This commit adds the OPENSSL_API_COMPAT macro and sets it to version
1.1.1 of OpenSSL when linking with a shared OpenSSL library.

The motivation for this is that when linking against OpenSSL 3.x there
are a lot of deprecation warnings and this allows them to be avoided.
When we later upgrade the code base to 3.x this value can then be
updated.

This commit adds min and max IV lengths to the test
crypto-cipheriv-decipher-iv when OpenSSL 3.x is in use.

The motivation for this is that OpenSSL 3.x has a check in
providers/implementations/ciphers/ciphercommon_gcm:

  if (iv != NULL) {
    if (ivlen < ctx->ivlen_min || ivlen > sizeof(ctx->iv)) {
      ERR_raise(ERR_LIB_PROV, PROV_R_INVALID_IV_LENGTH);
      return 0;
    }

And the ivlen_min is 8 and max is 64:
(lldb) expr ctx->ivlen_min
(size_t) $25 = 8

(lldb) expr sizeof(ctx->iv)
(unsigned long) $28 = 64
(size_t) $25 = 8

This commit adds a check and a skip if using OpenSSL 3.x. The reason for
this is that Blowfish (BF) is only available when using the legacy
provider in OpenSSL 3.x.

This commit fixes a number of test failures reported when using OpenSSL
3.x, for example:

Error: error:0500007E:Diffie-Hellman routines::modulus too small

This commit adds a constant to identify if the version of OpenSSl is
3 or above.

The motivation for this is it allows for checking this value in tests
to make sure that they work with OpenSSL 3.x, and also with earlier
versions.

Make sure that listeners are added properly if there has previously
been one but currently are none for a given event type.

PR-URL: https://github.com/nodejs/node/pull/34056


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>

This reverts commit b831b081.

This presumably unbreaks the ASAN github action build.

Example failure:

    2020-06-25T19:59:15.1448178Z === release test-repl-envvars ===
    2020-06-25T19:59:15.1448872Z Path: parallel/test-repl-envvars
    2020-06-25T19:59:15.1449449Z --- stderr ---
    2020-06-25T19:59:15.1449835Z assert.js:103
    2020-06-25T19:59:15.1450194Z   throw new AssertionError(obj);
    2020-06-25T19:59:15.1450524Z   ^
    2020-06-25T19:59:15.1450817Z
    2020-06-25T19:59:15.1451431Z AssertionError [ERR_ASSERTION]: Expected values to be strictly deep-equal:
    2020-06-25T19:59:15.1452000Z + actual - expected
    2020-06-25T19:59:15.1452298Z
    2020-06-25T19:59:15.1452634Z   {
    2020-06-25T19:59:15.1452978Z     terminal: true,
    2020-06-25T19:59:15.1453321Z +   useColors: false
    2020-06-25T19:59:15.1453861Z -   useColors: true
    2020-06-25T19:59:15.1454225Z   }
    2020-06-25T19:59:15.1454841Z     at /home/runner/work/node/node/test/parallel/test-repl-envvars.js:55:12
    2020-06-25T19:59:15.1455246Z     at internal/repl.js:33:5

PR-URL: https://github.com/nodejs/node/pull/34058


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>

Fixes: https://github.com/nodejs/node/issues/33348

PR-URL: https://github.com/nodejs/node/pull/33395
Refs: https://github.com/nodejs/node/issues/33348


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

This makes sure all REPL instances check for the NODE_REPL_MODE
environment variable in case the `replMode` is not passed through
as option.

At the same time this simplifies the internal REPL code significantly.

Signed-off-by: Ruben Bridgewater <ruben@bridgewater.de>

PR-URL: https://github.com/nodejs/node/pull/33461


Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

I added it in commit 57231d52 ("src: notify V8 profiler when we're
idle") from October 2013 as a stop-gap measure to measure CPU time
rather than wall clock time, otherwise processes that spend a lot
of time sleeping in system calls give a false impression of being
very busy.

That fix is not without drawbacks because the idle flag is set before
libuv makes I/O callbacks and cleared again after. I/O callbacks can
result into calls into JS code and executing JS code is as non-idle
as you can get.

In commit 96ffcb9a ("src: reduce cpu profiler overhead") from January
2015, I made Node.js block off the SIGPROF signal that V8's CPU profiler
uses before Node.js goes to sleep. The goal of that commit is to reduce
the overhead from EINTR system call wakeups but it also has the pleasant
side effect of fixing what the idle notifier tried to fix.

This commit removes the idle notifier and turns the JS process object
methods into no-ops.

Fixes: https://github.com/nodejs/node/issues/19009
Refs: https://github.com/nodejs/node/pull/33138

PR-URL: https://github.com/nodejs/node/pull/34010


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>

macOS app sandbox makes tcsetattr return EPERM. The CHECK_EQ(0, err)
here would fail when a sandboxed Node.js process is exiting. This commit
fixes this issue.

* test: add test for running in macOS app sandbox

Bare-bone command-line executables cannot run directly in the app sandbox.
To test that Node.js is able to run in the sandbox (and to test the fix in
317621b4a12562eb75055a67bb2c5556f53fe017), this commit creates a typical
Cocoa app bundle, puts the node executable in it and calles Apple's codesign
command to enable sandbox.

* test: use process.execPath to get path of testing node

PR-URL: https://github.com/nodejs/node/pull/33944


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

PR-URL: https://github.com/nodejs/node/pull/33680


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

PR-URL: https://github.com/nodejs/node/pull/33956


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

"utf-16LE" was parsed "UNKNOWN", this fixes to "UCS2"
"utf-buffer" was parsed "BUFFER", this fixes to "UNKNOWN"
"utf-16leNOT" was parsed "UCS2", this fixes to "UNKNOWN"

PR-URL: https://github.com/nodejs/node/pull/33957


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

Refs: https://github.com/nodejs/node/pull/31553
Refs: https://github.com/nodejs/node/pull/32953

PR-URL: https://github.com/nodejs/node/pull/34001
Refs: https://github.com/nodejs/node/pull/34002


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Zeyu Yang <himself65@outlook.com>
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>

We currently mark a number of `ArrayBuffer`s as not transferable,
including the `Buffer` pool and ones with finalizers provided
by C++ addons.

There is no good reason to assume that userland code might not
encounter similar problems, for example when doing `ArrayBuffer`
pooling similar to ours. Therefore, provide an API that lets
userland code also mark objects as not transferable.

PR-URL: https://github.com/nodejs/node/pull/33979


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Gus Caplan <me@gus.host>

PR-URL: https://github.com/nodejs/node/pull/33839


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: David Carlier <devnexen@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>

Fixes: https://github.com/nodejs/node/issues/33993

PR-URL: https://github.com/nodejs/node/pull/33994


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Zeyu Yang <himself65@outlook.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Gerhard Stöbich <deb2001-github@yahoo.de>

This makes sure the DEL character (ASCII 127) is detected as a zero
width character even if Node.js is not built with ICU.

Signed-off-by: Ruben Bridgewater <ruben@bridgewater.de>

PR-URL: https://github.com/nodejs/node/pull/33650


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>

This adds support for the "Combining Diacritical Marks for Symbols"
unicode group to calculate a zero length width even if Node.js is
built without ICU.

Signed-off-by: Ruben Bridgewater <ruben@bridgewater.de>

PR-URL: https://github.com/nodejs/node/pull/33650


Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>

This commit adds the list of agreed-upon info strings to the
documentation style guide to aid with future development and
maintenance.

Refs: https://github.com/nodejs/node/pull/33510
Refs: https://github.com/nodejs/node/pull/33507
Refs: https://github.com/nodejs/node/pull/33483
Refs: https://github.com/nodejs/node/pull/33531
Refs: https://github.com/nodejs/node/pull/33542
Refs: https://github.com/nodejs/node/pull/33028
Refs: https://github.com/nodejs/node/pull/33548
Refs: https://github.com/nodejs/node/pull/33486

PR-URL: https://github.com/nodejs/node/pull/34024


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>

PR-URL: https://github.com/nodejs/node/pull/33939


Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

Rename `ERR_MISSING_MESSAGE_PORT_IN_TRANSFER_LIST`
to `ERR_MISSING_TRANSFERABLE_IN_TRANSFER_LIST` in order to be more
accurate.

PR-URL: https://github.com/nodejs/node/pull/33872


Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>

PR-URL: https://github.com/nodejs/node/pull/33867
Refs: https://github.com/nodejs/node/blob/master/src/README.md#checked-conversion


Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>

This addresses a TODO comment, and aligns the behavior between
worker threads and the main thread.

The primary motivation for this change is to more strictly enforce
the invariant that no JS runs after the `'exit'` event is emitted.

PR-URL: https://github.com/nodejs/node/pull/33874


Reviewed-By: Gus Caplan <me@gus.host>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Shelley Vohr <codebytere@gmail.com>

Fixes: https://github.com/nodejs/node/issues/29536

PR-URL: https://github.com/nodejs/node/pull/33860


Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>

The build defaulted to the byte order of the host system but
that can be different from the endianness of the target system.

Refs: https://github.com/nodejs/node/issues/33703#issuecomment-644639158

PR-URL: https://github.com/nodejs/node/pull/33898


Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>