keepAliveTimeout` to make the test
robust.
PR-URL: https://github.com/nodejs/node/pull/15192
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reasons:
- `test-async-wrap-getasyncid` binds a handle, so move to
sequential because port cannot be already in use.
- `test-dgram-implicit-bind-failure` requires a hardcoded
port number to properly send socket packet.
- `test-http-agent-uninitialized-with-handle` requires a
hardcoded port number to properly send http request.
- `test-http-agent-uninitialized` requires a hardcoded port
number to properly send http request.
- `test-net-localport` requires a hardcoded port number
for assertions.
In addition this replaces two common.PORTs with a dynamic port.
PR-URL: https://github.com/nodejs/node/pull/15151
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Fix references within http2 core to HTTP_STATUS_CONTENT_RESET to point
to the correct HTTP_STATUS_RESET_CONTENT. Add tests for status 204,
205 & 304 in respond, respondWithFD & respondWithFile. Add general
error tests for respondWithFD & respondWithFile.
PR-URL: https://github.com/nodejs/node/pull/15153
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Claudio Rodriguez <cjrodr@yahoo.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
The only valid values for enablePush are 0 and 1. If validation
is requested, we should verify that it wasn't set to another
value rather than casting to Boolean regardless of value.
PR-URL: https://github.com/nodejs/node/pull/15167
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Anatoli Papirovski
cjihrig
Ruben Bridgewater
Rich Trott
James M Snell
Sampson Gao
Jason Ginchereau
Taylor Woll
Jon Moss
Anna Henningsen
Will Young
Refael Ackermann
Weijia Wang
XadillaX
Mike Kaufman
Damien O'Reilly
Bartosz Sosnowski
Gabriel Schulhof
geek
Rahul Mishra
David Benjamin