Andreas Madsen [Wed, 30 May 2012 18:32:50 +0000 (20:32 +0200)]
cluster: rename worker.unqiueID to worker.id
isaacs [Thu, 14 Jun 2012 15:37:44 +0000 (08:37 -0700)]
Update message test for new v8 behavior
Bert Belder [Thu, 14 Jun 2012 14:06:53 +0000 (16:06 +0200)]
test-bad-unicode: update to reflect V8 3.11 behavior
Igor Zinkovsky [Thu, 14 Jun 2012 08:33:56 +0000 (01:33 -0700)]
Windows: don't fire GC ETW events without ETW consumers
Nathan Rajlich [Thu, 14 Jun 2012 00:39:20 +0000 (17:39 -0700)]
test: use the debug build of node-weak when necessary
Nathan Rajlich [Thu, 14 Jun 2012 00:32:53 +0000 (17:32 -0700)]
test: add a .gitignore file to ignore weak's compiled "build" dir
Bert Belder [Thu, 14 Jun 2012 00:52:30 +0000 (02:52 +0200)]
LICENSE: add node-weak
Bert Belder [Thu, 14 Jun 2012 00:50:25 +0000 (02:50 +0200)]
LICENSE: punctuation fixes
yangguo@chromium.org [Wed, 13 Jun 2012 14:07:50 +0000 (14:07 +0000)]
Remove faster ticks for debug mode.
This solves the problem that code runs very slow when the debugger is enabled.
BUG=v8:2179
TEST=
Review URL: https://chromiumcodereview.appspot.com/
10544135
git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@11802
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
Bert Belder [Sat, 2 Jun 2012 14:01:18 +0000 (16:01 +0200)]
Unbreak the snapshotted build on Windows
The v8 team apparently decided that all build products should go
into ./build/«type», and updated their common.gypi file to do so.
Unfortunately v8's common.gypi is only used for some targets. All
the other targets would still look in the old place to find their
their dependencies, which effectively broke the build.
In the long run it would be good for node to send all build
output to ./build too, on all platforms.
Fedor Indutny [Wed, 13 Jun 2012 13:47:28 +0000 (15:47 +0200)]
v8: fix postmortem debugging tools build
Bert Belder [Wed, 13 Jun 2012 13:37:15 +0000 (15:37 +0200)]
v8: reapply floating patches
Note that
26f754d9 ('fix error handling in SendConnectMessage') is no
longer necessary; it was properly fixed upstream in V8 r11609.
mstarzinger@chromium.org [Wed, 13 Jun 2012 11:58:18 +0000 (11:58 +0000)]
Fix performance regression caused by r11202.
R=erik.corry@gmail.com
BUG=v8:2156,v8:2034
TEST=mjsunit/regress/regress-2156,mjsunit/regress/regress-2034
Review URL: https://chromiumcodereview.appspot.com/
10539131
git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@11800
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
Bert Belder [Wed, 13 Jun 2012 13:34:45 +0000 (15:34 +0200)]
v8: upgrade to v3.11.10
Bert Belder [Wed, 13 Jun 2012 23:28:51 +0000 (01:28 +0200)]
Inherited FDs should not accidentally be inherited by child processes
This makes test-child-process-detached pass on Windows.
Bert Belder [Wed, 13 Jun 2012 23:24:43 +0000 (01:24 +0200)]
Bert Belder [Wed, 13 Jun 2012 22:55:45 +0000 (00:55 +0200)]
vcbuild.bat: add duplicate arg-ok label to work around cmd.exe bug
Nathan Rajlich [Wed, 13 Jun 2012 22:55:29 +0000 (00:55 +0200)]
test: bundle node-weak in test/gc so that it doesn't need to be downloaded
Igor Zinkovsky [Mon, 11 Jun 2012 23:23:17 +0000 (16:23 -0700)]
Windows: Enable ETW events.
This commit enables ETW events to be fired on Windows for existing
DTrace probes. ETW instrumentation is enabled by default. It
is possible to build node.exe without ETW instrumentation by
using --without-etw option with configure script.
Bert Belder [Wed, 13 Jun 2012 20:10:59 +0000 (22:10 +0200)]
test-child-process-spawn-loop: make it pass on Windows.
It wasn't waiting for the child process' stderr to close, so not an
assertion was made *before* all the data that the child process sent
was received by node.
isaacs [Tue, 12 Jun 2012 17:02:52 +0000 (10:02 -0700)]
Fix #3388 Support listening on file descriptors
This implements server.listen({ fd: <filedescriptor> }). The fd should
refer to an underlying resource that is already bound and listening, and
causes the new server to also accept connections on it.
Not supported on Windows. Raises ENOTSUP.
Bert Belder [Wed, 13 Jun 2012 15:40:50 +0000 (17:40 +0200)]
Move resource files to src/res.
tools/msvs/res is not an appropriate place.
isaacs [Wed, 13 Jun 2012 02:05:51 +0000 (19:05 -0700)]
Fix #3407 os.tmpDir()
Bert Belder [Wed, 13 Jun 2012 01:02:53 +0000 (03:02 +0200)]
tests: make all message tests pass on Windows
Bert Belder [Wed, 13 Jun 2012 00:49:33 +0000 (02:49 +0200)]
test-net-timeout: make it pass on Windows
The test didn't take into account that some time might pass before
libuv makes close callbacks. This is now fixed.
Bert Belder [Tue, 12 Jun 2012 23:35:51 +0000 (01:35 +0200)]
Bert Belder [Tue, 12 Jun 2012 20:39:55 +0000 (22:39 +0200)]
test-child-process-kill: make it pass on windows
The test would fail if the child process writes anything to the stdout.
This doesn't happen on unix, since `cat` is spawned. However, on Windows
`cmd` is started, which *does* write stuff to it's stdout. This
meanlingless assert is now removed.
Bert Belder [Tue, 12 Jun 2012 20:37:36 +0000 (22:37 +0200)]
Fix child_process.kill oddities
* When the process is already dead, but the `exit` signal wasn't raised
yet, the ESRCH error should be ignored.
* When an invalid signal is specified, kill() should throw.
* Like process.kill(), child_process.kill() now preserves a `0` signal
which can be used to check the liveliness of the child process.
* process.kill() and child_process.kill() will now return true if the
signal was actually delivered, and false otherwise.
* When an `exec`-ed process is automatically killed because a time or
buffer limit is exceeded, and the kill() fails, this error should be
reported through the `exec` callback.
Fixes: #3409
Bert Belder [Tue, 12 Jun 2012 17:58:05 +0000 (19:58 +0200)]
test-exec: make it work on Windows
Ben Noordhuis [Tue, 12 Jun 2012 20:02:35 +0000 (22:02 +0200)]
doc: update crypto cipher/decipher docs
Ben Noordhuis [Tue, 12 Jun 2012 19:47:11 +0000 (21:47 +0200)]
crypto: make cipher/decipher accept buffer args
Bert Belder [Tue, 12 Jun 2012 17:19:26 +0000 (19:19 +0200)]
Bert Belder [Tue, 12 Jun 2012 17:52:38 +0000 (19:52 +0200)]
Revert "uv: upgrade to
b7e150ee"
Upgrade wasn't done correctly.
This reverts commit
b615077bab71db73cf065d6162cae8d6cb84d8d2.
Bert Belder [Tue, 12 Jun 2012 15:05:49 +0000 (17:05 +0200)]
Ben Noordhuis [Tue, 12 Jun 2012 14:04:56 +0000 (16:04 +0200)]
fs: fix infinite loop in fs.readFileSync()
Fix an infinite loop in the case where the file got truncated by a concurrent
writer while fs.readFileSync() was busy reading in the file.
Ben Noordhuis [Tue, 12 Jun 2012 13:32:40 +0000 (15:32 +0200)]
fs: fix infinite loop in fs.readFile()
Fix an infinite loop in the case where the file got truncated by a concurrent
writer while fs.readFile() was busy reading in the file.
Shigeki Ohtsu [Tue, 12 Jun 2012 07:01:03 +0000 (16:01 +0900)]
fs: fix fs.readFileSync to work on real empty file
Shigeki Ohtsu [Tue, 12 Jun 2012 06:01:34 +0000 (15:01 +0900)]
fs: fix typo in fs.readFile of lying size=0 stat
Bert Belder [Tue, 12 Jun 2012 00:22:30 +0000 (02:22 +0200)]
test-net-connect-econnrefused: don't take forever to complete
Bert Belder [Tue, 12 Jun 2012 00:04:44 +0000 (02:04 +0200)]
test-module-loading: convert backslashes to forward slashes
This makes the test pass on Windows.
Bert Belder [Mon, 11 Jun 2012 23:56:48 +0000 (01:56 +0200)]
test-http-get-pipeline-problem: don't fail if there are stray files in the temp dir
Bert Belder [Mon, 11 Jun 2012 23:40:04 +0000 (01:40 +0200)]
test-net-write-slow: increase the socket timeout period
In Windows the callbacks arrive in slightly different order. A bunch
of write operations complete immediately, and after that there is a
gap of a few hundred ms. This causes the timeout timer to fire, which
is not really warranted; the first few write operations just finished a
little quicker than expected.
isaacs [Mon, 11 Jun 2012 21:49:31 +0000 (14:49 -0700)]
fix fs.readFile with lying size=0 stat results
isaacs [Mon, 11 Jun 2012 22:24:38 +0000 (15:24 -0700)]
Add Buffer.concat method
We write out this loop a lot of places throughout node.
It clearly needs to be an API method.
Bert Belder [Mon, 11 Jun 2012 21:46:17 +0000 (23:46 +0200)]
cluster: don't silently drop messages when the write queue gets big
Bert Belder [Mon, 11 Jun 2012 19:22:50 +0000 (21:22 +0200)]
v8: force inlining of v8::internal::DescriptorArray methods
yangguo@chromium.org [Tue, 8 May 2012 12:38:24 +0000 (12:38 +0000)]
Force inlining CopyChars and String::Get.
BUG=
TEST=
Review URL: https://chromiumcodereview.appspot.com/
10332054
git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@11527
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
Charlie McConnell [Fri, 1 Jun 2012 04:23:05 +0000 (21:23 -0700)]
child_process: fix test implementation for options.detached
isaacs [Mon, 11 Jun 2012 16:30:44 +0000 (09:30 -0700)]
Do not gitignore npm's node_modules
isaacs [Mon, 11 Jun 2012 16:07:42 +0000 (09:07 -0700)]
Remove auto-unref
cc: @AvianFlu @AndreasMadsen
isaacs [Mon, 11 Jun 2012 15:58:33 +0000 (08:58 -0700)]
Now working on 0.7.11
isaacs [Mon, 11 Jun 2012 15:57:32 +0000 (08:57 -0700)]
Merge branch 'v0.7.10-release'
isaacs [Mon, 11 Jun 2012 04:25:41 +0000 (21:25 -0700)]
2012.06.11, Version 0.7.10 (unstable)
* Roll V8 back to 3.9.24.31
* build: x64 target should always pass -m64 (Robert Mustacchi)
* add NODE_EXTERN to node::Start (Joel Brandt)
* repl: Warn about running npm commands (isaacs)
* slab_allocator: fix crash in dtor if V8 is dead (Ben Noordhuis)
* slab_allocator: fix leak of Persistent handles (Shigeki Ohtsu)
* windows/msi: add node.js prompt to startmenu (Jeroen Janssen)
* windows/msi: fix adding node to PATH (Jeroen Janssen)
* windows/msi: add start menu links when installing (Jeroen Janssen)
* windows: don't install x64 version into the 'program files (x86)' folder (Matt Gollob)
* domain: Fix #3379 domain.intercept no longer passes error arg to cb (Marc Harter)
* fs: make callbacks run in global context (Ben Noordhuis)
* fs: enable fs.realpath on windows (isaacs)
* child_process: expose UV_PROCESS_DETACHED as options.detached (Charlie McConnell)
* child_process: new stdio API for .spawn() method (Fedor Indutny)
* child_process: spawn().ref() and spawn().unref() (Fedor Indutny)
* Upgrade npm to 1.1.25
- Enable npm link on windows
- Properly remove sh-shim on Windows
- Abstract out registry client and logger
isaacs [Mon, 11 Jun 2012 15:18:30 +0000 (08:18 -0700)]
Remove dep symlinks from tarball
Bert Belder [Mon, 11 Jun 2012 14:50:07 +0000 (16:50 +0200)]
process_wrap: set duplex flags when creating a pipe
isaacs [Mon, 11 Jun 2012 14:48:02 +0000 (07:48 -0700)]
lint
isaacs [Mon, 11 Jun 2012 04:29:47 +0000 (21:29 -0700)]
Upgrade npm to 1.1.25
Nuno Job [Sat, 9 Jun 2012 17:47:18 +0000 (18:47 +0100)]
test: add test for bad unicode sequences
This was a regression in 0.7.9.
isaacs [Sat, 9 Jun 2012 07:39:57 +0000 (00:39 -0700)]
Fix fs.realpath tests so that they actually run
isaacs [Sat, 9 Jun 2012 07:33:25 +0000 (00:33 -0700)]
Fix #3394 fs.realpath: Properly cache symlink targets
isaacs [Fri, 8 Jun 2012 23:49:03 +0000 (16:49 -0700)]
Fix fs.realpath to work on Windows
1. Make the isRoot check valid
2. Don't cache results based on dev/ino, since those are alwasy 0 on
windows.
isaacs [Fri, 8 Jun 2012 22:26:04 +0000 (15:26 -0700)]
Expose posix realpath on windows as well
Andreas Madsen [Fri, 8 Jun 2012 18:24:52 +0000 (20:24 +0200)]
domain: document and test dispose event
Andreas Madsen [Fri, 8 Jun 2012 18:06:10 +0000 (20:06 +0200)]
domain: run now return callback result
both domain.bind and domain.intercept act this way
Andreas Madsen [Fri, 8 Jun 2012 17:59:35 +0000 (19:59 +0200)]
domain: dry decorate using util._extend
Ben Noordhuis [Tue, 24 Apr 2012 01:41:49 +0000 (03:41 +0200)]
v8: debug: fix error handling in SendConnectMessage()
The old error handling code checked if the return value of Socket::Send() != 0,
which is wrong because Socket::Send() can write less bytes than requested or
return -1 on error.
isaacs [Tue, 29 May 2012 22:09:31 +0000 (15:09 -0700)]
v8: Floating patches
isaacs [Sat, 9 Jun 2012 15:09:42 +0000 (08:09 -0700)]
Roll V8 back to 3.9.24.31
Marc Harter [Wed, 6 Jun 2012 15:17:01 +0000 (10:17 -0500)]
Fix #3379 prevent domain.intercept passing 1st arg to cb
Charlie McConnell [Fri, 1 Jun 2012 04:23:05 +0000 (21:23 -0700)]
child_process: expose UV_PROCESS_DETACHED as options.detached
Bert Belder [Fri, 8 Jun 2012 18:46:34 +0000 (20:46 +0200)]
Revert "Unbreak the Windows build"
It is not needed for v8 3.10, and actually breaks the build.
This reverts commit
bd33fea732fa74d23fc31855631eba14a03a0c49.
Robert Mustacchi [Thu, 7 Jun 2012 23:21:04 +0000 (23:21 +0000)]
x64 target should always pass -m64
isaacs [Mon, 4 Jun 2012 18:10:31 +0000 (11:10 -0700)]
v8: Don't use 'echo -n' in configure script
Bert Belder [Sat, 2 Jun 2012 14:01:18 +0000 (16:01 +0200)]
Unbreak the Windows build
The v8 team apparently decided that all build products should go
into ./build/«type», and updated their common.gypi file to do so.
Unfortunately v8's common.gypi is only used for some targets. All
the other targets would still look in the old place to find their
their dependencies, which effectively broke the build.
In the long run it would be good for node to send all build
output to ./build too, on all platforms.
Conflicts:
deps/v8/build/common.gypi
Ben Noordhuis [Tue, 24 Apr 2012 01:41:49 +0000 (03:41 +0200)]
v8: debug: fix error handling in SendConnectMessage()
The old error handling code checked if the return value of Socket::Send() != 0,
which is wrong because Socket::Send() can write less bytes than requested or
return -1 on error.
isaacs [Tue, 29 May 2012 22:09:31 +0000 (15:09 -0700)]
v8: Floating patches
isaacs [Thu, 7 Jun 2012 18:45:30 +0000 (11:45 -0700)]
Roll V8 back to 3.10.8.13
Fedor Indutny [Wed, 6 Jun 2012 17:19:14 +0000 (21:19 +0400)]
child_process: spawn().ref() and spawn().unref()
isaacs [Wed, 6 Jun 2012 22:01:31 +0000 (15:01 -0700)]
build: Prevent duplication of doc/api folder
Ben Noordhuis [Wed, 6 Jun 2012 19:33:29 +0000 (21:33 +0200)]
fs: make callbacks run in global context
Callbacks that were passed to the binding layer ran in the context of the
(internal) binding object. Make sure they run in the global context.
Before:
fs.symlink('a', 'b', function() {
console.log(this); // prints "{ oncomplete: [Function] }"
});
After:
fs.symlink('a', 'b', function() {
console.log(this); // prints "{ <global object> }"
});
Ben Noordhuis [Wed, 6 Jun 2012 19:07:54 +0000 (21:07 +0200)]
fs: make fs.symlink() with no callback async
Fix a bug where fs.symlink('foo', 'bar') executed symlink(2) synchronously.
Joel Brandt [Wed, 6 Jun 2012 01:32:44 +0000 (18:32 -0700)]
add NODE_EXTERN to node::Start
Ben Noordhuis [Wed, 6 Jun 2012 03:21:07 +0000 (05:21 +0200)]
deps: upgrade libuv to 649ad50
isaacs [Tue, 5 Jun 2012 19:02:37 +0000 (12:02 -0700)]
Warn about running npm in the repl
Nathan Rajlich [Sun, 27 May 2012 07:47:40 +0000 (00:47 -0700)]
test: make the node-weak install build agaist this repo
This prevents node-gyp from attempting to download dev header files from
nodejs.org. Instead, the files in this repo will be used to build the module.
Ben Noordhuis [Tue, 5 Jun 2012 13:45:46 +0000 (15:45 +0200)]
deps: upgrade libuv to c8c9fe1
Ben Noordhuis [Tue, 5 Jun 2012 14:45:29 +0000 (16:45 +0200)]
udp_wrap, stream_wrap: lazy init slab allocator
Create slab allocator when binding is initialized.
Add an AtExit handler to destroy the slab before the VM shuts down, it can't be
disposed when V8 is dead and Valgrind will complain about memory leaks.
Ben Noordhuis [Tue, 5 Jun 2012 13:58:55 +0000 (15:58 +0200)]
slab_allocator: fix crash in dtor if V8 is dead
Don't try to dispose the persistent handles if the VM is already dead, it
triggers an assertion inside V8.
Shigeki Ohtsu [Tue, 5 Jun 2012 10:24:15 +0000 (19:24 +0900)]
slab_allocator: fix leak of Persistent handles
isaacs [Tue, 5 Jun 2012 00:32:46 +0000 (17:32 -0700)]
Upgrade npm to 1.1.24
isaacs [Mon, 4 Jun 2012 18:21:42 +0000 (11:21 -0700)]
build: Don't clobber symlinked npm
Bert Belder [Mon, 4 Jun 2012 19:39:23 +0000 (21:39 +0200)]
windows/msi: minor installer tweaks
Jeroen Janssen [Mon, 4 Jun 2012 16:59:01 +0000 (18:59 +0200)]
windows/msi: add node.js prompt to startmenu
isaacs [Mon, 4 Jun 2012 18:10:31 +0000 (11:10 -0700)]
v8: Don't use 'echo -n' in configure script
Ben Noordhuis [Mon, 4 Jun 2012 13:02:02 +0000 (15:02 +0200)]
net: rename flag FLAG_SHUTDOWNQUED
Jeroen Janssen [Sun, 3 Jun 2012 09:03:49 +0000 (11:03 +0200)]
windows/msi: fix adding node to PATH
Closes #3366
Bert Belder [Mon, 4 Jun 2012 12:04:15 +0000 (14:04 +0200)]
Improve child process stdio documentation
Jeroen Janssen [Sat, 2 Jun 2012 19:49:31 +0000 (21:49 +0200)]
windows/msi: add start menu links when installing
Jeroen Janssen [Sat, 2 Jun 2012 16:36:20 +0000 (18:36 +0200)]
windows/msi: cleanup WiX project files
The current WiX project files do some manual processing and generation
which WiX supports doing out of the box. This patch will use the
HeatDirectory task to generate the npm.wxs file and use the auto GUID
generation. I also changed the msi filename generation to include the
version number to match the currently used name for released msi files.
Closes #3360
Bert Belder [Sat, 2 Jun 2012 14:01:18 +0000 (16:01 +0200)]
Unbreak the Windows build
The v8 team apparently decided that all build products should go
into ./build/«type», and updated their common.gypi file to do so.
Unfortunately v8's common.gypi is only used for some targets. All
the other targets would still look in the old place to find their
their dependencies, which effectively broke the build.
In the long run it would be good for node to send all build
output to ./build too, on all platforms.