test-repl-tab-complete includes tests that ensure that certain keys do
not appear in tab completions or that completion does not crash the
repl. It performed these tests by comparing completion output to a
hardcoded list of expected keys. This list is made incorrect by any api
changes that occur when new versions of V8 are introduced.
With this change, we assert that the specific keys to be avoided are
not present in the output instead.
PR-URL: https://github.com/nodejs/node/pull/5772
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
This commit addresses https://github.com/nodejs/node/issues/5431 by
changing the way that the repl handles assignment to the global _
variable.
Prior to this commit, node sets the result of the last expression
evaluated in the repl to `_`. This causes problems for users of
underscore, lodash and other packages where it is common to assign
`_` to the package, e.g. `_ = require('lodash');`.
Changes in this commit now result in the following behavior.
- If unassigned on the repl, `_` continues to refer to the last
evaluated expression.
- If assigned, the default behavior of assigning `_` to the last
evaluated expression is disabled, and `_` now references whatever
value was explicitly set. A warning is issued on the repl -
'expression assignment to _ now disabled'.
- If `_` is assigned multiple times, the warning is only displayed once.
- When `.clear` is executed in the repl, `_` continues to refer to its
most recent value, whatever that is (this is per existing behavior).
If `_` had been explicitly set prior to `.clear` it will not change
again with the evaluation of the next expression.
PR-URL: https://github.com/nodejs/node/pull/5535
Fixes: https://github.com/nodejs/node/issues/5431
Reviewed-By: Roman Reiss <me@silverwind.io>
Reviewed-By: James M Snell <jasnell@gmail.com>
Previously, we emitted ip and addressType. This change includes the host
as the last argument to the lookup event.
PR-URL: https://github.com/nodejs/node/pull/5598
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
This commit makes input type checking consistent across all path
functions.
PR-URL: https://github.com/nodejs/node/pull/5348
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Roman Reiss <me@silverwind.io>
Several changes:
* Soft-Deprecate Buffer() constructors
* Add `Buffer.from()`, `Buffer.alloc()`, and `Buffer.allocUnsafe()`
* Add `--zero-fill-buffers` command line option
* Add byteOffset and length to `new Buffer(arrayBuffer)` constructor
* buffer.fill('') previously had no effect, now zero-fills
* Update the docs
PR-URL: https://github.com/nodejs/node/pull/4682
Reviewed-By: Сковорода Никита Андреевич <chalkerx@gmail.com>
Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
`Console` constructor checks that `stdout.write()` is a function but
does not do an equivalent check for `stderr.write()`. If `stderr` is not
specified in the constructor, then `stderr` is set to be `stdout`.
However, if `stderr` is specified, but `stderr.write()` is not a
function, then an exception is not thrown until `console.error()` is
called.
This change adds the same check for 'stderr' in the constructor that is
there for `stdout`. If `stderr` fails the check, then a `TypeError` is
thrown.
Took the opportunity to copyedit the `console` doc a little too.
PR-URL: https://github.com/nodejs/node/pull/5635
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rod Vagg <rod@vagg.org>
Per https://github.com/nodejs/node/issues/1817, there are many modules
that currently abuse the private `_events` property on EventEmitter.
One of the ways it is used is to determine if a particular event is
being listened for. This adds a simple `eventNames()` method that
returns an array of the events with currently registered listeners.
PR-URL: https://github.com/nodejs/node/pull/5617
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: Сковорода Никита Андреевич <chalkerx@gmail.com>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
According to the spec gzipped archives can contain more than one
compressed member. Previously Node's gzip implementation would only
unzip the first member and throw away the rest of the compressed data.
Issue #4306 is an example of this occurring in daily use.
Fixes: https://github.com/nodejs/node/issues/4306
PR-URL: https://github.com/nodejs/node/pull/5120
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
`isIPv4` and `isIPv6` are implemented on top of `isIP`, which in turn
checks the sting for being both IPv4 and IPv6, which can be inefficient
in some scenarios. This commit makes them use `uv_inet_pton` directly
instead.
PR-URL: https://github.com/nodejs/node/pull/5478
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Roman Reiss <me@silverwind.io>
net.createServer('aPipe') and net.createServer(8080) are mistakes,
and now throw a TypeError instead of silently being treated as an
object.
PR-URL: https://github.com/nodejs/node/pull/2904
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Make npm install a dependency that is defined as a relative path, so it
avoids any network interaction.
PR-URL: https://github.com/nodejs/node/pull/5613
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Claudio Rodriguez <cjrodr@yahoo.com>
This commit updates events doc to describe removeListener behaviour
when it is called within a listener. An example is added to make
it more evident.
A test is also incuded to make this behaviour consistent in future
releases.
Fixes: nodejs/node#4759
PR-URL: https://github.com/nodejs/node/pull/5201
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
In AsyncWrap::MakeCallback always return empty handle if there is an
error. In the future this should change to return a v8::MaybeLocal, but
that major change will have to wait for v6.x, and these changes are
meant to be backported to v4.x.
The HTTParser call to AsyncWrap::MakeCallback failed because it expected
a thrown call to return an empty handle.
In node::MakeCallback return an empty handle if the call is
in_makecallback(), otherwise return v8::Undefined() as usual to preserve
backwards compatibility.
Fixes: https://github.com/nodejs/node/issues/5555
PR-URL: https://github.com/nodejs/node/pull/5591
Reviewed-By: Julien Gilli <jgilli@nodejs.org>
Currently we are not testing that `npm install` works.
This is a very naive / basic test that shells out to `npm install`
in an empty `tempDir`. While this test will not be able to check
that `npm install` is 100% working, it should catch certain edge
cases that break it.
PR-URL: https://github.com/nodejs/node/pull/5166
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Alexis Campailla <orangemocha@nodejs.org>
Proxies support is now complete in V8. The tests needed slight modification to
match the spec implementation.
PR-URL: https://github.com/nodejs/node/pull/4722
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Michaël Zasso <mic.besace@gmail.com>
- An error message changed for undefined references
- `let` is now allowed in sloppy mode
- ES2015 proxies are shipped and the `Proxy` global is now a function
PR-URL: https://github.com/nodejs/node/pull/4722
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Ali Ijaz Sheikh <ofrobots@google.com>
Looping rapidly and making new connections causes problems on pi2.
Instead create a new connection when an old connection has already been
made. Running a stress test of 600 times and they all passed.
Fixes: https://github.com/nodejs/node/issues/5302
PR-URL: https://github.com/nodejs/node/pull/5537
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Alexis Campailla <orangemocha@nodejs.org>
Enable `no-extra-parens`. This rule restricts the use of parentheses to
only where they are necessary. It is set to be restricted to report only
function expressions.
PR-URL: https://github.com/nodejs/node/pull/5512
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Do not treat crypto inputs as `binary` strings, convert them to Buffers
using `new Buffer(..., 'utf8')`, or using newly updated StringBytes
APIs.
PR-URL: https://github.com/nodejs/node/pull/5522
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Array#pop() is known to be faster than Array#shift().
To be exact, it's O(1) vs. O(n). In this case there's no difference
from which side of the "pool" array the object is retrieved,
so .pop() should be preferred.
PR-URL: https://github.com/nodejs/node/pull/2174
Reviewed-By: mscdex - Brian White <mscdex@mscdex.net>
Reviewed-By: jasnell - James M Snell <jasnell@gmail.com>
Reviewed-By: ofrobots - Ali Ijaz Sheikh <ofrobots@google.com>
Make `HTTPParser` an instance of `AsyncWrap` and make it use
`MakeCallback`. This means that async wrap hooks will be called on
consumed TCP sockets as well as on non-consumed ones.
Additional uses of `AsyncCallbackScope` are necessary to prevent
improper state from progressing that triggers failure in the
test-http-pipeline-flood.js test. Optimally this wouldn't be necessary,
but for the time being it's the most sure way to allow operations to
proceed as they have.
Fix: https://github.com/nodejs/node/issues/4416
PR-URL: https://github.com/nodejs/node/pull/5419
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Change types of all PBKDF2 params to `int` as they are `int` in `evp.h`.
Check that `raw_keylen` fits into `int` before passing it to OpenSSL.
Fix: #5396
PR-URL: https://github.com/nodejs/node/pull/5397
Reviewed-By: Shigeki Ohtsu <ohtsu@iij.ad.jp>
Reviewed-By: Ben Noorhduis <info@bnoordhuis.nl>
The test is still failing sometimes because when trying to establish the
second connection, the server is already closed. Bring back the code
that handled this case and was removed in the last refactoring of the
test. Also ignore the errors that might happen when sending the second
handle to the worker because it may already have exited.
PR-URL: https://github.com/nodejs/node/pull/5422
Reviewed-By: Rich Trott <rtrott@gmail.com>
Fix path.win32.parse("/foo/bar") retuns `{root: '' ...}`(v5.7.0),
but not `{root: '/' ...}`(v5.6.0).
PR-URL: https://github.com/nodejs/node/pull/5484
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: Roman Reiss <me@silverwind.io>
On strict mode, "'use strict'; void 0; " is added as prefix
in order to prevent "use strict" as the result value
for let/const statements. It causes wrong column number in
stack trace.
PR-URL: https://github.com/nodejs/node/pull/5416
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Roman Reiss <me@silverwind.io>
Fixes a regression introduced by: https://github.com/nodejs/node/pull/4374.
Adds a new test to avoid similar issue in the future.
The test is disabled on windows, because this feature never worked
there.
Fixes: https://github.com/nodejs/node/issues/5398
Reviewed-By: Roman Reiss <me@silverwind.io>
Reviewed-By: Rod Vagg <rod@vagg.org>
Reviewed-By: Evan Lucas <evanlucas@me.com>
adds posix test cases for paths similar to those that caused #5447
PR-URL: https://github.com/nodejs/node/pull/5456
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: Roman Reiss <me@silverwind.io>
win32 normalize() will output a trailing '\' for some UNC paths. trim
them before processing
Change by @mscdex
Add basic UNC path tests to win32 relative()
PR-URL: https://github.com/nodejs/node/pull/5456
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: Roman Reiss <me@silverwind.io>
Just send 10 messages recursively and check that the send calls are
asynchronous by asserting that a `setImmediate` callback has been called
in-between. It avoids a race condition in the test when the recursive
limit is reached without having received at least 10 messages.
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/5079
This test was sometimes timing out in `OS X`. Remove the timeout and
clean up the code.
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/5339
Default to FIPS off even in FIPS builds.
Add JS API to check and control FIPS mode.
Add command line arguments to force FIPS on/off.
Respect OPENSSL_CONF variable and read the config.
Add testing for new features.
Fixes: https://github.com/nodejs/node/issues/3819
PR-URL: https://github.com/nodejs/node/pull/5181
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-by: Michael Dawson <michael_dawson@ca.ibm.com>
Currently, there is a check to ensure that the user either
provides an object or a string to repl.start(). The string case
is used to set a REPL prompt. However, a default of '> ' already
exists, so forcing the user to specify a prompt is a bit
redundant. This commit removes this restriction.
Fixes: https://github.com/nodejs/node/issues/5385
PR-URL: https://github.com/nodejs/node/pull/5388
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Julian Duque <julianduquej@gmail.com>
It's documented as such but didn't actually behave that way.
Bug introduced in commit 66fc8ca ("cluster: emit 'message' event on
cluster master"), which is the commit that introduced the event.
Fixes: https://github.com/nodejs/node/issues/5126
PR-URL: https://github.com/nodejs/node/pull/5361
Reviewed-By: Ali Ijaz Sheikh <ofrobots@google.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Enable additional likely-uncontroversial lint rules:
* `comma-dangle` set to prohibit dangling commas on objects and arrays
that are defined on a single line. Multi-line definitions can use or
omit a trailing comma.
* `no-unused-labels` Prohibits defining a label that is not used.
* `no-path-concat` Prohibits string-concatenation using i`__dirname` and
`__filename`. Use `path.join()`, `path.resolve()`, or template strings
instead.
* `no-new-symbol` disallow use of `new` operator with `Symbol` object.
Violating this rule would result in a `TypeError` at runtime.`
PR-URL: https://github.com/nodejs/node/pull/5357
Reviewed-By: Rod Vagg <rod@vagg.org>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michaël Zasso <mic.besace@gmail.com>
Matt Loring
Prince J Wesley
Lance Ball
HUANG Wei
Brian White
James M Snell
Rich Trott
Kári Tristan Helgason
Vladimir Kurchatkin
Sam Roberts
Santiago Gimeno
Vaibhav
Trevor Norris
Myles Borins
Evan Lucas
Matteo Collina
Ali Ijaz Sheikh
Michaël Zasso
Wyatt Preul
Fedor Indutny
Anton Khlynovskiy
Ben Noordhuis
Owen Smith
Zheng Chaoping
Stefan Budeanu
cjihrig
Jeremiah Senkpiel