Original commit message:
darwin: fix size calculation in select() fallback
Apple's `fd_set` stores its bits in an array of 32-bit integers, which
means `FD_ISSET()` may read out of bounds if we allocate storage at
byte granularity. There's also a chance that the `select()` call could
corrupt the heap, although I didn't investigate that.
This issue was discovered by LLVM's AddressSanitizer which caught
`FD_ISSET()` trying to read out of bounds.
Ref: https://github.com/libuv/libuv/pull/241
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
PR-URL: https://github.com/joyent/node/pull/9179
Float patch to fix setsockopt for multicast on Solaris and derivatives.
Original commit message:
solaris: fix setsockopt for multicast options
On Solaris and derivatives such as SmartOS, the length of socket options
for multicast and ttl options is not always sizeof(char).
This fixes the udp_options and udp_options6 tests.
Ref: https://github.com/libuv/libuv/pull/243
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
PR-URL: https://github.com/joyent/node/pull/9179
Float patch to fix pipe on Windows. Original commit message:
win: fix pipe blocking writes
In the code path for pipe blocking writes, WriteFile is already
posting a completion packet to the I/O completion port.
POST_COMPLETION_FOR_REQ was causing the same request to get
returned twice by GetCompletionStatusEx.
Also on the same code path, we were waiting on the wrong event.
We need to update queued_bytes and write_queue_size when a
blocking write request completes asynchronously.
Ref: https://github.com/libuv/libuv/pull/238
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
PR-URL: https://github.com/joyent/node/pull/9179
process.send() should be synchronous, it should block until the message
has been sent in full, but it wasn't after the second-to-last libuv
upgrade because of commit libuv/libuv@393c1c5 ("unix: set non-block
mode in uv_{pipe,tcp,udp}_open"), which made its way into io.js in
commit 07bd05b ("deps: update libuv to 1.2.1").
Commit libuv/libuv@b36d4ff ("unix: implement uv_stream_set_blocking()")
as landed in io.js in commit 9681fca ("deps: update libuv to 1.4.0")
makes it possible to restore the synchronous behavior again and that's
precisely what this commit does.
The same line of reasoning applies to `net.Socket({ fd: 1 })`: creating
a socket object from a stdio file descriptor, like the `process.stdout`
getter does, should put the file descriptor in blocking mode for
compatibility reasons.
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
PR-URL: https://github.com/joyent/node/pull/9179
When slicing global pool - ensure that the underlying buffer's data ptr
is 8-byte alignment to do not ruin expectations of 3rd party C++ addons.
NOTE: 0.10 node.js always returned aligned pointers and v0.12 should do
this too for compatibility.
PR-URL: https://github.com/joyent/node/pull/9375
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
caeb67735b introduced a regression where
the domains stack would not be cleared after an error had been handled
by the top-level domain.
This change clears the domains stack regardless of the position of the
active domain in the stack.
PR: #9364
PR-URL: https://github.com/joyent/node/pull/9364
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
This reverts commit ad0684807c.
Initially, this bug fix targeted master, and I pushed to have it
included in v0.10. In retrospect, I'm not sure it should have made into
v0.10 as it seems it could break a lot of existing working code.
In my opinion, this change is still a bug fix, and it is not backward
incompatible per se. However, I'm not sure that taking the risk to break
a lot of users with a new 0.10.x release that would include this fix is
reasonable, especially now that 0.10.x releases are entering
maintenance mode.
PR-URL: https://github.com/joyent/node/pull/9257
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
A significant performance regressions has been introduced in 1fddc1f for
GET requests which send data through response.end(). The number of
requests per second dropped to somewhere around 6% of their previous
level.
The fix consists of removing a part of the lines added by 1fddc1f,
lines which were supposed to affect only HEAD requests, but interfered
with GET requests instead.
The lines removed would not have affected the behaviour in the case of
a HEAD request as this._hasBody would always be false. Therefore, they
were not required to fix the issue reported in #8361.
Fixes#8940.
PR: #9026
PR-URL: https://github.com/joyent/node/pull/9026
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
This test setups two event listeners: one on a child process' exit event
, another for the same child process' stdandard output's 'data' event.
The data even listener writes to a stream, and the exit event listener
ends it.
Because the exit event can be emitted before the data event, there is a
chance that something will be written to the stream after it's ended,
and that an error is thrown.
This change makes the test end the stream in the listener for the child
process' standard output's end event, which is guaranteed to be emitted
after the last data event, thus avoiding the race.
PR: #9301
PR-URL: https://github.com/joyent/node/pull/9301
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: James M Snell <jasnell@users.noreply.github.com>
Reviewed-By: Bert Belder <bertbelder@gmail.com>
Currently, lib/_tls_legacy.js and lib/crypto.js cannot be loaded when
--use-strict is passed to node. In addition to that, console.trace
throws because it uses arguments.callee.
This change fixes these issues and adds a test that makes sure
every external built-in module can be loaded with require when
--use-strict is enabled.
Please note that this change does not fix all issues with built-in
modules' code running with --use-strict. It is very likely that some
code in the built-in modules still fails when passing this flag.
However, fixing all code would require us to enable strict mode by
default in all builtins modules, which would certainly break existing
applications.
Fixes#9187.
PR: #9237
PR-URL: https://github.com/joyent/node/pull/9237
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
test/simple/test-http-destroyed-socket-write2.js validates
that you get an appropriate error when trying to write to
a request when the response on the other side has been destroyed.
The test uses http.request to get a request and then keeps writing
to it until either it hits 128 writes or gets the expected error.
Since the writes are asynchronous we see that the writes just end
up adding events to the event loop, which then later get processed
once the connection supporting the request is fully ready.
The test is timing dependent and if takes too long for the connection
to be made the limit of 128 writes is exceeded and the test fails.
The fact that the test allows a number of writes is probably to allow
some delay for the connection to be ready for writing.
On AIX, in the default configuration using the loopback interface
is slower and the test fails because the delay is such that many
more writes can be queued up before the connection takes place.
If we use the host ip instead of defaulting to the loopback then
the test passes.
The test needs to be made more robust to delays. Since each write
simply enqueues an additional write to the event queue there is
probably no point in doing the second write until the first has
completed. This patch schedules the next write when the first one
completes and allows the test to pass even if it takes longer for
the connection to be ready for writing
PR-URL: https://github.com/joyent/node/pull/9270
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
In the documentation for querystring.parse, the documentation mentions
that the default value for options.decodeURIComponent is the
decodeURIComponent function, but it's actually the querystring.unescape
function.
PR-URL: https://github.com/joyent/node/pull/9259
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
`TLSSocket` wraps the original `net.Socket`, but writes/reads to/from
`TLSSocket` do not touch the timers of original `net.Socket`.
Introduce `socket._parent` property, and iterate through all parents
to unref timers and prevent timeout event on original `net.Socket`.
Fix: https://github.com/joyent/node/issues/9242
PR-URL: https://github.com/iojs/io.js/pull/891
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
This is a backport of 6c3647c38d from
v0.12 to v0.10.
Console.prototype.timeEnd() returns NaN if the timer label
corresponds to a property on Object.prototype. In v0.12, this was fixed
by using Object.create(null) to construct the _times object
However, the version of V8 in the v0.10 branch makes this fix not work
as expected. In v0.10, this commit changes the _times object into a
array of objects of the form:
{ label: someLabel, time: staringWallClockTime }
someLabel can thus be any string, including any string that represents
any Object.prototype field.
Fixes#9116.
PR: #9215
PR-URL: https://github.com/joyent/node/pull/9215
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, fs.truncate() silently fails when a file descriptor
is passed as the first argument. This commit changes this
behavior to properly call fs.ftruncate().
PR-URL: https://github.com/joyent/node/pull/9161
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
The commit in v0.10 (431eb172f9) that
backported the original change
(https://codereview.chromium.org/806143002) did add an extra newline
because the logging facilities in v0.10's V8 do not add one.
When merging this commit in v0.12, V8's logging facilities now
automatically add the newline character, and the debug builds assert if
one is already present.
If the Buffer allocation isn't a slice then there's no need to adjust
the pool offset after realloc'ing the space available.
Fixes: 6462519 "buffer, doc: misc. fix and cleanup"
The NativeModule system passes NativeModule.require transparently and so
is unnecessary to call explicitly.
The only one which should have the prefix is the in line 295, where
actually implements a big fs-based module system and actually requires a
native module. That is left unchanged.
PR-URL: https://github.com/joyent/node/pull/9201
Ref: https://github.com/joyent/node/issues/2009
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Timeout#unref() call returns undefined, not this. The test already
worked before, because the interval was still unref'd, and the test also
succeeds without clearing the interval.
PR-URL: https://github.com/joyent/node/pull/9171
Reviewed-by: Colin Ihrig <cjihrig@gmail.com>
Reviewed-by: Timothy J Fontaine <tjfontaine@gmail.com>
A number -> uint32 type coercion bug made buffer sizes
larger than kMaxLength (0x3fffffff) wrap around.
Instead of rejecting the requested size with an exception,
the constructor created a buffer with the wrong size.
PR-URL: https://github.com/iojs/io.js/pull/657
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
AsyncWrap should always properly propagate asynchronous calls to any
child that is created. Regardless whether kCallInitHook is currently
active. The previous logic would always return early if kCallInitHook
wasn't set.
PR-URL: https://github.com/joyent/node/pull/9146
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
V8_CONSTANT_REMOVED_SINCE(major, minor) can be used to mark a constant
has being removed from V8 since V8 version major.minor.
Reviewed-By: Dave Pacheco <dap@joyent.com>
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Bugs fixed:
* v0.12 and later: in-object properties not printing correctly.
* 64-bit: not printing external strings correctly (offset was hardcoded
for 32-bit). This would happen with "::jsstack -vn0" because the
script "node.js" wasn't printed correctly, at least with 0.10 core
files.
* 64-bit: printing JS source (via "::jsstack -v") emits errors and shows
the wrong code.
* Several build warnings.
* Two-byte strings are unnecessarily truncated.
* Could print friendlier note when given obviously bogus function token
positions.
New features:
* ::jsstack prints much cleaner output by default.
* ::jsprint keys are now quoted.
* ::jsstack -v includes "this" value for each function on the stack.
* ::jsstack -v includes more details about each argument (constructor
names for each object).
* new commands: ::jsconstructor, ::jsfunctions, ::jssource, ::nodebuffer
and ::v8internal.
* ::findjsobjects and ::jsprint hidden flags for developers to measure
and improve test coverage.
* internal jsobj_properties() function is much better documented.
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Building node with GCC > 4.4 on CentOS makes the node binary depend on a
more recent version of the C/C++ runtime that is not installed by
default on these older CentOS platforms, and probably on other platforms
as well.
Building node with the default gcc and g++ compilers that come with
these older versions of CentOS allows to ship a node binary that runs
out of the box on these setups with older C/C++ runtimes.
This change works around a bug that was fixed in GCC 4.5. Versions of
GCC < 4.5 would not support using the injected-class-name of a
template base class as a type name.
This change also disables aliasing optimizations for toolchains using
GCC <= 4.4 as they're not able to deal with the aliasing in the queue
implementation used by libuv and node (see src/queue.h).
Fixes#9079.
PR: #9098
PR-URL: https://github.com/joyent/node/pull/9098
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Currently, JSON.stringify() is used to create error messages
on failed assertions. This causes an error when stringifying
objects with circular references. This commit switches out
JSON.stringify() for util.inspect(), which can handle
circular references.
PR: #8734
PR-URL: https://github.com/joyent/node/pull/8734
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
This reverts commit d312b6d15c.
d312b6d15c introduced some confusion in
the existing API of url.format and url.parse.
The way the 'path' property overrides other properties in url.format's
input is too confusing for existing users compared to the issues it
fixes.
Fixes such as https://github.com/joyent/node/pull/9081 have been
proposed, but they do not make the API less confusing.
Instead, this change just reverts the original breaking change so that
it gives us more time after v0.12.0 is released to come up with a better
API for url.format, url.parse and other related APIs in the v0.13
development branch.
Fixes#9070.
Conflicts:
doc/api/url.markdown
PR: #9109
PR-URL: https://github.com/joyent/node/pull/9109
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
On Windows, when building the "build-release" rule, enable both small-icu
and download-all.
This change also slightly refactors the build-release rule, mainly so that
it's easier to read.
The resulting MSI package was tested on Windows 7.
Fixes#9099.
PR: #9100
PR-URL: https://github.com/joyent/node/pull/9100
Reviewed-By: Steven R. Loomis <srloomis@us.ibm.com>
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Make this test less prone to race conditions by using synchronous
interprocess communication instead of a timer to determine when the
child process is ready to receive messages from its parent.
Also, remove a superfluous timer since the tests suite already makes
tests time out after a while.
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
In order to preserve the potential for a flush method being added to the
streams API, rename flush to flushHeaders which is much more clear about
the behavior of this method.
PR: #9048
PR-URL: https://github.com/joyent/node/pull/9048
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
Since the current environment is in scope use ThrowError on that,
instead of having to lookup the Environment again.
Added benefit, lint the source code.
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Julien Gilli <julien.gilli@joyent.com>
On non-windows supported platforms, fs.access(readOnlyFile, W_OK, ...)
is expected to fail, but always succeeds if node runs as the super user,
which is often the case for tests running on our continuous integration
platform.
This change makes the test try to change its process user id to nobody
on non-windows platforms so that the above mentioned test can pass and
still perform the actual desired test. If changing the process user id
to a nobody is not possible, then the test checks that
fs.access(readOnlyFile, W_OK, ...) actually succeeds.
Fixes#9033.
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
unref one superfluous timer (as the test suite already has a global
timeout), and improve the state machine to iterate the messages more
reliably.
Ultimately make the test complete more quickly.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
Previously if a worker's state machine had already transitioned into the
'listening' state when it received the message enabling the debugger,
the worker would never enable its debugger.
Change the logic to allow the 'listening' as a valid state for enabling
the debugger.
Fixes#6440
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
In general path functions don't change the case of a path. Making an
exception for windows drive letters violates the principle of least
surprise.
Changing the drive letter case has caused a lot of issues, including
joyent/node#7031, joyent/node#7806 and lots of bikeshedding about
whether uppercase is the right case or lowercase.
This effectively reverts joyent/node@a05f973
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
Invokes the configure script used to build binary packages
(OSX pkg, binary tarballs, pkgsrc, MSI) with --download=all
--with-intl=small-icu.
Also makes PACKAGEMAKER customizable, because PackageMaker is not
necessarily installed in /Developer on OSX anymore.
Tested all binary packages on Windows, OSX, Linux and SmartOS.
Fixes#7676.
Reviewed-by: Steven R. Loomis <srl@icu-project.org>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The tests suite available in test/external/ssl-options was originally
written for security fixes made in the v0.10 branch. In this branch, the
client's default ciphers list is compatible with SSLv2.
After merging this change from v0.10 to v0.12, this tests suite was
broken because commits 5d2aef17ee and
f4c8020d10 make SSL/TLS clients use a
default ciphers list that is not compatible with the SSLv2 protocol.
This change fixes two issues:
1) The cipher list that was setup for a given test was not passed
properly to the client.
2) When either or both of clients/servers were using SSLv2, tests were
expected to succeed when at least the server end was using SSLv2
compatible ciphers. Now, tests are expected to succeed only if
SSLv2 compatible ciphers are used on both ends.
Fixes#9020.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
A recent merge of v0.10 to v0.12
(0c7f6ca830) upgraded OpenSSL to version
1.0.1j. In v0.10, this required test-crypto-stream.js to be fixed with
commit 707cc25011. Basically, instead of
returning the proper error, Err_get_error() would return 0 and the test
for the error message needed to be updated in test-crypto-stream.js.
However, in the v0.12 branch, crypto error messages are handled a bit
differently since commit 26a1b712ec
landed. Instead of returning the default OpenSSL error message, it makes
the decipher stream return a default message specific to Node.js.
This commit updates test-crypto-stream.js to test the error object
against the proper default error message.
Fixes#9019.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Workers that are already disconnected but not yet exited should not be
disconnected, trying to do so raises exceptions.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
- `sh.css` already exists in `api_assets`
- `sh_vim-dark.css` is unused, but used in the repo `node-website`
now
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
Backport 10703774f0 from the v0.12 branch.
Currently, the test-child-process-spawn-typeerror.js is
calling execFile() on a JavaScript source file, which is
causing failures on Windows. This commit switches to calling
spawn() on an actual executable.
Fixes#8930.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
child_process.spawn's argument parsing is stricter in v0.12 than in
v0.10. Changes in tests merged from v0.10 that relied on the less-strict
argument parsing would fail. This change updates the test so that it
makes sure that the stricter argument parsing fails as expected.
This change also fixes a small typo introduced during the conflicts
resolution of said merge.
fs.exists() and fs.existsSync() do not follow the typical error first
callback convention. access() and accessSync() are added as alternatives
in this commit.
PR-URL: https://github.com/joyent/node/pull/8714
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Currently, the test-child-process-spawn-typeerror.js is
calling execFile() on a JavaScript source file, which is
causing failures on Windows. This commit switches to calling
spawn() on an actual executable.
Reviewed-by: Sam Roberts <sam@strongloop.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
Some tests that rely on some environment variables being passed to child
processes would fail because they reset the child processes'
environement instead of appending to it. This would break on test
environments where some custom environment variables are needed to make
node work properly.
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reduce the overhead of the CPU profiler by replacing sched_yield() with
nanosleep() in V8's tick event processor thread. The former only yields
the CPU when there is another process scheduled on the same CPU.
Before this commit, the thread would effectively busy loop and consume
100% CPU time. By forcing a one nanosecond sleep period rounded up to
the task scheduler's granularity (about 50 us on Linux), CPU usage for
the processor thread now hovers around 10-20% for a busy application.
PR-URL: https://github.com/joyent/node/pull/8789
Ref: https://github.com/strongloop/strong-agent/issues/3
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The latest merge resulted in uv__loop_configure being defined twice on
Windows. This changes removes one of these duplicates to fix the build
on this platform.
If the data length passed to smalloc.alloc() the array_length will be
zero, causing an overflow check to fail. This prevents that from
happening.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Make "--with-intl=none" the default and add "intl-none" option to
vcbuild.bat.
If icu data is missing print a warning unless either --download=all or
--download=icu is set. If set then automatically download, verify (MD5)
and unpack the ICU data if not already available.
There's a "list" of URLs being used, but right now only the first is
picked up. The logic works something like this:
* If there is no directory deps/icu,
* If no zip file (currently icu4c-54_1-src.zip),
* Download zip file (icu-project.org -> sf.net)
* Verify the MD5 sum of the zipfile
* If bad, print error and exit
* Unpack the zipfile into deps/icu
* If deps/icu now exists, use it, else fail with help text
Add the configuration option "--with-icu-source=..."
Usage:
* --with-icu-source=/path/to/my/other/icu
* --with-icu-source=/path/to/icu54.zip
* --with-icu-source=/path/to/icu54.tgz
* --with-icu-source=http://example.com/icu54.tar.bz2
Add the configuration option "--with-icu-locals=...". Allows choosing
which locales are used in the "small-icu" case.
Example:
configure --with-intl=small-icu --with-icu-locales=tlh,grc,nl
(Also note that as of this writing, neither Klingon nor Ancient Greek
are in upstream CLDR data. Serving suggestion only.)
Don't use hard coded ../../out paths on windows. This was suggested by
@misterdjules as it causes test failures. With this fix, "out" is no
longer created on windows and the following can run properly:
python tools/test.py simple
Reduce space by about 1MB with ICU 54 (over without this patch). Also
trims a few other source files, but only conditional on the exact ICU
version used. This is to future-proof - a file that is unneeded now may
be needed in future ICUs.
Also:
* Update distclean to remove icu related files
* Refactor some code into tools/configure.d/nodedownload.py
* Update docs
* Add test
PR-URL: https://github.com/joyent/node/pull/8719
Fixes: https://github.com/joyent/node/issues/7676#issuecomment-64704230
[trev.norris@gmail.com small change to test's whitespace and logic]
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The url.parse() function now checks whether an escapable character is in
the URL before trying to escape it.
PR-URL: https://github.com/joyent/node/pull/8638
[trev.norris@gmail.com: Switch to use continue instead of if]
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Commit 934bfe23a1 had introduced a
regression where node would crash trying to access a null unref timer if
a given unref timer's callback would remove other unref timers set to
fire in the future.
More generally, it makes the unrefTimeout function more solid by not
mutating the unrefList while traversing it.
Fixes#8897.
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
This change fixes a regression introduced by commit
0d051238be, which contained a typo that
would cause every unrefd interval to fire only once.
Fixes#8900.
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Add documentation for the callback parameter of http.ClientRequest's and
http.ServerResponse's end methods.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
In cases where many small writes are made to a stream
lacking _writev, the array data structure backing the
WriteReq buffer would greatly increase GC pressure.
Specifically, in the fs.WriteStream case, the
clearBuffer routine would only clear a single WriteReq
from the buffer before exiting, but would cause the
entire backing array to be GC'd. Switching to [].shift
lessened pressure, but still the bulk of the time was
spent in memcpy.
This replaces that structure with a linked list-backed
queue so that adding and removing from the queue is O(1).
In the _writev case, collecting the buffer requires an
O(N) loop over the buffer, but that was already being
performed to collect callbacks, so slowdown should be
neglible.
PR-URL: https://github.com/joyent/node/pull/8826
Reviewed-by: Timothy J Fontaine <tjfontaine@gmail.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fix a Windows-only build error that was introduced in
commit 1183ba4 ("zlib: support concatenated gzip files").
Rename the NO_ERROR and FAILED enumerations, they conflict
with macros of the same name in <winerror.h>.
PR-URL: https://github.com/joyent/node/pull/8893
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Reviewed-By: Rod Vagg <rod@vagg.org>
Reviewed-by: Timothy J Fontaine <tjfontaine@gmail.com>
Original commit message:
Fix Unhandled ReferenceError in debug-debugger.js
This fixes following exception in Sky on attempt to set a breakpoint
"Unhandled: Uncaught ReferenceError: break_point is not defined"
I think this happens in Sky but not in Chrome because Sky scripts are executed in strict mode.
BUG=None
LOG=N
R=yangguo@chromium.org
Review URL: https://codereview.chromium.org/741683002
Cr-Commit-Position: refs/heads/master@{#25415}
Before this change, _unrefActive would keep the unrefList sorted when
adding a new timer.
Because _unrefActive is called extremely frequently, this linear scan
(O(n) at worse) would make _unrefActive show high in the list of
contributors when profiling CPU usage.
This commit changes _unrefActive so that it doesn't try to keep the
unrefList sorted. The insertion thus happens in constant time.
However, when a timer expires, unrefTimeout has to go through the whole
unrefList because it's not ordered anymore.
It is usually not large enough to have a significant impact on
performance because:
- Most of the time, the timers will be removed before unrefTimeout is
called because their users (sockets mainly) cancel them when an I/O
operation takes place.
- If they're not, it means that some I/O took a long time to happen, and
the initiator of subsequents I/O operations that would add more timers
has to wait for them to complete.
With this change, _unrefActive does not show as a significant
contributor in CPU profiling reports anymore.
Fixes#8160.
PR-URL: #8751
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
ICU support in v0.12 generates a new icu_config.gypi. This was
accidentally committed after switching branches. The file has been removed
and added to .gitignore.
Fixes: 0d051238 "timers: fix unref() memory leak"
Previously, 'configure' would not return an exit status
if gyp blows up. This can be tested via:
date >> node.gyp ; ./configure && echo A-OK
You will get "A-OK" even though gyp had failed.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
The original documentation was slightly confusing. It seemed that the
list of items described the properties of the urlObj object, while it
was actually describing the formatting process. This change makes this
clearer.
Fixes#8796.
Reviewed-by: Chris Dickinson <christopher.s.dickinson@gmail.com>
Documentation mentioned 'securityOptions', where it should have read
'secureOptions'.
Fixes#8608.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
Dtrace probes were removed from libuv recently, but their usage by node
was not completely removed, causing build breaks on SmartOS.
Even though the build is working on other platforms, these probes are
not fired by libuv anymore, so there's no point in using them on these
platforms too.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Code examples in documentation for net.createServer and
net.createConnection contained confusing log messages. This change makes
them clearer.
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
Currently there's an example using http.ServerResponse stream, which
has a known bug and will not throw an error while writing after end().
Changed to a writable stream from fs which behaves as expected.
fix#8814
Signed-off-by: Julien Gilli <julien.gilli@joyent.com>
In `tls.markdown`, there was a misuse of 'a' which has been replaced
with 'an'.
In `timers.markdown`...
line 31: misuse of 'a', replaced with 'an'
line 59: unclear wording, haywire 'a', added new comma
Call a user-defined callback at specific points in the lifetime of an
asynchronous event. Which are on instantiation, just before/after the
callback has been run.
**If any of these callbacks throws an exception, there is no forgiveness
or recovery. A message will be displayed and a core file dumped.**
Currently these only tie into AsyncWrap, meaning no call to a hook
callback will be made for timers or process.nextTick() events. Though
those will be added in a future commit.
Here are a few notes on how to make the hooks work:
- The "this" of all event hook callbacks is the request object.
- The zero field (kCallInitHook) of the flags object passed to
setupHooks() must be set != 0 before the init callback will be called.
- kCallInitHook only affects the calling of the init callback. If the
request object has been run through the create callback it will always
run the before/after callbacks. Regardless of kCallInitHook.
- In the init callback the property "_asyncQueue" must be attached to
the request object. e.g.
function initHook() {
this._asyncQueue = {};
}
- DO NOT inspect the properties of the object in the init callback.
Since the object is in the middle of being instantiated there are some
cases when a getter is not complete, and doing so will cause Node to
crash.
PR-URL: https://github.com/joyent/node/pull/8110
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
When instantiating a new AsyncWrap allow the parent AsyncWrap to be
passed. This is useful for cases like TCP incoming connections, so the
connection can be tied to the server receiving the connection.
Because the current architecture instantiates the *Wrap inside a
v8::FunctionCallback, the parent pointer is currently wrapped inside a
new v8::External every time and passed as an argument. This adds ~80ns
to instantiation time.
A future optimization would be to add the v8::External as the data field
when creating the v8::FunctionTemplate, change the pointer just before
making the call then NULL'ing it out afterwards. This adds enough code
complexity that it will not be attempted until the current approach
demonstrates it is a bottle neck.
PR-URL: https://github.com/joyent/node/pull/8110
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
Instead of simply creating a new v8::Object to contain the connection
information, instantiate a new instance of a FunctionTemplate. This will
allow future improvements for debugging and performance probes.
Additionally, the "provider" argument in the ReqWrap constructor is no
longer optional.
PR-URL: https://github.com/joyent/node/pull/8110
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
C++ won't deoptimize like JS if specific conditional branches are
sporadically met in the future. Combined with the amount of code
duplication removal and simplified maintenance complexity, it makes more
sense to merge MakeCallback and MakeDomainCallback.
Additionally, type casting in V8 before verifying what that type is will
cause V8 to abort in debug mode if that type isn't what was expected.
Fix this by first checking the v8::Value before casting.
PR-URL: https://github.com/joyent/node/pull/8110
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Julien Gilli <julien.gilli@joyent.com>
Adding --flaky-tests option, to allow regarding flaky tests failures
as non-fatal.
Currently only observed by the TapProgressIndicator, which will
add a # TODO directive to tests classified as flaky. According to the
TAP specification, the test harness is supposed to treat failures
that have a # TODO directive as non-fatal.
In very unlikely case, where `deflateInit2()` may return error (right
now happening only on exhausting all memory), the `ZCtx::Error()` will
be called and will try to `Unref()` the handle. But the problem is that
this handle was never `Ref()`ed, so it will trigger an assertion error
and crash the program.
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
PR-URL: https://github.com/joyent/node/pull/8687
Remove unneeded functionality and tweak the generated output so we
can #include it in C++ source code.
This is a back-port of commit e159073 from the master branch.
It was my mistake to change an assert check. This changes it back to how
the assert was originally done.
Fixes: c131c1f "modules: adding load linked modules feature"
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
- introduced NM_F_LINKED flag to identify linked modules
- setting node_is_initialized after calling V8::Initialize in order to
make the right decision during initial module registration
- introduced modlist_linked in order to track modules that were
pre-registered in order to complete it once node is initialized
- completing registration of linked module similarly to the way it's
done inside DLOpen
PR-URL: https://github.com/joyent/node/pull/8386
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
In 59658a8de7
the return of uv_thread_self() was changed from unsigned long to
uv_thread_t.
uv_thread_t is a HANDLE (pointer-sized) on Windows, which means that
on Win64 it cannot be stored with CRYPTO_THREADID_set_numeric without
data loss.
Furthermore, without this change there will be a build break on Windows
when the libuv change is integrated into Node, because of the
conversion from HANDLE to unsigned long.
Other related commits:
5845a6bcd5919d8ec63a
Should work in all platforms and it fixes this compilation problem
on OSX:
../src/node_crypto.cc:154:3: error: no matching function for call to
'CRYPTO_THREADID_set_numeric'
CRYPTO_THREADID_set_numeric(tid, uv_thread_self());
^~~~~~~~~~~~~~~~~~~~~~~~~~~
../deps/openssl/openssl/include/openssl/../../crypto/crypto.h:435:6:
note: candidate function not viable: no known conversion from
'uv_thread_t' (aka '_opaque_pthread_t *') to 'unsigned long'
for 2nd argument
void CRYPTO_THREADID_set_numeric(CRYPTO_THREADID *id, unsigned
long val);
^
1 error generated.
PR-URL: https://github.com/joyent/node/pull/8785
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The normalizeArray() function now avoids using the slow Array#splice()
method to improve performance and now also filters out empty path parts.
Code that pre-filtered empty parts has been removed.
PR-URL: https://github.com/joyent/node/pull/8724
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
- Add hyperlinks from spawn options to subsections detailing what
those options do.
- Clarify some verbiage around ChildProcess.prototype.std{in,out,err}.
- Remove second-person pronoun.
PR-URL: https://github.com/joyent/node/pull/8639
Reviewed-by: Chris Dickinson <christopher.s.dickinson@gmail.com>
execFile and spawn have same API signature with respect to optional arg
array and optional options object, they should have same behaviour with
respect to argument validation.
PR-URL: https://github.com/joyent/node/pull/8454
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The test wasn't checking directly that an assertion was thrown. Instead,
it was checking that spawn did not sucessfully spawn a non-existent
command.
However, the command chosen, dir, exists in GNU coreutils, so it exists
on Linux (though not on BSD derived OS X). The test as written passed on
Linux, even with the TypeError it is supposed to be checking for deleted
from spawn(). It would also pass on Windows if a ls.exe existed.
The approach is unnecessarily obscure, assert.throw() is for asserting
code throws, using it is more clear and works regardless of what
commands do or do not exist.
PR-URL: https://github.com/joyent/node/pull/8454
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
So what I did here is to make the icutools library actually contain the
entire ICU dependencies needed for host-side tools. Sadly, this will
build ICU twice, but avoids conflicts between host and target side.
This all seems like a gyp bug of some sort, but without docs for
toolsets, who’s to say?
I removed the icuio library as a separate target, because it was only
used by the host-side tools.
PR-URL: https://github.com/joyent/node/pull/8681
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Revert uv_thread_self() to return unsigned long instead of uv_thread_t.
This was causing a build failure on Windows and is only a temporary fix
until the proper patch lands upstream.
Reverts: https://github.com/joyent/libuv/commit/59658a8
Fixes: ce112c2 "deps: update uv to v1.0.0-rc2"
Add a test that goes through the whole matrix of:
- command line options (--enable-ssl*)
- secureOptions
- secureProtocols
and makes sure that compatible test setups actually work as expected.
The test works by spawning two processes for each test case: one client
and one server. The test passes if a SSL/TLS connection from the client
to the server is successful and the test case was supposed to pass, or
if the connection couldn't be established and the test case was supposed
to fail.
The test is currently located in the directory 'test/external' because
it has external dependencies.
* openssl: Update to 1.0.1j (Addressing multiple CVEs)
* uv: Update to v0.10.29
* child_process: properly support optional args (cjihrig)
* crypto: Disable autonegotiation for SSLv2/3 by default (Fedor Indutny,
Timothy J Fontaine, Alexis Campailla)
This is a behavior change, by default we will not allow the negotiation to
SSLv2 or SSLv3. If you want this behavior, run Node.js with either
`--enable-ssl2` or `--enable-ssl3` respectively.
This does not change the behavior for users specifically requesting
`SSLv2_method` or `SSLv3_method`. While this behavior is not advised, it is
assumed you know what you're doing since you're specifically asking to use
these methods.
Always set ssl2/ssl3 disabled based on whether they are enabled in Node.
In some corner-case scenario, node with OPENSSL_NO_SSL3 defined could
be linked to openssl that has SSL3 enabled.
In the case of a pipe'd input, i.e. from the CI the fd will be a PIPE
and when listen() is called it will return ENOTSOCK instead of EINVAL.
Backport: cd2d3aedaa
The order of the callbacks is non-deterministic, so don't expect the
error messages to come back in the same order every time, instead just
verify they are expected messages.
This change disables SSLv2/SSLv3 use by default, and introduces a
command line flag to opt into using SSLv2/SSLv3.
SSLv2 and SSLv3 are considered unsafe, and should only be used in
situations where compatibility with other components is required and
they cannot be upgrade to support newer forms of TLS.
Because of constant-timeness change made in openssl-1.0.1j the error is
no longer returned from EVP_DecryptFinal_ex. Now it just return 0, and
thus the error message does not contain proper error code. Adapt to this
change, there is not much that we could do about it.
vcbuild.bat is calling vcvars.bat, which doesn't detect if the environment
has already been set. This causes repeated entries to be added to the PATH,
which after a few invocations will lead to an error:
The input line is too long.
You cannot spawn 'dir' on Windows because it's not an executable. Also,
some people might have 'ls' on their path on Windows, so I changed
invalidCmd to something that's highly unlikely to exist.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The cluster children are hitting breakpoint at `cluster.onread` and
hanging on a Semaphore wait now. This prevents them from disconnecting
gracefully. Considering that the test is checking different thing, the
cluster children needs to be force killed from the grand parent process.
Reviewed-By: Trevor Norris <trevnorris@gmail.com>
PR-URL: https://github.com/joyent/node/pull/8476
Because of behavior change of some V8 APIs (they mostly became more
strict), following modules needed to be fixed:
* crypto: duplicate prototype methods are not allowed anymore
* contextify: some TryCatch trickery, the binding was using it
incorrectly
* util: maximum call stack error is now crashing in a different place
Reviewed-By: Trevor Norris <trevnorris@gmail.com>
PR-URL: https://github.com/joyent/node/pull/8476
Do not send signal to children if they are already in debug mode.
Node.js on Windows does not register signal handler, and thus calling
`process._debugProcess()` will throw an error.
Reviewed-By: Trevor Norris <trevnorris@gmail.com>
PR-URL: https://github.com/joyent/node/pull/8476
* Add official documentation that a Buffer instance is a viable
argument when instantiating a new Buffer.
* Properly set the poolOffset when a buffer needs to be truncated.
* Add comments clarifying specific peculiar coding choices.
* Remove a level of unnecessary indentation.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The constants in enum v8::ExternalArrayType have been changed. The old
values are there for legacy reasons, but it's best to update anyway.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
* `util.inspect` cannot accept es6 symbol primitive
* It will throw exception if do `util.inspect(Symbol())`
* This also affects repl, console.log, etc.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Chris Dickinson <christopher.s.dickinson@gmail.com>
Add generic functions for (U)Int read/write operations on Buffers. These
support up to and including 48 bit reads and writes.
Include documentation and tests.
Additional work done by Trevor Norris to include 40 and 48 bit write
support. Because bitwise operations cannot be used on values greater
than 32 bits, the operations have been replaced with mathematical
calculations. Regardless, they are still faster than floating point
operations.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Performance improvement by moving checks for floating point operations
to JS and doing the operation on a protected internal function that
assumes all arguments are correct. Still abort if the operation
overflows memory. This can only be caused if the Buffer's length
property isn't the same as the actual internal length.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Passing null as the output stream to readline.Interface()'s constructor
is now supported. Any output written by readline is just discarded. It
makes it easier to use readline just as a line parser.
Fixes: https://github.com/joyent/node/issues/4408
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Since we are taking control of the microtask queue it makes sense to
disable autorun and only run microtasks when necessary. Just setting
isolate->SetAutorunMicrotasks(false) would cause _tickCallback() not to
be called.
Automatically running the microtask queue will cause it to run:
* After callback invocation
* Inside _tickCallback()
* After _tickCallback() invocation
The third one is unnecessary as the microtask queue is guaranteed to be
empty at this point. The first only needs to be run manually when
_tickCallback() isn't going to be called by MakeCallback().
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Match the behavior of the slow path by setting url.query to an empty
object when the url contains no query, but query parsing is requested.
Also add a test for this case, and update the documents to clearly
reflect this behavior.
Fixes: https://github.com/joyent/node/issues/8332
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
The two main goals of this change are:
- To make it easier to build the Intl option using ICU (particularly,
using a newer ICU than v8/Chromium's version)
- To enable a much smaller ICU build with only English support The goal
here is to get node.js binaries built this way by default so that the
Intl API can be used. Additional data can be added at execution time
(see Readme and wiki)
More details are at https://github.com/joyent/node/pull/7719
In particular, this change adds the "--with-intl=" configure option to
provide more ways of building "Intl":
- "full-icu" picks up an ICU from deps/icu
- "small-icu" is similar, but builds only English
- "system-icu" uses pkg-config to find an installed ICU
- "none" does nothing (no Intl)
For Windows builds, the "full-icu" or "small-icu" options are added to
vcbuild.bat.
Note that the existing "--with-icu-path" option is not removed from
configure, but may not be used alongside the new option.
Wiki changes have already been made on
https://github.com/joyent/node/wiki/Installation
and a new page created at
https://github.com/joyent/node/wiki/Intl
(marked as provisional until this change lands.)
Summary of changes:
* README.md : doc updates
* .gitignore : added "deps/icu" as this is the location where ICU is
unpacked to.
* Makefile : added the tools/icu/* files to cpplint, but excluded a
problematic file.
* configure : added the "--with-intl" option mentioned above.
Calculate at config time the list of ICU source files to use and data
packaging options.
* node.gyp : add the new files src/node_i18n.cc/.h as well as ICU
linkage.
* src/node.cc : add call into
node::i18n::InitializeICUDirectory(icu_data_dir) as well as new
--icu-data-dir option and NODE_ICU_DATA env variable to configure ICU
data loading. This loading is only relevant in the "small"
configuration.
* src/node_i18n.cc : new source file for the above Initialize..
function, to setup ICU as needed.
* tools/icu : new directory with some tools needed for this build.
* tools/icu/icu-generic.gyp : new .gyp file that builds ICU in some new
ways, both on unix/mac and windows.
* tools/icu/icu-system.gyp : new .gyp file to build node against a
pkg-config detected ICU.
* tools/icu/icu_small.json : new config file for the "English-only" small
build.
* tools/icu/icutrim.py : new tool for trimming down ICU data. Reads the
above .json file.
* tools/icu/iculslocs.cc : new tool for repairing ICU data manifests
after trim operation.
* tools/icu/no-op.cc : dummy file to force .gyp into using a C++ linker.
* vcbuild.bat : added small-icu and full-icu options, to call into
configure.
* Fixed toolset dependencies, see
https://github.com/joyent/node/pull/7719#issuecomment-54641687
Note that because of a bug in gyp {CC,CXX}_host must also be set.
Otherwise gcc/g++ will be used by default for part of the build.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
Running fill() with an empty string would cause Node to hang
indefinitely. Now it will return without having operated on the buffer.
User facing function has been pulled into JS to perform all initial
value checks and coercions. The C++ method has been placed on the
"internal" object.
Coerced non-string values to numbers to match v0.10 support.
Simplified logic and changed a couple variable names.
Added tests for fill() and moved them all to the beginning of
buffer-test.js since many other tests depend on fill() working properly.
Fixes: https://github.com/joyent/node/issues/8469
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The parameter parser specifically looked for the old bracket syntax.
This generated a lot of warnings when building the docs. Those warnings
have been fixed by changing the parsing logic.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Several fields on OutgoingMessage were set after instantiation. These
have been included in the constructor to prevent mutation of the object
map after instantiation.
"name" is now explicitly checked to be a string. Where before if a
non-string was passed the following cryptic error was thrown:
_http_outgoing.js:334
var key = name.toLowerCase();
^
TypeError: undefined is not a function
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Calls from JS to C++ have an implicit HandleScope. So there is no need
to instantiate a new HandleScope in these basic cases.
Check if the returned int64_t is an SMI and cast the return value to
uint32_t instead of a double. Prevents needing to box the return value,
and saves a small amount of execution time.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Don't allocate any BIO buffers initially, do this on a first read from
the TCP connection. Allocate different amount of data for initial read
and for consequent reads: small buffer for hello+certificate, big buffer
for better throughput.
see #8416
Add stricter argument type checking to normalizeSpawnArguments().
Removes a number of extraneous checks in spawn().
Fix regression in handling of the optional args argument.
Add more thorough testing of spawn() arguments.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Only run lineEnding.test() on the newly acquired chunk of string instead
of on the entire line buffer.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Documentation incorrectly used bracket notation for optional parameters.
This caused inconsistencies in usage because of examples like the
following:
fs.write(fd, data[, position[, encoding]], callback)
This simply fixes all uses of bracket notation in documentation.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
The property server.connections should no longer be enumerable because
it has been deprecated. This will prevent deprecation warnings when
server objects are accessed by functions such as JSON.stringify.
Fixes: https://github.com/joyent/node/issues/8373
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Adds a test for benchmarking the module loader, needed for benchmarking
changes / refacortings in the module loader.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Timothy J Fontaine <tjfontaine@gmail.com>
4c9b30d introduced a regression in buffer.slice that 7c3c51b fixed, but
no test had been added to make sure that a similar regression is caught
by the tests suite in the future.
In 4c9b30d removal of the prototype attributes meant NativeBuffer() no
longer had the same object map as Buffer(). By now setting the same
properties in the same order both constructors will produce the same
map.
The same commit changed "parent" from undefined to null. This caused a
failure in Buffer#slice() where it was checked if parent === undefined.
Causing the incorrect parent to be set.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Expands the paragraph in the transform stream
implementation docs about the callback that is passed
to the _transform method to include details about how
two arguments may be passed, error and data. A code
example is also included.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
When calling write() after end() has been called on an OutgoingMessage,
an error is emitted and the write's callback is called with an instance
of Error.
Fix#7477.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Increase the performance of new Buffer construction by initializing all
properties before SetIndexedPropertiesToExternalArrayData call.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
When V8 started supporting Promises natively it also introduced a
microtack queue. This feature operates similar to process.nextTick(),
and created an issue where neither knew when the other had run. This
patch has nextTick() call the microtask queue runner at the end of
processing callbacks in the nextTickQueue.
Fixes: https://github.com/joyent/node/issues/7714
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Add a simple test to cover workers' implementation of
Worker.prototype.destroy(). Before adding this test, this code wouldn't
be covered by the tests suite, and any regression introduced in workers'
implementation of Worker.prototype.destroy wouldn't be caught.
Fixes: https://github.com/joyent/node/issues/8223
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
* npm: Update to 1.4.28
* v8: fix a crash introduced by previous release (Fedor Indutny)
* configure: add --openssl-no-asm flag (Fedor Indutny)
* crypto: use domains for any callback-taking method (Chris Dickinson)
* http: do not send `0rnrn` in TE HEAD responses (Fedor Indutny)
* querystring: fix unescape override (Tristan Berger)
* url: Add support for RFC 3490 separators (Mathias Bynens)
When replying to a HEAD request, do not attempt to send the trailers and
EOF sequence (`0\r\n\r\n`). The HEAD request MUST not have body.
Quote from RFC:
The presence of a message body in a response depends on both the
request method to which it is responding and the response status code
(Section 3.1.2). Responses to the HEAD request method (Section 4.3.2
of [RFC7231]) never include a message body because the associated
response header fields (e.g., Transfer-Encoding, Content-Length,
etc.), if present, indicate only what their values would have been if
the request method had been GET (Section 4.3.1 of [RFC7231]).
fix#8361
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
This adds domains coverage for pdbkdf2, pseudoRandomBytes, and randomBytes.
All others should be covered by event emitters.
Fixes#5801.
Reviewed-By: Timothy J Fontaine <tjfontaine@gmail.com>
Compiles and executes source code in V8's debugger context. Provides
a programmatic way to get access to the debug object by executing:
var Debug = vm.runInDebugContext('Debug');
Fixes#7886.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Strings are treated as UTF8 instead of one-byte strings when
names are processed and when OpenSSL's ..._print functions are used.
This commit fixes simple/test-tls-peer-certificate-encoding test.
fix#8366
The behavior of the `node_modules` lookup algorithm was
changed in #1177, but the documentation was not updated completely
to describe the new behavior.
The pseudocode of the lookup algorithm did not metion that
`index.json` is tried to be loaded if you require a folder.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
V8 3.26.31 has received 14 patches since the upgrade to 3.26.33. Since
3.26.33 is technically a tag on the 3.27 branch, reverting back to
3.26.31 would remove now default functionality like WeakMaps. Because of
that the patches have simply been cherry-picked and squashed.
Here is a summary of all patches:
* Fix index register assignment in LoadFieldByIndex for arm, arm64, and
mips.
* Fix invalid attributes when generalizing because of incompatible map
change.
* Skip write barriers when updating the weak hash table.
* MIPS: Avoid HeapObject check in HStoreNamedField.
* Do GC if CodeRange fails to allocate a block.
* Array.concat: properly go to dictionary mode when required.
* Keep CodeRange::current_allocation_block_index_ in range.
* Grow heap slower if GC freed many global handles.
* Do not eliminate bounds checks for "<const> - x".
* Add missing map check to optimized f.apply(...).
* In GrowMode, force the value to the right representation to avoid
deopts between storing the length and storing the value.
* Reduce max executable size limit.
* Fix invalid condition in check elimination effects.
* Fix off-by-one error in Array.concat slow mode check.
For more information see: https://github.com/v8/v8/commits/3.26
Reviewed-By: Fedor Indutny <fedor@indutny.com>
There is only one call site that uses it and that can do the checks
itself. Removes ~15 lines of code.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Unexport the http.parsers freelist. It was originally exported by Ryan
in commit 0003c701 but the commit log doesn't mention why and it's never
been documented. It's unclear if there are any users.
The lifecycle of parser objects changed recently and it seems better to
not let people shoot themselves in the foot so easily.
If it turns out there are actually users, we can always re-export it
again - probably under a slightly different name, to force people to
update their code to the new way of things.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Make it a little harder to slip in use-after-free bugs by nulling out
references to the parser object after handing it off to freeParser().
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
It's safe to call BaseObject::object() from your destructor _unless_
the handle is weak; then it's the weak callback that is calling your
destructor and the object will have been released by the time the
destructor runs.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Weak handles put strain on the garbage collector and the parser handle
doesn't need to be weak in the first place. This change should improve
GC times on busy servers a little.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Counterpart to Wrap(), clears the previously assigned internal field.
Will be used in an upcoming commit.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fix a resource leak where an intermediate Local<Context> handle in
Environment::GetCurrent() got leaked into whatever HandleScope was
further up the stack.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Allow cluster workers to listen on exclusive ports for TCP and UDP,
instead of forcing all calls to go through the cluster master.
Fixes: #3856
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Reviewed-by: Fedor Indutny <fedor@indutny.com>
node::StringBytes::Write() has appropriate support to write strings with
'binary' encoding. So expose that API through StreamWrap and allow
inheriting classes to use it.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The REPL global object lazy loads modules by placing getters for each.
This causes MakeDomainCallback() to be run if a native module is loaded
from the REPL, but if the domain module hasn't been loaded then there
are no enter/exit callbacks to be called. Causing an assert() to fail.
Fix the issue by conditionally running the callback instead of asserting
it is available. Also add "addon" test to verify the fix.
Fixes: #8231
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
PR #8034 came with a test to make sure that timers expiry is based on
monotonic time and not on wall-clock time. However, a bug in the
implementation broke timers with non-integer delays. A fix for this
issue was provided with PR #8073, but it didn't come with a test.
Because #8073 fixed a subtle issue that could reappear in the future,
and because the impact of such an issue would be significant, I suggest
adding this test.
The test would timeout after 1 minute if the issue was reproduced.
Otherwise it will run very quickly.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Allow to create an executable with no external dynamic libraries, also the
ones from the system. This is somewhat dependent of the used C lib, for
example glibc has some internal dynamic libraries loaded by itself, but for
other ones like eglibc or dietlib, this would produce a true static linked
executable. This can be of interest for embebers or resource constraints
platforms, but the main reason for this is to allow to use a Javascript
file as Linux kernel 'init' on NodeOS.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Currently, cluster workers can be removed from the workers list in three
different places:
- In the exit event handler for the worker process.
- In the disconnect event handler of the worker process.
- In the disconnect event handler of the cluster master.
However, handles for a given worker are cleaned up only in one of these
places: in the cluster master's disconnect event handler.
Because these events happen asynchronously, it is possible that the
workers list is empty before we even clean up one handle. This makes
the assert that makes sure that no handle is left when the workers
list is empty fail.
This commit removes the worker from the cluster.workers list only when
the worker is dead _and_ disconnected, at which point we're sure that
its associated handles are cleaned up.
Fixes#8191 and #8192.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
In case of an invalid DH parameter file, it is sliently discarded. To
use auto DH parameter in a server and DHE key length check in a
client, we need to wait for the next release of OpenSSL-1.0.2.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
There is no need to split the host by hand in `url.js` – Punycode.js
takes care of it anyway. This not only simplifies the code, but also
adds support for RFC 3490 separators (i.e. not just U+002E, but U+3002,
U+FF0E, and U+FF61 as well).
Closes#6055.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Documentation states that `querystring.unescape` may be overridden to
replace unescaper during parsing. However, the function was only
being used as a fallback for when the native decoder throws (on a
malformed URL). This patch moves the call to the native function and
the try/catch around it into querystring.unescape then has the parser
always invoke it, so that an override will always be used.
Fixes#4055
Reviewed-By: Fedor Indutny <fedor@indutny.com>
fd80a31e06 has introduced a segfault
during redundant boundary check elimination (#8208).
The problem consists of two parts:
1. Abscense of instruction iterator in
`EliminateRedundantBoundsChecks`. It was present in recent v8, but
wasn't considered important at the time of backport. However, since
the function is changing instructions order in block, it is
important to not rely at `i->next()` at the end of the loop.
2. Too strict ASSERT in `MoveIndexIfNecessary`. It is essentially a
backport of a45c96ab from v8's upstream. See
https://github.com/v8/v8/commit/a45c96ab for details.
fix#8208
net Sockets were calling read(0) to start reading, without
checking to see if they were paused first. This would result
in paused Socket objects keeping the event loop alive.
Fixes#8200
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fix assertion failure from poor argument parsing logic introduced in
6ea5d16. Add tests to make sure arguments are properly parsed.
Fixes: 6ea5d16 "dns: always set variable family in lookup()"
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fix a few issues in test/internet/test-dns.js:
- 'hint' should be 'hints'
- reverse name lookup is not guaranteed to return 'localhost'
- V4MAPPED hint requires IPV6 address family
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Tests in test-net-remote-address-port.js assume that client and server
sockets always use IPv4. However, depending on the OS and the network
interfaces setup, this is not true. This change makes the test consider
that both IPv4 or IPv6 sockets are valid
Fixes#8096.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Instead of hard-coding http service name in test-dns, retrieve it from
/etc/services. This is not ideal, but it's still better than hard-coding
it.
Fixes#8047.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
1 is actually a valid flag on SmartOS. More generally, hints flags'
values are defined by the underlying native flags, and these can have
different values on different systems.
Using (ADDRCONFIG | V4MAPPED) + 1 ensure that the flag will be invalid,
since it will always be different from ADDRCONFIG, V4MAPPED, ADDRCONFIG
| V4MAPPED, 0 and any other combination of even flags.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
The slot 0 and 1 had already been taken by "gin" and "blink" in Chrome,
and the size of isolate's slots is 4 by default, so using 3 should hopefully
make node work independently when embedded into other application.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Fixes an issue that caused the first querystring to be parsed prepending
a "?" in the first variable name on relative urls with no #fragment
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Fix issue where output of a native prototype method would simply print
[Function]. It will now print [Function: name].
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Regression occurred that prevented the variable "family" from being set
properly when the lookup() function's "options" parameter was passed a
number instead of an object.
Also included a sanity check by setting the default value of "family" to
a value that will not pass verification.
Fixes: e643fe4 "dns: fix GetAddrInfo assert"
Reviewed-by: Alexis Campailla <alexis@janeasystems.com>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Linking CoreFoundation for OSX is needed for OSX debugging features to
function properly.
For instance Instruments cannot record Heap Allocations if the
CoreFoundation is not linked.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Disabling the part of the test that relies on dispatching SIGHUP,
because sending SIGHUP is not supported on Windows.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Send messages until both the parent and the child process have received
at least one message. If at least one of them doesn't receive any
message, the test runner will make the test timeout.
Fixes#8046.
This is the Node side of the fix for Node's cluster module on Windows.
https://github.com/joyent/node/issues/7691
The other required part is
https://github.com/joyent/libuv/pull/1384
Windows and Unix return certain socket errors (i.e. EADDRINUSE) at
different times: bind on Windows, and listen on Unix.
In an effort to hide this difference, libuv on Windows stores such
errors in the bind_error field of uv_tcp_t, to defer raising it at
listen time.
This worked fine except for the case in which a socket is shared in
a Node cluster and a bind error occurs.
A previous attempt to fix this (
d1e6be14603da36fe00e
) was flawed becaused in an attempt to relay the error at the JS level
it caused the master to start accepting connections.
With this new approach, libuv itself is relaying the bind errors,
providing for a uniform behavior of uv_tcp_listen.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
The method GetAddrInfo() is used by more than just dns.lookup(), and in
those cases a third argument isn't passed. This caused the following
check to abort:
assert(args[3]->IsInt32());
Fixes: 4306786 "net: don't prefer IPv4 addresses during resolution"
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Currently the address resolution family defaults to IPv4. Instead remove
the preference and instead resolve to a family suitable for the host.
Expose the getaddrinfo flags and allow them to be passed.
Add documentation about new flags.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
On Windows, path.isAbsolute() returns an empty string on failed cases.
This forces the return value to always be boolean.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
When backporting f8193ab into v0.10, a regression was introduced. Timers
with non-integer timeout could trigger a infinite recursion with 100%
cpu usage. This commit backports 93b0624 which fixes the regression.
After backporting f8193ab, instead of using Date.now(), timers would use
Timer.now() to determine if they had expired. However, Timer.now() is
based on loop->time, which is not updated when a timer's remaining time
is > 0 and < 1. Timers would thus never timeout if their remaining time
was at some point > 0 and < 1.
With this commit, Timer.now() updates loop->time itself, and timers
always timeout eventually.
Fixes#8065 and #8068.
Callbacks in node are usually asynchronous, and should never be
sometimes synchronous, and sometimes asynchronous.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Internal function trim(arr). 2nd parameter of slice() should be slice's
end index (not included). Because of function normalize() (called before
trim()), "start" is always zero so the bug -for now- has no effect, but
its a bug waiting to happen.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
A streams1 stream will have its falsy values such as 0, false, or ""
eaten by the upgrade to streams2, even when objectMode is enabled.
Include test for said cases.
Reviewed-by: isaacs <i@izs.me>
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
This patch adds a fast path for parsing of simple path-only URLs, as commonly
found in HTTP requests received by a server.
Benchmark results [ms], before / after patch:
/foo/bar 0.008956 0.000418 (fast path used)
http://example.com/ 0.011426 0.011437 (normal slow path, no change)
In a simple 'ab' benchmark of a single-threaded web server, this patch
increases the request rate from around 6400 to 7400 req/s.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
Original commit message:
timers: use uv_now instead of Date.now
This saves a few calls to gettimeofday which can be expensive, and
potentially subject to clock drift. Instead use the loop time which
uses hrtime internally.
In addition to the backport, this commit:
- keeps _idleStart timers' property which is still set to
Date.now() to avoid breaking existing code that uses it, even if
its use is discouraged.
- adds automated tests. These tests use a specific branch of
libfaketime that hasn't been submitted upstream yet. libfaketime
is git cloned if needed when running automated tests.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
This is the Node side of the fix for Node's cluster module on Windows.
https://github.com/joyent/node/issues/7691
The other required part is
https://github.com/joyent/libuv/pull/1384
Windows and Unix return certain socket errors (i.e. EADDRINUSE) at
different times: bind on Windows, and listen on Unix.
In an effort to hide this difference, libuv on Windows stores such
errors in the bind_error field of uv_tcp_t, to defer raising it at
listen time.
This worked fine except for the case in which a socket is shared in
a Node cluster and a bind error occurs.
A previous attempt to fix this (
d1e6be14603da36fe00e
) was flawed becaused in an attempt to relay the error at the JS level
it caused the master to start accepting connections.
With this new approach, libuv itself is relaying the bind errors,
providing for a uniform behavior of uv_tcp_listen.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
v0.10 and node docs specific that in a worker, the 'message' and 'error'
event emits on process, and on cluster.worker.
Reviewed-By: Fedor Indutny <fedor@indutny.com>
internet/test-dns.js assumes that ::1 always resolves to "localhost" on
all platforms. This is not what happens in reality. Some platforms
resolve it to "ip6-localhost" too. There doesn't seem to be any consensus
on what's the right thing to do. However, most sane platforms will use
either one of these two values.
Reviewed-by: Trevor Norris <trev.norris@gmail.com>
Prevent test-process-kill-pid.js tests suite from sending SIGHUP
to its process group, which was causing the test runner to terminate.
Fix jenkins' jobs for nodejs-master.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Between 0.11.1 and 0.11.2, the message and error events stopped
being usable via the cluster.worker object. This commit makes
them usable again. Closes#7998.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Currently, invalid usage such as:
process.kill('SIGTERM')
process.kill(null)
process.kill(undefined);
all coerce the pid to 0, and signal the current process.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Currently, stringification of an empty array outputs a single
separator character. This commit causes an empty array to output
the empty string.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
When TLS Alert is occured in handshake, ClearOut only write it into
wbio and does not flush to socket. TLS Alert should be written to
socket with EncOut before socket is destroyed within its error
callback.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
According to V8 changelog, `armv7` config variable was replaced by
`arm_version`, with value either '7', '6' or 'default'.
Detect ARMv7 and ARMv6 CPUs and default to 'default'.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Currently, checkExecSyncError() attempts to access the contents
of stderr. When stdio is set to 'ignore', this causes a crash.
This commit adds a check on the access of stderr. Closes#7966.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Documentation claimed it accepted a single label argument, as time and
timeEnd do, which was incorrect.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
This allows embedders enough control to initialize node, run the
event loop, and cleanly exit (including calling handlers).
Signed-off-by: Fedor Indutny <fedor@indutny.com>
A ReadableStream with a base64 StringDecoder backed by only
one or two bytes would fail to output its partial data before
ending. This fix adds a check to see if the `read` was triggered
by an internal `flow`, and if so, empties any remaining data.
fixes#7914.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Emits on every call to cluster.setupMaster(), even if no new settings
are given. This is because calling cluster.setupMaster() without
arguments (or with an empty options object) results in the settings
being restored to their defaults.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Only attributes of 'cluster.settings' will be modified after the first
call, leaving all other cluster initialization alone. Each call that
includes a 'settings' argument triggers a 'setup' event to be emitted.
Instead of each call resetting all values to their defaults, use the
current settings (if any) as the default. This retains setupMaster's
support how cluster.fork() uses setupMaster() to ensure
cluster.settings has been populated.
Update example in docs to use current node coding style and include
an example of progressive configuration.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Pass in the v8::Context, instead of creating it
within CreateEnvironment. This allows callers
to use a pre-existing context.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
The spawnSync() cwd option was being copied to the incorrect
location. This commit copies to the correct location.
Closes#7824
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Switch condition order to check for null before calling isNaN().
Also remove two unnecessary calls to isNaN() that are already
covered by calls to isFinite(). This commit targets v0.10, as
opposed to #7891, which targets master (suggested by
@bnoordhuis). Closes#7840.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Use a util.deprecate wrapper to issue warnings like any other
deprecated API. The option has been marked as deprecated in the docs
since v0.5.11.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Default highWaterMark is now set properly when using stream Duplex's
writableObjectMode and readableObjectMode options.
Added condition to the already existing split objectMode test to ensure
the highWaterMark is being set to the correct default value on both the
ReadableState and WritableState for readableObjectMode and
writableObjectMode.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Documentation for console.assert incorrectly described message as a
single message, but it is a format.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Fix for `error` events emitting only once when reconnecting
a single instance of net.Socket.
Fixesjoyent/node#7888
Signed-off-by: Fedor Indutny <fedor@indutny.com>
A udp packet can have 0 content. In that case nread will be equal to 0,
but addr != NULL.
Add test case for empty data gram packets and fixed test that checked
for OOB when length == 0.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
V8 seems to ignore the default value for want_separate_host_toolset and
would override it at build time. Instead always explicitly set the value.
Fixes#7833
The refactor in 3ae0b17c broke the multiline input's visual appearence.
While actually switching to this mode, the `...` prefix is not
displayed.
Additionally, account only SyntaxErrors that are happening at the parse
time, everything else should not be switching repl to the multiline
mode.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
The [Stream documentation for .push](http://nodejs.org/api/stream.html#stream_readable_push_chunk_encoding)
explicitly states multiple times that null is a special cased value
that indicates the end of a stream. It is confusing and undocumented
that undefined *also* ends the stream, even though in object mode
there is a distinct and important difference.
The docs for Object-Mode also explicitly mention null as the *only*
special cased value, making no mention of undefined.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Creating a new buffer from the toJSON() output of another
buffer does not currently work. This commit adds that
support. Closes#7849.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Calling dns.lookup with arguments that generate an error from c-ares
previously sent those errors back to the callback. This commit restores
the ca9eb71 behavior.
Fixes#7731.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Instrumentation code might need to find out the entry point of the
process in a global context.
Documenting the existing process.mainModule to officially support this.
Fixes#7808
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Currently child_process.exec() assumes that cmd.exe is on the PATH,
and fails with a spawn ENOENT error if it is not.
The Windows 'comspec' environment variable contains the full filepath
to the default command interpreter, eg "C:\Windows\System32\cmd.exe".
Should it not be set, we fall-back to using 'cmd.exe' from PATH, as
before.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Increase the performance and simplify the logic of Buffer#write{U}Int*
and Buffer#read{U}Int* methods by placing the byte manipulation code
directly inline.
Also improve the speed of buffer-write benchmarks by creating a new
call directly to each method by using Function() instead of calling by
buff[fn].
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Conflicts:
lib/buffer.js
It's possible to construct a typed array from a buffer but the buffer
is treated as an array, not a byte array as one might expect.
Fixes#7786.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Building MSIs for different arch's can sometimes confuse MSBuild and
Wix, isntead run the toolchain externally so we don't have to worry
about which arch cmd.exe is running as.
This features comes from the need of adding extra options when displaying
the object using console.dir().
console.dir() accepts now a second parameter that is passed to util.inspect()
in order to provide extra options to the output. These options are: depth, color
and showHidden. More information about these options in util.inspect() documentation.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
WriteItem callback may add new item to the `pending_write_items`. Ensure
that this item won't be called in the same `InvokeQueued` call, as it
may result in way-to-early `finish` event on js-side.
fix#7733
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Tests for the behaviour in v0.10.x which allows process.argv changes
to be honoured by cluster.setupMaster().
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
In v0.10.x, process.argv and process.execArgv would only be
evaluated and copied into cluster.settings on the first call to
cluster.setupMaster() (either directly or via cluster.fork()),
allowing them to be modified as needed before initializing the
settings.
In 41b75ca the behaviour was changed so that these values are
initialized at the time of the first require('cluster').
Fixes#7670.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
* openssl: to 1.0.1h (CVE-2014-0224)
* npm: upgrade to 1.4.10
* utf8: Prevent Node from sending invalid UTF-8 (Felix Geisendörfer)
- *NOTE* this introduces a breaking change, previously you could construct
invalid UTF-8 and invoke an error in a client that was expecting valid
UTF-8, now unmatched surrogate pairs are replaced with the unknown UTF-8
character. To restore the old functionality simply have NODE_INVALID_UTF8
environment variable set.
* child_process: do not set args before throwing (Greg Sabia Tucker)
* child_process: spawn() does not throw TypeError (Greg Sabia Tucker)
* constants: export O_NONBLOCK (Fedor Indutny)
* crypto: improve memory usage (Alexis Campailla)
* fs: close file if fstat() fails in readFile() (cjihrig)
* lib: name EventEmitter prototype methods (Ben Noordhuis)
* tls: fix performance issue (Alexis Campailla)
Previously we were only shifting the address space for ASLR on 32bit
processes, apply the same shift for 64bit so processes don't
get artificially limited native heap.
Previously v8's WriteUtf8 function would produce invalid utf-8 output
when encountering unmatched surrogate code units [1]. The new
REPLACE_INVALID_UTF8 option fixes that by replacing invalid code points
with the unicode replacement character.
[1]: JS Strings are defined as arrays of 16 bit unsigned integers. There
is no unicode enforcement, so one can easily end up with invalid unicode
code unit sequences inside a string.
- https://codereview.chromium.org/121173009/
- https://code.google.com/p/v8/source/detail?r=18683
Note: The v8 test case did not cleanly apply, so it's missing from this
patch. I'm assuming this is not a problem if the v8 test suite is not
part of the node build / test system. If that's the case I'll fix it.
Otherwise the test case will be integrated once v8 is upgraded.
This patch simplifies the implementation of StringDecoder, fixes the
failures from the new test cases, and also no longer relies on v8's
WriteUtf8 function to encode individual surrogates.
The test cases are still essentially the same, but now all possible ways
of writing a buffer into the decoder are tested, which has exposed a few
failing scenarios that had not been discovered so far!
The test is supposed to measure the performance of the base64 encoder
so move the Buffer#write() calls out of the benchmark section.
The overhead of the calls isn't terrible (about 1-3%) but having
unrelated activity in a micro-benchmark is never a good idea.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Replace the CONTAINER_OF macro with a template function that is as
type-safe as a reinterpret_cast<> of an arbitrary pointer can be made.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Currently, if fstat() fails in readFile(), the callback
is invoked without closing the file. This commit closes
the file before calling back.
Closes#7697
Currently, if fstat() fails in readFile(), the callback
is invoked without closing the file. This commit closes
the file before calling back.
Closes#7697
See https://github.com/joyent/node/issues/7675
net.server.listen() behaves inconsistently depending on whether the port
number is provided.
1. port === 0 && host == '' (i.e. false-y), node creates an AF_INET
socket but does not call bind().
2. port > 0 && host == '', node creates an AF_INET6 socket and calls
bind().
The fix makes 1 consistent with 2.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Rework the fix from commit 6810132 in a way that removes ~60 lines of
code.
The bug was introduced in commit e87ceb2 (mea culpa) and is at its core
a pointer aliasing bug where sometimes two independent pointers existed
that pointed to the same chunk of heap memory.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Otherwise it's not possible to check from inside a destructor if V8 is
still alive with v8::V8::IsDead(). In V8 3.25, that function returns
true until the last isolate is destroyed.
This used to work in v0.10 and is a standard trick to dispose persistent
handles conditionally.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
dispose() free's the memory when executed and sets the external array
data to NULL and length to zero.
To prevent the same memory from being free'd twice when the object is
garbage collected we first check if the object's external array data
length == 0. Since alloc() passes NULL to
SetIndexedPropertiesToExternalArrayData() if length == 0 there's no
opportunity for memory leak.
Fix up a bad assumption in pummel/test-net-pingpong, namely that binding
to 'localhost' or '' means that incoming connections will have an IPv4
address.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
EMFILE and ENFILE mean 'out of file descriptors'. It's a run-time error
and as such should emit an error on the child process object, not throw
an exception.
Fixes#7453.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Avoid sending unsent data and destroying otherwise legitimate sockets
for requests that are aborted while still in the agent queue. This
reduces stress on upstream systems who will likely respond to the
request but client app already knows that it will be dropped on the
floor and also helps avoid killing keep-alive connections.
See also commit e7bfbaf. Don't depend on deps/v8/build/features.gypi
to disable handle zapping, be explicit about it.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Slashes should be documented, because 3rd-party protocols -- those
postfixed with `://` -- would incorrectly `format` and `parse` if they
didn't set/get the `slashes` option.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Not all querystring are utf-8 encoding, make querystring can be used
to encode / decode `non-utf8` encoding string if necessary.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Do not ever call `Delete()` on `proxy_global_`, it will invoke
`GlobalPropertyDeleteCallback` and cause crash because of the infinite
recursion.
fix#7529
Adds a section to the transform stream docs to clarify the
difference between the `end` event and the `finish` events.
Also clarifies the wording on the `end` event.
64bit constants are keyed for x64 platforms only, add PowerPC based
platform constants.
Node's "ucs2" encoding wants LE character data stored in the Buffer, so
we need to reorder on BE platforms. See
http://nodejs.org/api/buffer.html regarding Node's "ucs2" encoding
specification
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Storing it in SSL_CTX is incorrect as it may go away and get destructed
earlier, also it'll yield invalid results in SelectSNIContextCallback.
Use `SSL_get_app_data()` instead.
fix#7484
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
The spawnsync test was written wrong, the timeout can never fire before
the sync process has returned, the delta is immaterial and times when
it was succeeding are not reliable cases.
Instead verify that the timeout doesn't fire while the sync process is
happening.
When close() is called on a non-listening server, a synchronous
error is thrown. This commit causes the error to be passed to
the asynchronous callback function instead.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
ClientHelloParser used to contain an 18k buffer that was kept around
for the life of the connection, even though it was not needed in many
situations. I changed it to be deallocated when it's determined to
be no longer needed.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Fix the following compiler warning on systems where _XOPEN_SOURCE is
defined by default:
../src/node_constants.cc:35:0: warning: "_XOPEN_SOURCE" redefined
#define _XOPEN_SOURCE 500
Move the (re)definition of _XOPEN_SOURCE to the top of the file while
we're here. Commit 00890e4 adds a `#define _XOPEN_SOURCE 500` in order
to make <fcntl.h> expose O_NONBLOCK but it does so after other system
headers have been included. If those headers include <fcntl.h>, then
the #include in node_constants.cc will be a no-op and O_NONBLOCK won't
be visible.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
stringToFlags() has fall throughs in a case statement.
However, they are not consistently implemented. This commit adds
consistency.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
There was an underlying assumption in readline.emitKeypressEvents (and
by extension emitKey) that the given stream (usually process.stdin)
would emit 'data' once per keypress, which is not always the case.
This commit buffers the input stream and ensures a 'keypress' event is
triggered for every keypress (including escape codes).
Signed-off-by: Fedor Indutny <fedor@indutny.com>
This test is still in test/disabled because it requires a tty, however
when executed directly this test now passes.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Ensure TypeError is thrown, fix a bug where `env` option was
assuming the option was actually an object.
This case is especially bad because it then sets `env == null`
instead of using `process.env`.
Fix#7456
Signed-off-by: Fedor Indutny <fedor@indutny.com>
A recent change to v8's API now makes it impossible to memcpy to a
v8::ArrayBuffer without causing it to be externalized. This means that
the garbage collector will not automatically free the memory when the
object is collected.
When/If the necessary API is included to allow the above
Buffer#toArrayBuffer() will be reintroduced.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Not removing 'end' listeners for input and output on the 'close' event
resulted in an EventEmitter related memory leak.
This issue also might be reproduced at:
https://github.com/npm/npm/issues/5203
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
When process._setupNextTick() was introduced as the means to properly
initialize the mechanism behind process.nextTick() a chunk of code was
left behind that assigned memory to process._tickInfo. This code is no
longer needed.
Before this commit the EventEmitter methods were anonymous functions.
V8 tries to infer names for anonymous functions based on the execution
context but it frequently gets it wrong and when that happens, the
stack trace is usually confusing and unhelpful. This commit names all
methods so V8 can fall back to the method.name property.
The above gotcha applies to all anonymous functions but is exacerbated
for EventEmitter methods because those are invoked with a plenitude of
different receivers.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
On windows you can see ECONNABORTED instead of ECONNRESET in various
scenarios, and they are both applicable we're testing that Node is not
swallowing these errors which it was known to do prior to 0.10
As a comment in the test states: "This test should not be ported to
v0.10 and higher, because the problem is fixed by not ignoring
ECONNRESET in the first place."
The test is checking whether write returns false instead of whether an
ECONNRESET has been raised.
Replace with test-http-destroyed-socket-write2, this test verifies that
ECONNRESET is raised when writing to an http request where the server
has destroyed the socket.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
* Don't set referer if already set
* fetch: Send referer and npm-session headers
* run-script: Support --parseable and --json
* list runnable scripts (Evan Lucas)
* Use marked instead of ronn for html docs
On Windows we cannot get the server address until a connection
is accepted.
From MSDN:
The getsockname function does not always return information about
the host address when the socket has been bound to an unspecified
address, unless the socket has been connected with connect or accept
(for example, using ADDR_ANY). A Windows Sockets application must not
assume that the address will be specified unless the socket is
connected.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
* Check SHA before using files from cache
* adduser: allow change of the saved password
* Make `npm install` respect `config.unicode`
* Fix lifecycle to pass `Infinity` for config env value
* Don't return 0 exit code on invalid command
* cache: Handle 404s and other HTTP errors as errors
* bump tap dep, make tests stderr a bit quieter
* Resolve ~ in path configs to env.HOME
* Include npm version in default user-agent conf
* npm init: Use ISC as default license, use save-prefix for deps
* Many test and doc fixes
Because of differences in memcmp() implementation, normalize output to
return -1, 0 or 1 only.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
This commit introduces `readableObjectMode` and
`writableObjectMode` options for Duplex streams.
This can be used mostly to make parsers and
serializers with Transform streams.
Also the docs section about stream state objects
is removed, because it is not relevant anymore.
The example from the section is remade to show
new options.
fixes#6284
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
When ExitCallback was not called with an error such as ENOENT in
uv_spawn, the process handle still remains refed and needs to be closed.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
1) ThrowCryptoTypeErrors was not actually used for
type-related errors. Removed it.
2) For AEAD modes, OpenSSL does not set any internal
error information if Final does not complete suc-
cessfully. Therefore, "TypeError:error:00000000:l
ib(0):func(0):reason(0)" would be the error mess-
age. Use a default message for these cases.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
compare() works like String.localeCompare such that:
Buffer.compare(a, b) === a.compare(b);
equals() does a native check to see if two buffers are equal.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Commit f9ced08 switches V8 on Linux over from gettimeofday() to
clock_getres() and clock_gettime(). As of glibc 2.17, those functions
live in libc. For older versions, we need to pull them in from librt.
Fixes the following link-time error;
Release/obj.target/deps/v8/tools/gyp/libv8_base.a(platform-posix.o):
In function `v8::internal::OS::Ticks()':
platform-posix.cc:(.text+0x93c):
undefined reference to `clock_gettime'
platform-posix.cc:(.text+0x989):
undefined reference to `clock_getres'
Fixes#7514.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Date.now() indirectly calls gettimeofday() on Linux and that's a system
call that is extremely expensive on virtualized systems when the host
operating system has to emulate access to the hardware clock.
Case in point: output from `perf record -c 10000 -e cycles:u -g -i`
for a benchmark/http_simple bytes/8 benchmark with a light load of
50 concurrent clients:
53.69% node node [.] v8::internal::OS::TimeCurrentMillis()
|
--- v8::internal::OS::TimeCurrentMillis()
|
|--99.77%-- v8::internal::Runtime_DateCurrentTime(v8::internal::Arguments, v8::internal::Isolate*)
| 0x23587880618e
That's right - over half of user time spent inside the V8 function that
calls gettimeofday().
Notably, nearly all system time gets attributed to acpi_pm_read(), the
kernel function that reads the ACPI power management timer:
32.49% node [kernel.kallsyms] [k] acpi_pm_read
|
--- acpi_pm_read
|
|--98.40%-- __getnstimeofday
| getnstimeofday
| |
| |--71.61%-- do_gettimeofday
| | sys_gettimeofday
| | system_call_fastpath
| | 0x7fffbbaf6dbc
| | |
| | |--98.72%-- v8::internal::OS::TimeCurrentMillis()
The cost of the gettimeofday() system call is normally measured in
nanoseconds but we were seeing 100 us averages and spikes >= 1000 us.
The numbers were so bad, my initial hunch was that the node process was
continuously getting rescheduled inside the system call...
v8::internal::OS::TimeCurrentMillis()'s most frequent caller is
v8::internal::Runtime_DateCurrentTime(), the V8 run-time function
that's behind Date.now(). The timeout handling logic in lib/http.js
and lib/net.js calls into lib/timers.js and that module will happily
call Date.now() hundreds or even thousands of times per second.
If you saw exports._unrefActive() show up in --prof output a lot,
now you know why.
That's why this commit makes V8 switch over to clock_gettime() on Linux.
In particular, it checks if CLOCK_REALTIME_COARSE is available and has
a resolution <= 1 ms because in that case the clock_gettime() call can
be fully serviced from the vDSO.
It speeds up the aforementioned benchmark by about 100% on the affected
systems and should go a long way toward addressing the latency issues
that StrongLoop customers have been reporting.
This patch will be upstreamed as a CR against V8 3.26. I'm sending it
as a pull request for v0.10 first because that's what our users are
running and because the delta between 3.26 and 3.14 is too big to
reasonably back-port the patch. I'll open a pull request for the
master branch once the CR lands upstream.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Forcibly flushes the request headers. You need this with long-lived
HTTP connections where the first data isn't written until the connection
has been established (think: tunneling requests over HTTP CONNECT.)
Fixes#7296.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
In tls.connect a unix socket connection to a path may be made in
recent versions of node by specifying the value for the path
property.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
OpenSSL behaves oddly: on client `cert_chain` contains
the `peer_certificate`, but on server it doesn't.
Signed-off-by: Fedor Indutny <fedor@indutny.com>
When creating a TLSSocket instance based on the existing connecting
socket, `_connecting` property is copied after the initialization of
`net.Socket`. However, since `net.Socket` constructor will call
`.read(0)` if the `readable` is true - error may happen at this code
chunk in net.js:
Socket.prototype._read = function(n) {
debug('_read');
if (this._connecting || !this._handle) {
debug('_read wait for connection');
this.once('connect', this._read.bind(this, n));
...
Leading to a test failures on windows:
- test/simple/test-tls-connect-given-socket.js
Signed-off-by: Fedor Indutny <fedor@indutny.com>
* isaacs, Robert Kowalski, Benjamin Coe: Test Improvements
* isaacs doc: Add canonical url
* isaacs view: handle unpublished packages properly
* Raynos (Jake Verbaten) do not log if silent
* Julian Gruber fix no such property error
* isaacs npmconf@0.1.14
* Thorsten Lorenz adding save-prefix configuration option
* isaacs npm-registry-client@0.4.7
* isaacs cache: treat missing versions as a 404
* isaacs cache: Save shasum, write resolved/etc data to cache
* isaacs cache: Always fetch root doc
* isaacs cache: don't repack unnecessarily from tmp
* Andrey Kislyuk Don't crash if shrinkwrap-dependencies were not passed in pkginfo
* Robert Kowalski fix link in faq
* Jean Lauliac Add a peerDependencies section in package.json doc
* isaacs read-installed@2.0.2
Oversight to not pass blksize to fs.Stats on initialization.
Also added a test to make sure the object property has been set. Since
now on Windows both blksize and blocks will simply be set to undefined.
Fix possible deadlock, when handles are sent in both direction
simultaneously. In such rare cases, both sides may queue their
`NODE_HANDLE_ACK` replies and wait for them.
fix#7465
Fix issue where a signed integer is returned.
Example:
var b = new Buffer(4);
b.writeUInt32BE(0xffffffff);
b.readUInt32BE(0) == -1
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
When our estimates for a storage size are higher than the actual length
of decoded data, the destination buffer should be truncated. Otherwise
`Buffer::Length` will give misleading information to C++ layer.
fix#7365
Signed-off-by: Fedor Indutny <fedor@indutny.com>
`process.uptime()` interface will return the amount of time the
current process has been running. To achieve this it was caching the
`uv_uptime` value at program start, and then on the call to
`process.uptime()` returning the delta between the two values.
`uv_uptime` is defined as the number of seconds the operating system
has been up since last boot. On sunos this interface uses `kstat`s
which can be a significantly expensive operation as it requires
exclusive access, but because of the design of `process.uptime()` node
*had* to always call this on start. As a result if you had many node
processes all starting at the same time you would suffer lock
contention as they all tried to read kstats.
Instead of using `uv_uptime` to achieve this, the libuv loop already
has a concept of current loop time in the form of `uv_now()` which is
in fact monotonically increasing, and already stored directly on the
loop. By using this value at start every platform performs at least
one fewer syscall during initialization.
Since the interface to `uv_uptime` is defined as seconds, in the call
to `process.uptime()` we now `uv_update_time` get our delta, divide by
1000 to get seconds, and then convert to an `Integer`. In 0.12 we can
move back to `Number::New` instead and not lose precision.
Caveat: For some platforms `uv_uptime` reports time monotonically
increasing regardless of system hibernation, `uv_now` interface is
also monotonically increasing but may not reflect time spent in
hibernation.
Introduce new signature for both `dgram.createSocket` method and
`dgram.Socket` constructor:
dgram.createSocket(options, [listener])
Options should contain `type` property and may contain `reuseAddr`
property. When `reuseAddr` is `true` - SO_REUSEADDR will be issued on
socket on bind.
fix#7415
Signed-off-by: Fedor Indutny <fedor@indutny.com>
Socket may become not `readable`, but http should not rely on this
property and should not think that it means that no data will ever
arrive from it. In fact, it may arrive in a next tick and, since
`this.push(null)` was already called, it will result in a error like
this:
Error: stream.push() after EOF
at readableAddChunk (_stream_readable.js:143:15)
at IncomingMessage.Readable.push (_stream_readable.js:123:10)
at HTTPParser.parserOnBody (_http_common.js:132:22)
at Socket.socketOnData (_http_client.js:277:20)
at Socket.EventEmitter.emit (events.js:101:17)
at Socket.Readable.read (_stream_readable.js:367:10)
at Socket.socketCloseListener (_http_client.js:196:10)
at Socket.EventEmitter.emit (events.js:123:20)
at TCP.close (net.js:479:12)
fix#6784
This prevents segfaults when a native method is reassigned to a
different object (which corrupts args.This()). When unwrapping,
clients should use args.Holder() instead of args.This().
Closes#6690.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Increase the performance and simplify the logic of Buffer#write{U}Int*
and Buffer#read{U}Int* methods by placing the byte manipulation code
directly inline.
Also improve the speed of buffer-write benchmarks by creating a new
call directly to each method by using Function() instead of calling by
buff[fn].
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
The two biggest changes are that v8::Script::New() has been removed and
that a v8::Script object now has to be explicitly bound to a context if
you want to run it from another context.
We can accommodate both changes without breaking the vm module's public
API or even the internal JS API.
The test/simple/test-smalloc.js has an implicit assumption
of the byte order of the data stored for Double and Uint32
values. On a big endian platform this test fails without
these patches.
Use os.endianness() to detect the endian of the platform
and use it to gate the static value used for comparison.
Improve on commit b55c9d6 by not requiring that switches are comma
separated. This commit makes `./configure --v8-options="--foo --bar"`
work and takes special care to properly escape quotes in the options
string.
By building the fs.Stats object in JS, which is returned by all fs stat
functions, calls to v8::Object::Set() are removed. This also includes
creating all associated Date objects in JS, rather than using
v8::Date::New(). Both these changes have significant performance gains.
Note that the returned value from fs.stat changes slightly for non-POSIX
systems. Whereas before the stats object would be missing blocks and
blksize keys, it now has these keys with undefined as the value.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Move `createCredentials` to `tls` module and rename it to
`createSecureContext`. Make it use default values from `tls` module:
`DEFAULT_CIPHERS` and `DEFAULT_ECDH_CURVE`.
fix#7249
Include the "expected protocol" in the Error message
string, which evaluates to "http:" for the `http`
core module, and "https:" for the `https` module.
Closes#7355.
These are an old and deprecated properties that was used by previous
stream implementation, and are still in use in some user-land modules.
Prior to this commit, they were read from the underlying socket, which
may be non-readable/non-writable while connecting or while staying
uninitialized.
Force set them to `true`, just to make sure that there will be no
inconsistency.
fix#7152
Previously the build artifacts did not include a signed timestamp, so
when the certificate expired the validation of the artifact would fail.
Now we sign against a timestamp server such that the artifact will
always be valid regardless of the disposition of the certificate.
Closes#7360 and #7059.
Ensure that OpenSSL has enough entropy (at least 256 bits) for its PRNG.
The entropy pool starts out empty and needs to fill up before the PRNG
can be used securely.
OpenSSL normally fills the pool automatically but not when someone
starts generating random numbers before the pool is full: in that case
OpenSSL keeps lowering the entropy estimate to thwart attackers trying
to guess the initial state of the PRNG.
When that happens, we wait until enough entropy is available, something
that normally should never take longer than a few milliseconds.
Fixes#7338.
The default entropy source is /dev/urandom on UNIX platforms, which is
okay but we can do better by seeding it from OpenSSL's entropy pool.
On Windows we can certainly do better; on that platform, V8 seeds the
random number generator using only the current system time.
Fixes#6250.
NB: This is a back-port of commit 7ac2391 from the master branch that
for some reason never got back-ported to the v0.10 branch.
The default on UNIX platforms in v0.10 is different and arguably worse
than it is with master: if no entropy source is provided, V8 3.14 calls
srandom() with a xor of the PID and the current time in microseconds.
That means that on systems with a coarse system clock, the initial
state of the PRNG may be easily guessable.
The situation on Windows is even more dire because there the PRNG is
seeded with only the current time... in milliseconds.
* Documentation upgrades
* Fix glob bug which prevents proper README publishing
* node-gyp upgrade to 0.13
* Documentation updates
* Add --save-exact to save an exact dep (instead of a range)
* alias 't' to 'test'
The `Agent#request()` function was removed in
f3189ace6b, so don't
use it in the documentation example. The function
wasn't documented in the first place.
Default to the `defaultAgent.protocol` when comparing the
user-specified `options.protocol` string. This is so that
`http.Agent` instances do not need to specify their own
`protocol` field, since we have the relevant information
already from the `defaultAgent`.
Note that the test case could be separately cherry-picked
to the `v0.10` branch, since it already passes correctly.
Fixes#7349.
Fixes the regression described in: http://git.io/2ds-WQ
Turn off -Werror when building V8, it hits -Werror=unused-local-typedefs
with g++ 4.8. The warning itself is harmless so don't abort the build.
This was originally implemented in commit d2ab314e back in 2011 but the
build process has gone through a few iterations since then, that change
no longer works.
`env.h` is an internal header file and should not be copied or exposed
to the users.
Additionally, export convenience `Throw*` methods with `v8::Isolate*` as
a first argument.
Fix up the dtrace/etw/systemtap infrastructure after the V8 upgrade in
commit 1c7bf24. The win32 changes are untested but can hardly make
things worse because node doesn't build on windows right now.
Fixes#7313 with some luck.
Don't call DecodeWrite() with a Buffer as its argument because it in
turn calls StringBytes::Write() and that method expects a Local<String>.
"Why then does that function take a Local<Value>?" I hear you ask.
Good question but I don't have the answer. I added a CHECK for good
measure and what do you know, all of a sudden a large number of crypto
tests started failing.
Calling DecodeWrite(BINARY) on a buffer is nonsensical anyway: if you
want the contents of the buffer, just copy out the data, there is no
need to decode it - and that's exactly what this commit does.
Fixes a great many instances of the following run-time error in debug
builds:
FATAL ERROR: v8::String::Cast() Could not convert to string
Fix a regression that was introduced in commit ce04c726 after the
upgrade to V8 3.24.
The new weak persistent handle API no longer gives you the original
persistent but still requires that you clear it inside your weak
callback.
Rearrange the code in src/smalloc.cc to keep track of the persistent
handle with the least amount of pain and try hard to share as much
code as possible between the 'just free it' and 'invoke my callback'
versions of the smalloc API.
Fixes#7309.
Conform to the Google styleguide more and make cpplint happy, add more
CHECK macros.
Preemptively addresses cpplint's readability/check warnings ("Consider
using CHECK_GT instead of CHECK(a > b)".)
Make calls to v8::Isolate::AdjustAmountOfExternalAllocatedMemory() take
special care when negating 32 bits unsigned types like size_t.
Before this commit, values were negated before they got promoted to
64 bits, meaning that on 32 bits architectures, a value like 42 got
cast to 4294967254 instead of -42.
That in turn made the garbage collector start scavenging like crazy
because it thought the system was out of memory.
That's bad enough but calls to AdjustAmountOfExternalAllocatedMemory()
were made from weak callbacks, i.e. at a time when the garbage collector
was already busy. It triggered asserts in debug builds and caused
random crashes and memory corruption in release builds.
The behavior in release builds is arguably a V8 bug and should perhaps
be reported upstream.
Partially fixes#7309 but requires further bug fixes to src/smalloc.cc
that I'll address in a follow-up commit.
The variable isn't actually used uninitialized but g++ 4.8 doesn't know
that. Set it to NULL to silence the following compiler warning:
../src/string_bytes.cc:247:29: warning: 'data' may be used
uninitialized in this function [-Wmaybe-uninitialized]
unsigned a = hex2bin(src[i * 2 + 0]);
^
../src/string_bytes.cc:299:15: note: 'data' was declared here
const char* data;
^
V8 was upgraded from 3.22 to 3.24 in commit 1c7bf24. Upgrade source
files in test/addons/ and automatically generated tests from
doc/api/addons.markdown to the new V8 API.
This coincidentally fixes a bug in src/node_object_wrap.h where it was
still using the old V8 weak persistent handle interface, which is gone
in 3.24.
* ::jsstack -v prints function defintion
* ::jsprint works with objects with only numeric properties
* update tests to use builtin mdb_v8
* add more symbols to postmortem script - pending upstream
inclusion
Previously if you wanted to be notified of pending handles for pipes
you needed to use uv_read2_start, however in v0.11.22 you can query for
pending handles independently.
Internally we use hrtime to schedule when a timer will fire, to avoid
the perils of clock drift or other external operation making time go
backward. The timers ordering test should use the same timing mechanism
Fix the following valgrind warning:
Conditional jump or move depends on uninitialised value(s)
at 0x7D64E7: v8::internal::GlobalHandles::IterateAllRootsWithClassIds(v8::internal::ObjectVisitor*) (global-handles.cc:613)
by 0x94DCDC: v8::internal::NativeObjectsExplorer::FillRetainedObjects() (profile-generator.cc:2849)
# etc.
This was fixed upstream in r12903 and released in 3.15.2 but that commit
was never back-ported to the 3.14 branch that node.js v0.10 uses.
The code itself works okay; this commit simply shuffles the clauses in
an `if` statement to check that the node is in use before checking its
class id (which is uninitialized if the node is not in use.)
When sending a socket to a child process via IPC pipe,
`child_process.js` picks a raw UV handle from `_handle` property, sends
it, and assigns `null` to the property. Sending the same socket twice
was resulting in a runtime error, since we weren't handling the empty
`_handle` case.
In case of `null` `_handle` we should send just a plain text message
as passed it was passed to `.send()` and ignore the handle, letting
users handle such cases themselves instead of throwing the error at
runtime.
fix#5469
It's currently not really possible to compile native add-ons with
-fvisibility=hidden because that also hides the struct containing
the module definition.
The NODE_MODULE() and NODE_MODULE_DECL() macros are structured in
a way that makes it impossible to add a visibility attribute manually
so there is no escape hatch there.
That's why this commit adds an explicit visibility attribute to
the module definition. It doesn't help with node.js releases that
are already out there but at least it improves the situation going
forward.
If two timers run on the same tick, and the first timer uses a domain,
and then catches an exception and disposes of the domain, then the
second timer never runs. (And even if the first timer does not dispose
of the domain, the second timer could run under the wrong domain.)
This happens because timer.js uses "process.nextTick()" to schedule
continued processing of the timers for that tick. However, there was
an exception inside a domain, then "process.nextTick()" runs under
the domain of the first timer function, and will do nothing if
the domain has been disposed.
To avoid this, we temporarily save the value of "process.domain"
before calling nextTick so that it does not run inside any domain.
Previously if you cached process.nextTick and then require('domain')
subsequent nextTick() calls would not be caught because enqueued
functions were taking the wrong path. This keeps nextTick to a single
function reference and changes the implementation details after domain
has been required.
When `setImmediate(cb)` is called in `beforeExit` event handler the
consequent `uv_run(..., UV_RUN_NOWAIT)` may return `0`, even if there
was some active handles at start.
Fixes simple/test-beforeexit-event.js.
Unlike the 'exit' event, this event allows the user to schedule more
work and thereby postpone the exit. That also means that the
'beforeExit' event may be emitted many times, see the attached test
case for an example.
Refs #6305.
libuv gyp builds now require you to define the library disposition
(static or shared).
Also, libuv now supports vectored IO for file system reads and writes,
update to those function signatures
Between `ClientRequest` and `Agent`. The circular require was doing
weird things at load time, like making the `globalAgent` property
be `undefined` from within the context of the "_http_client"
module.
Removing the circular dependency completely fixes this.
This commit effectively removes the undocumented `Agent#request()`
and `Agent#get()` functions.
Don't invoke the `agent.requst()` or `agent.get()` functions
directly. Instead, use the public API and pass the agent
instance in as the `agent` option.
For the `request()` and `get()` functions. I could never
really understand why these two functions go through agent
first... Especially since the user could be passing `agent: false`
or a different Agent instance completely, in which `globalAgent`
will be completely bypassed.
Moved the relevant logic from `Agent#request()` into the
`ClientRequest` constructor.
Incidentally, this commit fixes#7012 (which was the original
intent of this commit).
This makes it so that the user may pass in a
`createConnection()` option, and they don't have
to pass `agent: false` at the same time.
Also adding a test for the `createConnection` option,
since none was in place before.
See #7014.
Expose `setBlocking` on Pipe's and if a pipe is being created for stdio
on windows then make the pipes blocking.
This fixes test-stream2-stderr-sync.js on Windows.
Fixes#3584
One test case in test-stream2-stderr-sync.js was creating a TTY
object using an undocumented constructor and passing in fd 2.
However, this is running in a child process and fd 2 is actually
a pipe, not a TTY.
The constructor fails on Windows and causes the handle type to be
left uninitialized, which later causes an assert to fail.
On Unix, the constructor fails to retrieve the windows size but unlike
on Windows, it just leaves the size fields undefined and continues
with initializing the stream type, yielding a semi-usable object.
I could make the Windows version match Unix behavior, but it
seems to me that the test is relying on an implementation detail of
an undocumented API, and the Unix behavior is not necessarily more
correct than the Windows one. Thus it makes more sense to remove this
test.
NodeBIO::Gets was reading off the end of a buffer if it
didn't find a "\n" before the EOF. This behavior
was causing X509 certificates passed to `https.Agent`
via the "ca" option to be silently discarded. It also
was causing improper parsing of certs and keys
passed to https.Agent, but those problems were worked
around in cdde9a3.
Backed out workaround in `lib/crypto.js` from ccde9a3,
which now isn't needed. But keep the test introduced
in that commit, which tests properly for this
bug.
This bug was first introduced in a58f93f
Gist containing test code, bisection log, and notes:
https://gist.github.com/maxtaco/9211605
The linker was optimizing the static variables that were supposed
to trigger module initialization.
I am making them non-static, and dllexport so that they don't get
optimized away.
Fixes#7116
Ensure that the behavior of `assert.deepEqual` does not depend on
argument ordering when comparing an `arguments` object with a
non-`arguments` object.
This test is particularly pathological, and requires a ton of time to
run, we need to find a better way to manage it but in general this path
is fairly safe these days.
bufferSize is now a getter that shows all that has not been
acknowledged by the os, as well as in the buffer state. The test is
only looking to verify the js verified state.
Scheduling of next ticks from within the next tick handler will result
in a tight execution loop where a timer cannot break into.
This test was invalid
We were being very aggressive in our connection creations, resulting
in the pipeline flood detection to drop us. Relax how fast we're
creating these connections so the gc can run all its tests.
It wasn't doing anything, and actually due to
3ae0b17c76, it was causing
the readline `prompt()` function to be overwritten
which throws an error in the REPL shortly after.
Expose localPort for binding to a specific port for outbound
connections.
If localAddress is not specified '0.0.0.0' is used for ip4 and '::'
for ip6 connections.
Fixes#7092
The reason this wasn't working was because after restart, when restoring
breakpoints the scripts wasn't loaded, so the breakpoint.script was
undefined. As a fix I added another check to use breakpoint.scriptReq
instead of breakpoint.script, which is the same except when the
breakpoint is a function.
fixes#7027
In some scenarios this will strip the DOF sections for DTrace, and in a
future world where we re-export all static libraries it would defeat
that purpose.
If an input stream would emit `end` event, like
`fs.createReadStream`, then readline need to get the last line
correctly even though that line isnt ended with `\n`.
Given the assert message, and the fact that endCb is always true
in the assert, I am pretty sure the test author was intending
to test for finishEvent, not endCb.
In this test, an HTTP server was ending the response before
consuming all the data sent in the PUT request.
Ending the response would cause the socket to be destroyed,
and since there is some data still to be read, an ECONNRESET is
surfaced on the client side, event though the client has already
ended its side and even seen a 'finish' event.
See:
http://www.w3.org/Protocols/rfc2616/rfc2616-sec8.html#sec8.2.2
While it is certainly admissible for the server to send a response
before consuming the entire request, it seems reasonable to
expect that the server would close the connection afterwards
and that the ECONNRESET would be raised on the client.
So I have changed the test to wait until the entire request has been
consumed before sending the response.
Before sending a socket from a cluster master to a worker,
we would call listen in UV but not handle the error.
I made createServerHandle call listen on Windows so we get a chance
the handle any bind/listen errors early.
This fix is 100% windows specific.
It fixes test-cluster-bind-twice and
test-cluster-shared-handle-bind-error on Windows.
This implements the user-facing APIs that lets one run a child process
and block until it exits.
Logic shared with the async counterpart of each function was refactored
to enable code reuse.
Docs and tests are included.
Don't use argument as callback if it's not a valid callback function.
Throw a valid exception instead explaining the issue.
Adds to #7070 ("DNS — Throw meaningful error(s)").
The number of connections achieved by the test can vary by platform
and by machine. Lowering the acceptance threshold so that the
test passes on Windows.
Don't use argument as callback if it's not a valid callback function.
Throw a valid exception instead explaining the issue. Adds to #7070
("DNS — Throw meaningful error(s)").
Make vm.runInContext() and vm.runInNewContext() stop copying the Proxy
object from the parent context into the new context when --harmony or
--harmony_proxies is in effect because it overwrites the new context's
native Proxy object.
This commit also adds a regression test for Harmony symbols. They work
okay in the current implementation and the test should ensure it stays
that way.
Conditional globals like 'gc' should only be recognized when --expose_gc
is set. The global.gc feature check works only when done eagerly, else
it lets through a leaked variable called 'gc'.
Before, `new String('foo')` would be inspected as `"{}"` which
is simply not very helpful. Now, a more meaningful
`"[String: 'foo']"` result will be returned from `util.inspect()`.
Boxed String, Boolean, and Number types are all supported.
Closes#7047
Try embedding the ` ... ^` lines inside the `SyntaxError` (or any other
native error) object before giving up and printing them to the stderr.
fix#6920fix#1310
The AsyncListener API has been moved into the "tracing" module in order
to keep the process object free from unnecessary clutter.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Now the second field in asyncFlags will tell if the provider is
currently being watched, or listened for.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
These will be used to allow users to filter for which types of calls
they wish their callbacks to run.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
"flags" could mean one of many things, and multiple flag types could be
checked. So make the field more explicit on what type of flags are being
stored.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
Add a new 'tracing' module with a v8 property that lets the user
register listeners for gc events. The listeners are invoked after
every garbage collection cycle with 'before' and 'after' statistics.
Useful for monitoring tools that want to keep track of memory usage.
Create a new HandleScope before looking up the object context with
v8::Object::CreationContext(), else we leak the Local<Context> into
the current HandleScope.
That's relatively harmless unless the HandleScope is long-lived and
MakeCallback() is called a lot. In a scenario like that, we may end
up leaking a lot of memory.
What is unfortunate about this change is that we're trying hard to
eradicate the node_isolate global. Longer term, we will probably have
to change the MakeCallback() prototype to one that requires an explicit
v8::Isolate* argument.
Make it possible to invoke MakeCallback() on a v8::Value but only for
the variant that takes a v8::Function as the thing to call.
The const char* and v8::String variants still require a v8::Object
because the function to call is looked up as a property on the receiver,
but that only works when the receiver is an object, not a primitive.
If the call to writeBuffer completes asynchronously, we need to have
an oncomplete callback on the request object no matter what. The
writeQueueSize seems irrelvant to that regard.
Note that on Windows writeBuffer always completes asynchronously.
See related commit 9836a4eeda
Update the list of root certificates in src/node_root_certs.h with
tools/mk-ca-bundle.pl and update src/node_crypto.cc to make use of
the new format.
Fixes#6013.
`tls_wrap.cc` was crashing in an `Unwrap` call, when non
`SecureContext` object was passed to it. Check that the passed object
is a `SecureContext` instance before unwrapping it.
fix#7008
Original commit message:
VS2013 contains a number of improvements, most notably the addition
of all C99 math functions.
I'm a little bit concerned about the change I had to make in
cpu-profiler.cc, but I spent quite a bit of time looking at it and was
unable to figure out any rational explanation for the warning. It's
possible it's spurious. Since it seems like a useful warning in
general though, I chose not to disable globally at the gyp level.
I do think someone with expertise here should probably try to
determine if this is a legitimate warning.
BUG=288948
R=dslomov@chromium.org
Review URL: https://codereview.chromium.org/23449035
NOTE: Path applied without `cpu-profiler.cc` changes because in our
version it was looking totally different.
The test is no longer valid for the original scenario.
It now fails intermittently because of two other issues:
1. Since the client is only processing one readable event, the
client request is not enough to keep the process alive and the
process can exit before the desired events have been raised.
2. Reading just 1 byte is not enough to guarantee that the parser
will eventually consume all the data and raise the desired
parse error. I tried postponing the server.close() to address
the issue at [1], but then the test just hangs sometimes.
Even if stdio streams are opened as file streams, we should not ever try
to close them. This could be accomplished by passing `autoClose: false`
in options on their creation.
Even if stdio streams are opened as file streams, we should not ever try
to close them. This could be accomplished by passing `autoClose: false`
in options on their creation.
This was originally introduced in 6034701 to prevent the closing
brace being pushed onto the next line if an object is longer than
the max width, however the functionality was removed in d164989 but
the supplementary variables (and operations) were left behind
This matches how libuv handles the definition of ssize_t, by
typedef'ing intptr_t to ssize_t.
However, in the future we will use portable types from stddef.h
It's saner to check exit codes or signals to determine if the process
actually aborted. On OSX and Linux the exit code is 134, on SunOS it
propagates the SIGABRT signal
Built-in modules should be automatically registered, replacing the
static module list. Add-on modules should also be automatically
registered via DSO constructors. This improves flexibility in adding
built-in modules and is also a prerequisite to pure-C addon modules.
Right now no default ciphers are use in, e.g. https.get, meaning that
weak export ciphers like TLS_RSA_EXPORT_WITH_DES40_CBC_SHA are
accepted.
To reproduce:
node -e "require('https').get({hostname: 'www.howsmyssl.com', \
path: '/a/check'}, function(res) {res.on('data', \
function(d) {process.stdout.write(d)})})"
The test was not waiting for all the worker-created sockets
to be listening before calling cluster.disconnect().
As a result, the channels with the workers could get closed
before all the socket handles had been passed to them, leading
to various errors.
Original commit message:
ares_parse_txt_reply: return a ares_txt_reply node for each sub-string
Previously, the function would wrongly return all substrings merged into
one.
fix#6931
Socket may become not `readable`, but http should not rely on this
property and should not think that it means that no data will ever
arrive from it. In fact, it may arrive in a next tick and, since
`this.push(null)` was already called, it will result in a error like
this:
Error: stream.push() after EOF
at readableAddChunk (_stream_readable.js:143:15)
at IncomingMessage.Readable.push (_stream_readable.js:123:10)
at HTTPParser.parserOnBody (_http_common.js:132:22)
at Socket.socketOnData (_http_client.js:277:20)
at Socket.EventEmitter.emit (events.js:101:17)
at Socket.Readable.read (_stream_readable.js:367:10)
at Socket.socketCloseListener (_http_client.js:196:10)
at Socket.EventEmitter.emit (events.js:123:20)
at TCP.close (net.js:479:12)
fix#6784
The test was calling server.close() after write on the socket
had completed. However the fact that the write had completed was
not valid indication that the server had received the data.
This would result in a premutaure closing of the server and
an ECONNRESET event on the client.
When creating TLSSocket on top of the regular socket that already
contains some received data, `_tls_wrap.js` should try to write all that
data to the internal `SSL*` instance.
fix#6940
* uv: Upgrade to v0.10.23
* npm: Upgrade to v1.3.24
* v8: Fix enumeration for objects with lots of properties
* child_process: fix spawn() optional arguments (Sam Roberts)
* cluster: report more errors to workers (Fedor Indutny)
* domains: exit() only affects active domains (Ryan Graham)
* src: OnFatalError handler must abort() (Timothy J Fontaine)
* stream: writes may return false but forget to emit drain (Yang Tianyang)
When the domain specific code was reintroduced in 828f145 the
conditional to check and clear the nextTickQueue if many items had run
was not introduced. This allows for the application to run out of memory
if domains are being used in an infinite recursive loop.
ERR_load_crypto_strings() registers the error strings for
all libcrypto functions, SSL_load_error_strings() does the
same, but also registers the libssl error strings.
Make the HMAC digest method configurable. Update crypto.pbkdf2() and
crypto.pbkdf2Sync() to take an extra, optional digest argument.
Before this commit, SHA-1 (admittedly the most common method) was used
exclusively.
Fixes#6553.
Now that the context stores the active execution stack, and because
removeAsyncListener() always removed the AsyncListener from the queue
and the stack, there's no need to keep a stack around anymore. Instead
the active asyncQueue and the currentContext is able to handle it all.
Signed-off-by: Forrest L Norvell <ogd@aoaioxxysz.net>
Performance gains are ~4x (~1.5us), but still much slower than a naive
approach. There is some duplicate work done between join(), normalize()
and normalizeArray() so additional optimizations are possible.
Note that this only improves the POSIX implementation.
Thanks to @isaacs and @othiym23 for helping with this optimization.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
After one of OpenSSL updates we have stopped accepting PEM private keys
and certificates that doesn't end with a newline (`\n`) character.
Handle this regression in `crypto.js` to make less trouble to our users.
fix#6892
Do not throw in internal C++ methods, that clobbers logic and may lead
to the situations, where both exception was thrown and the value was
returned (via `args.GetReturnValue().Set()`). That doesn't play nicely
with v8.
fix#6912
Now that process.createAsyncListener() returns a unique object instance
it is no longer necessary to compare the uid's of the objects.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
It was possible that the same AL instance was run twice if it were both
attached to the currentContext then again added to the new asyncQueue
generated for the new stack.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
The ability to add/remove an AsyncListener to an object after its
creation was an artifact of trying to get AL working with the domain
module. Now that is no longer necessary and other features are going to
be implemented that would be affected by this functionality. So the code
will be removed for now to simplify the implementation process.
In the future this code will likely be reintroduced, but after some
other more important matters have been addressed.
None of this functionality was documented, as is was meant specifically
for domain specific implementation work arounds.
Signed-off-by: Timothy J Fontaine <tjfontaine@gmail.com>
All C++ code should be using `args.GetIsolate()` or `env->isolate()`.
Using static `node_isolate` var limits possible future functionality
(like multi-isolate support).
This test was originally intended to guard against regressions for
commit 16b59cbc74.
As such, it only needs to ensure that process exit has not been held up
by the date cache timer, which would fire on the next second.
We now wait to connect to the debuggee until we know that
its error stream has data, to ensure that the output message
"connecting..... ok" appears after "Debugger listening on port xyz"
I also increased the test timeout to let the more complex
tests finish in time on Windows
This change fixes the following unit tests on Windows:
test-debugger-repl.js
test-debugger-repl-term.js
test-debugger-repl-utf8.js
test-debugger-repl-restart.js
addon_register_func and its cousin addon_context_register_func are type
definitions, dllimport and dllexport are name mangling directives, i.e.
they're quite unrelated concepts. MinGW complains about mixing them
when cross-compiling native add-ons.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Signed-off-by: Bert Belder <bertbelder@gmail.com>
Before this commit, verification exceptions had err.message set to the
OpenSSL error code (e.g. 'UNABLE_TO_VERIFY_LEAF_SIGNATURE').
This commit moves the error code to err.code and replaces err.message
with a human-readable error. Example:
// before
{
message: 'UNABLE_TO_VERIFY_LEAF_SIGNATURE'
}
// after
{
code: 'UNABLE_TO_VERIFY_LEAF_SIGNATURE',
message: 'unable to verify the first certificate'
}
UNABLE_TO_VERIFY_LEAF_SIGNATURE is a good example of why you want this:
the error code suggests that it's the last certificate that fails to
validate while it's actually the first certificate in the chain.
Going by the number of mailing list posts and StackOverflow questions,
it's a source of confusion to many people.
Spawn's arguments were documented to be optional, as they are for the
other similar child_process APIs, but the code was missing. Result was
`child_process.spawn('node', {})` errored when calling slice() on an
Object, now it behaves as the documentation said it would.
domain.create().exit() should not clear the domain stack if the domain
instance does not exist within the stack.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
spawn stdio options can be a 'stream', but the following code
fails with "Incorrect value for stdio stream: [object Object]",
despite being a stream. The problem is the test isn't really
for a stream, its for an object with a numeric `.fd` property,
and streams do not have an fd until their async 'open' event
has occurred. This is reasonable, but was not documented.
child_process.spawn('date', [], {stdio: [
'ignore',
fs.createWriteStream('out.txt',{flags:'a'}),
'ignore']})
The RR cluster scheduler replaces the normal StreamWrap handle. Because
of this the AsyncListener method failed to be in place when domains were
in use.
The issue was resolved in 828f145 by reverting having domains use
AsyncListeners.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
Before when an AsyncListener object was created and the "create"
callback returned a value, it was necessary to construct a new Object
with the same callbacks but add a place for the new storage value.
Now, instead, a separate storage array is kept on the context which is
used for any return value of the "create" callback. This significantly
reduces the number of Objects that need to be created.
Also added a flags property to the context to quickly check if a
specific callback was available either on the context or on the
AsyncListener instance itself.
Few other minor changes for readability that were difficult to separate
into their own commit.
This has not been optimized yet.
This is a slightly modified revert of bc39bdd.
Getting domains to use AsyncListeners became too much of a challenge
with many edge cases. While this is still a goal, it will have to be
deferred for now until more test coverage can be provided.
Forcibly disable -Werror, the old { 'werror': '' } hack in node.gyp
no longer works with newer versions of V8.
We support a wide range of compilers, it's simply not feasible to
squelch all warnings, never mind that the libraries in deps/ are
not under our control.
Fixes#6817.
If a write is above the highWaterMark, _write still manages to
fully send it synchronously, _writableState.length will be adjusted down
to 0 synchronously with the write returning false, but 'drain' will
not be emitted until process.nextTick.
If another small write which is below highWaterMark is issued before
process.nextTick happens, _writableState.needDrain will be reset to false,
and the drain event will never be fired.
So we should check needDrain before setting it up, which prevents it
from inproperly resetting to false.
Instead of checking the uid on the array index of the queue, instead the
object property "uid" was checked on the queue iteself. Because this
will always evaluate to "undefined" the same listener could be added
multiple times to the same context.
There was a flaw in the old API that has been fixed. Now the
asyncListener callback is now the "create" object property in the
callback object, and is optional.
The fact that the "exit" event passes the exit code as an argument
as omitted from the documentation. This adds the explanation and
augments the example code to show that.
Master was disconnecting its workers as soon as they both started up.
Meanwhile, the workers were trying to listen. Its a race, sometimes the
disconnect would happen between when worker gets the response message,
and acks that message with a 'listening'. This worked OK after v0.11
introduced a behaviour where disconnect would always exit the worker,
but once that backwards-incompatible behaviour is removed, the worker
lives long enough to try and respond to the master, and child_process
errors at the attempt to send from a disconnected child.
This is a problem present in both v0.10, and v0.11, where the 'setup'
event is synchronously emitted by `cluster.setupMaster()`, a mostly
harmless anti-pattern.
Fix inadvertent v0.11 changes to the definition of suicide, particularly
the relationship between suicide state, the disconnect event, and when
exit should occur.
In v0.10, workers don't forcibly exit on disconnect, it doesn't give
them time to do a graceful finish of open client connections, they exit
under normal node rules - when there is nothing left to do. But on
unexpected disconnect they do exit so the workers aren't left around
after the master.
Note that a test as-written was invalid, it failed against the v0.10
cluster API, demonstrating that it was an undocumented API change.
Fixes issue in 0.11 where callback doesn't occur if worker count is
currently zero. In 0.10 callback occurs after worker count is zero, and
occurs in next tick if worker count is currently zero.
QueryString.stringify() allowed a fourth argument that was used as a
conditional in the return value, but was undocumented, not used by core
and always was always false/undefiend. So the argument and conditional
have been removed.
Signed-off-by: Trevor Norris <trev.norris@gmail.com>
After landing 6ed861d it is no longer possible to reliably monkey-patch
the EventEmitter constructor. However there's valid use cases for that,
and makes for easier debugging. Therefore, move the guts of the
constructor to a separate function which is monkey-patchable.
Closes#6693
The UNIX domain is also known as the LOCAL domain (AF_LOCAL), and
node/libuv implements it on Windows using named pipes. The API
documentation did not describe the naming rules for named pipes, and
also repeatedly described `listen(path)` as being UNIX, which it is not
on Windows.
Closes#6743
Removing a hack intended to shortcut the resolution of 'localhost'
but which doesn't work for ipv6.
This was introduced in 2876141c42.
However it seems that the problems that this was trying to
circumvent has gone away ages ago, when dns resolution on
Windows started relying on Win32 GetAddrInfoW, which was
probably with be2320d408.
Fixes test-net-connect-options-ipv6.js on Windows.
The %p is replaced with the current PID. This used to work in node.js
v0.9.7 but it seems to have been lost somewhere along the way.
This commit makes the fix from 6b713b52 ("cluster: make --prof work for
workers") work again. Without it, all log data ends up in a single
file and is unusable because the addresses are all wrong.
Eliminate a race condition between uv_async_send and the closing of the
corresponding handle.
Also made errors in Watchdog constructor call abort()
Fixes#6088
The 1.3.19 release had a critical bug: any packages published with it
could not be installed, because the shasum would be incorrect.
Thankfully, 1.3.19 was published using 1.3.19, so could not be installed
by any users! However, if it goes out as part of a Node.js release,
then obviously that would be a problem.
This is a comment change, where it originally says disabling TLS
Compression protects against BEAST attack. But in fact, it is the
CRIME attack(Compression Ratio Info-leak Made Easy) that makes use
of TLS Compression and not BEAST.
BEAST(Browser Exploit Against SSL/TLS) is an entirely another variant
making use of the chosen boundary attack against CBC mode in
encryption.
Just making sure, that the exact reason for disabling TLS compression
must be made clear and not be misleading with some other attack.
Quoting CVE-2013-6639:
The DehoistArrayIndex function in hydrogen-dehoist.cc in Google V8
before 3.22.24.7, as used in Google Chrome before 31.0.1650.63,
allows remote attackers to cause a denial of service (out-of-bounds
write) or possibly have unspecified other impact via JavaScript code
that sets the value of an array element with a crafted index.
Quoting CVE-2013-6640:
The DehoistArrayIndex function in hydrogen-dehoist.cc in Google V8
before 3.22.24.7, as used in Google Chrome before 31.0.1650.63,
allows remote attackers to cause a denial of service (out-of-bounds
read) via JavaScript code that sets a variable to the value of an
array element with a crafted index.
Like 6b92a7, this is unlikely to affect node.js because it only runs
local, trusted code. However, if there exists some module somewhere
that populates an array index with remotely provided data this could
very well be used to crash a remote server running node. Defense in
depth and all.
This is a backport of upstream commit r17801. Original commit log:
Limit size of dehoistable array indices
LOG=Y
BUG=chromium:319835,chromium:319860
R=dslomov@chromium.org
Review URL: https://codereview.chromium.org/74113002
The test is expecting an invalid result for the loopback
interface network mask, but this issue was fixed in
libuv commit 1d5c61a8b31257733c41fb507762d3eb56eecb2d
Closes#5262#6673
When `symlink`, `link` or `rename` report EEXIST, ENOTEMPTY or EPERM -
the destination file name should be included in the error message,
instead of source file name.
fix#6510
NOTE: Also removed `.receivedShutdown` method of `Connection` it wasn't
documented anywhere, and was rewritten with `true` after receiving
`close_notify`.
fix#6638
FSEventStream may emit events that happened right before it has started.
Ignore changes emitted for the directory itself, since they may come
from the stale events.
This was failing if the file didn't already exist.
Fixes unit tests on Windows:
* test\simple\test-http-curl-chunk-problem.js
* test\simple\test-pipe-file-to-http.js
This adds two new member functions getAuthTag and setAuthTag that
are useful for AES-GCM encryption modes. Use getAuthTag after
Cipheriv.final, transmit the tag along with the data and use
Decipheriv.setAuthTag to have the encrypted data verified.
gyp by default now tries to process gyp files in parallel by using
python's multiprocessing module, but it has problems on oddball
platforms. We don't have many files or complex dependency chains that
would benefit from parallel processing so disable by deafult
fixes#6640
The android generator for gyp currently doesn't support
--generator-output - this makes embedding node.js as project dependency
difficult for android projects.
Note: the generated files in deps/uv should be ignored in libuv's
.gitignore
1. Swallow errors when sending internal NODE_HANDLE_ACK messages, so
they don't crash the process.
2. Queue process.disconnect() if there are any pending queued messages.
Fixes test-child-process-fork-net2.js on win.
When calling `encOut` in loop, `maybeInitFinished()` may invoke
`clearOut`'s loop, leading to the writing of interleaved data
(encrypted and cleartext) into the one shared pool.
Move `maybeInitFinished()` out of the loop and add assertion for
future.
The null signal test existed, but only tested the case where the target
process existed, not when it did not exist.
Also clarified that SIGUSR1 is reserved by Node.js only for receiveing,
its not at all reserved when sending a signal with kill().
kill(pid, 'O_RDWR'), or any other node constant, "worked". I fixed this
by also checking for 'SIG'. The same as done in the isSignal() function.
Now the signal names supported by process.kill() are the same as those
supported by process.on().
2013-12-02 10:41:37 -08:00
6561 changed files with 807425 additions and 291879 deletions
"p3p: policyref=\"http://www.amazon.com/w3c/p3p.xml\",CP=\"CAO DSP LAW CUR ADM IVAo IVDo CONo OTPo OUR DELi PUBi OTRi BUS PHY ONL UNI PUR FIN COM NAV INT DEM CNT STA HEA PRE LOC GOV OTC \"\r\n"
,.headers={{"Date","Wed, 15 May 2013 17:06:33 GMT"}
,{"Server","Server"}
,{"x-amz-id-1","0GPHKXSJQ826RK7GZEB2"}
,{"p3p","policyref=\"http://www.amazon.com/w3c/p3p.xml\",CP=\"CAO DSP LAW CUR ADM IVAo IVDo CONo OTPo OUR DELi PUBi OTRi BUS PHY ONL UNI PUR FIN COM NAV INT DEM CNT STA HEA PRE LOC GOV OTC \""}
@@ -7,6 +7,8 @@ npm-tag(1) -- Tag a published version
## DESCRIPTION
THIS COMMAND IS DEPRECATED. See npm-dist-tag(1) for details.
Tags the specified version of the package with the specified tag, or the
`--tag` config if not specified.
@@ -23,6 +25,29 @@ This also applies to `npm dedupe`.
Publishing a package always sets the "latest" tag to the published version.
## PURPOSE
Tags can be used to provide an alias instead of version numbers. For
example, `npm` currently uses the tag "next" to identify the upcoming
version, and the tag "latest" to identify the current version.
A project might choose to have multiple streams of development, e.g.,
"stable", "canary".
## CAVEATS
Tags must share a namespace with version numbers, because they are
specified in the same slot: `npm install <pkg>@<version>` vs `npm
install <pkg>@<tag>`.
Tags that can be interpreted as valid semver ranges will be
rejected. For example, `v1.4` cannot be used as a tag, because it is
interpreted by semver as `>=1.4.0 <1.5.0`. See
<https://github.com/npm/npm/issues/6082>.
The simplest way to avoid semver problems with tags is to use tags
that do not begin with a number or the letter `v`.
## SEE ALSO
* npm-publish(1)
@@ -31,4 +56,5 @@ Publishing a package always sets the "latest" tag to the published version.
* npm-registry(7)
* npm-config(1)
* npm-config(7)
* npm-tag(3)
* npmrc(5)
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
