This commit attempts to address one of the TODOs in
https://github.com/nodejs/node/issues/4641 regarding making the
AtExit callback's per environment, instead of the current global.
bnoordhuis provided a few options for solving this, and one was to
use a thread-local which is what this commit attempts to do.
PR-URL: https://github.com/nodejs/node/pull/9163
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
Currently, the following warning will be reported when configuring
without-ssl:
../src/node.cc:3653:8: warning: unused variable 'use_bundled_ca'
[-Wunused-variable]
bool use_bundled_ca = false;
^
../src/node.cc:3654:8: warning: unused variable 'use_openssl_ca'
[-Wunused-variable]
bool use_openssl_ca = false;
^
I missed this when working on
commit 8a7db9d4b5 ("src: add
--use-bundled-ca --use-openssl-ca check").
Refs: https://github.com/nodejs/node/pull/12087
PR-URL: https://github.com/nodejs/node/pull/12302
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
When configuring and building --without-ssl the following warning is
reported:
../src/node_debug_options.cc:12:11: warning: unused variable
'default_inspector_port' [-Wunused-const-variable]
const int default_inspector_port = 9229;
The commit adds a HAVE_INSPECTOR guard to this constant.
PR-URL: https://github.com/nodejs/node/pull/12303
Reviewed-By: cjihrig - Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
I'm currently seeing a timeout error for test-signal-handler.js on
macosx when using the following configuration:
./configure --debug --without-ssl && make -j8 test
--without-ssl implies that there will be no inspector but the signal
SIGUSR1 is blocked in PlatformInit just the same. But in this case
never unblocked which is causing the signal to never be delivered to
the handlers in test-signal-handler.js and it loops until it times out.
Not sure if this is the best way of fixing this but hopefully more eyes
on this will help.
PR-URL: https://github.com/nodejs/node/pull/12266
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This commit adds lldbinit files from upstream V8 and also adds these so
that they get installed when `make install` is run.
Original commit message:
[tools] add lldbinit
The goal of this commit is to add the equivalent to gdbinit but
for lldb. I've tried to replicate the commands as close as possible
but I'm unsure about the jss command and hoping to get some feedback
on it in addition to the bta command which I'm not sure how/when
this could be used. This is probably just inexperience on my part.
The lldbinit file can be placed into a directory prefixed with dot
(.lldbinit) and the python script is currently expected to be in the
same directory. The path to the script can be changed manually if
needed as well.
NOTRY=true
Review-Url: https://codereview.chromium.org/2758373002
Cr-Commit-Position: refs/heads/master@{#44136}
PR-URL: https://github.com/nodejs/node/pull/12061
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Currently it is possible to configure using --without-ssl and
--shared-openssl/--openssl-no-asm/openssl-fips without an error
occuring.
The commit add check for these combinations:
$ ./configure --without-ssl --shared-openssl
Error: --without-ssl is incompatible with --shared-openssl
$ ./configure --without-ssl --openssl-no-asm
Error: --without-ssl is incompatible with --openssl-no-asm
$ ./configure --without-ssl --openssl-fips=dummy
Error: --without-ssl is incompatible with --openssl-fips
PR-URL: https://github.com/nodejs/node/pull/12175
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
At the moment Argv only supports three arguments which fulfilled my
requirements when working on #9163.
This commit adds support for a variable number of arguments. There is
also a no-args constructor that is the equivalent to running "node -p
process.version" which is hopefully alright as a default.
PR-URL: https://github.com/nodejs/node/pull/12166
Ref: https://github.com/nodejs/node/pull/9163
Reviewed-By: James M Snell <jasnell@gmail.com>
This commit tries to make it simpler to add unit tests (cctest) for
code that needs to test node core funtionality but that might not be
appropriate as an addon or a JavaScript test. An example of this could
be adding functionality targeted for situations when Node itself is
embedded.
Currently it was not as easy, or efficient, as one would have hoped to
add such tests. The object output directories vary for different
operating systems which we need to link to so that we don't have an
additional compilation step.
PR-URL: https://github.com/nodejs/node/pull/11956
Ref: https://github.com/nodejs/node/pull/9163
Reviewed-By: James M Snell <jasnell@gmail.com>
When configuring node with --openssl-use-def-ca-store the root certs
from OpenSSL should be used and not the ones in src/node_root_certs.h.
I noticed that src/node_root_certs.h is still included even when
using --openssl-use-def-ca-store.
This commit adds check and does not include node_root_certs.h if
--openssl-use-def-ca-store is specified.
PR-URL: https://github.com/nodejs/node/pull/11939
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
I think there might be an issue with the value of OBJ_DIR when
using a "mac" os. The value is currently specified in common.gypi
which is included by node.gyp:
'OBJ_DIR': '<(PRODUCT_DIR)/obj',
In the generated Makefile (out/Makefile) the object output directory
is:
obj := $(builddir)/obj
And in the included node.target.mk we have the OBJS declared:
OBJS := \
$(obj).target/$(TARGET)/src/async-wrap.o \
$(obj).target/$(TARGET)/src/cares_wrap.o \
If OBJ_DIR is used in node.gyp to point to generated object files
on mac they will not be found.
PR-URL: https://github.com/nodejs/node/pull/11857
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
When configuring --without-ssl and then running make install
openssl headers will be copied from deps/openssl to the target
installation directory.
This commit adds a check for is node_use_openssl is set in which
case the headers are not copied.
PR-URL: https://github.com/nodejs/node/pull/11766
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Brian White <mscdex@mscdex.net>
Currently test-tls-legacy-deprecated will fail if node
was built using --without-ssl. This commit adds a check to
verify is crypto support exists and if not skip this test.
PR-URL: https://github.com/nodejs/node/pull/11747
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently when node is build --without-ssl and the test are run,
there are a number of failing test due to tests expecting crypto
support to be available. This commit fixes fixes the failure and
instead skips the tests that expect crypto to be available.
PR-URL: https://github.com/nodejs/node/pull/11631
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
When building --without-ssl and then running:
out/Release/node --expose-internals
test/parallel/test-internal-util-assertCrypto.js
The following error is displayed:
assert.js:420
throw actual;
^
Error: Node.js is not compiled with openssl crypto support
at Object.exports.assertCrypto (internal/util.js:82:11)
at assert.throws
(/Users/danielbevenius/work/nodejs/node/test/parallel/test-internal-util-assertCrypto.js:8:28)
at _tryBlock (assert.js:379:5)
at _throws (assert.js:398:12)
at Function.throws (assert.js:428:3)
at Object.<anonymous>
(/Users/danielbevenius/work/nodejs/node/test/parallel/test-internal-util-assertCrypto.js:8:10)
at Module._compile (module.js:571:32)
at Object.Module._extensions..js (module.js:580:10)
at Module.load (module.js:488:32)
at tryModuleLoad (module.js:447:12)
It looks like the regex is not taking into account the 'Error: ' hence
the failure to match it.
PR-URL: https://github.com/nodejs/node/pull/11620
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently when building with the following configuration options:
$ ./configure --without-ssl && make
The following link error is reported:
Undefined symbols for architecture x86_64:
"node::openssl_config", referenced from:
node::Init(int*, char const**, int*, char const***) in node.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see
invocation)
Adding an HAVE_OPENSSL directive around this code allows the build to
pass.
PR-URL: https://github.com/nodejs/node/pull/11618
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
It might be useful to sometimes run a single cctest while debugging/
developing. This commit adds support for Google Test's --gtest_filter
to enable this.
To list the tests available:
$ make list-gtests
UtilTest.
ListHead
StringEqualNoCase
StringEqualNoCaseN
ToLower
Malloc
Calloc
UncheckedMalloc
UncheckedCalloc
InspectorSocketTest.
ReadsAndWritesInspectorMessage
BufferEdgeCases
AcceptsRequestInSeveralWrites
ExtraTextBeforeRequest
ExtraLettersBeforeRequest
RequestWithoutKey
KillsConnectionOnProtocolViolation
CanStopReadingFromInspector
CloseDoesNotNotifyReadCallback
CloseWorksWithoutReadEnabled
ReportsHttpGet
HandshakeCanBeCanceled
GetThenHandshake
WriteBeforeHandshake
CleanupSocketAfterEOF
EOFBeforeHandshake
Send1Mb
ErrorCleansUpTheSocket
InspectorSocketServerTest.
InspectorSessions
ServerDoesNothing
ServerWithoutTargets
ServerCannotStart
StoppingServerDoesNotKillConnections
Then run a single test:
$ env GTEST_FILTER=InspectorSocketTest.GetThenHandshake make cctest
Note: Google Test filter = InspectorSocketTest.GetThenHandshake
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from InspectorSocketTest
[ RUN ] InspectorSocketTest.GetThenHandshake
[ OK ] InspectorSocketTest.GetThenHandshake (0 ms)
[----------] 1 test from InspectorSocketTest (0 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (0 ms total)
[ PASSED ] 1 test.
PR-URL: https://github.com/nodejs/node/pull/11474
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
I noticed that the link to http-parser is pointing to the joyent
organization. There is a redirect to the nodejs organization but
perhaps this should be updated anyway.
PR-URL: https://github.com/nodejs/node/pull/11477
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, the sources list contains sources and headers which are
separated by a comment. I noticed two .cc files after the headers
comment and this commit moves those files the start of the list
where the rest of source files are.
PR-URL: https://github.com/nodejs/node/pull/10850
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Italo A. Casas <me@italoacasas.com>
Currently, when test/cctest/test_inspector_socket_server.cc is run there
is output written to stderr by src/inspector_socket_server.cc which is
interleaved with the gtest report:
Debugger listening on port 9229.
Warning: This is an experimental feature and could change at any time.
To start debugging, open the following URLs in Chrome:
...
The goal of this commit is to remove the above logged information
by introducing an out_ member in the InspectorSocketServer class
which defaults to stderr (keeping the current behavior).
Setting out_ to NULL is supported in which case nothing will be written
and is what the test has been configured with. When working on specific
test case the appropriate output stream can be specified for the
ServerHolder constructor to limit logging to that test case.
PR-URL: https://github.com/nodejs/node/pull/10537
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Eugene Ostroukhov <eostroukhov@google.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, there are a number of popups that get displayed when running
the tests asking to accept incoming network connections. Rules can be
added manually to the socket firewall on Mac OS X but getting this right
might not be obvious and quite a lot of time can be wasted trying to get
the rules right. This script hopes to simplify things a little so that
it can be re-run when needed.
The script should be runnable from both the projects root directory and
from the tools directory, for example:
$ sudo ./tools/macosx-firewall.sh
Fixes: https://github.com/nodejs/node/issues/8911
PR-URL: https://github.com/nodejs/node/pull/10114
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, some of the docs use different indentation for comments
in the code examples. This commit makes the indentation consistent
by putting the comments at the beginning of the line (really no
indentation that is).
PR-URL: https://github.com/nodejs/node/pull/9518
Reviewed-By: Teddy Katz <teddy.katz@gmail.com>
Reviewed-By: Roman Reiss <me@silverwind.io>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Currently, when building the addons the following warning is displayed:
make[2]: warning: jobserver unavailable: using -j1. Add `+' to parent
make rule.
Adding the MAKEFLAGS="-j1" to avoid the warning.
Also updated the log message to say that it is building the addon and
not running the test as I think that is more accurate.
PR-URL: https://github.com/nodejs/node/pull/9450
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently, make lint is failing with the following error:
3:7 error 'common' is assigned a value but never used no-unused-vars
✖ 1 problem (1 error, 0 warnings)
PR-URL: https://github.com/nodejs/node/pull/9334
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
The doc/api/addons.md document contains examples of Addon
Initialization functions that take a parameter named exports.
This also matches the name used in node.cc when calling:
mp->nm_register_func(exports, module, mp->nm_priv);
Currently, a number of the tests name this same parameter target. This
commit renames target to exports for consistency.
PR-URL: https://github.com/nodejs/node/pull/9135
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Commit fdca79fbc0 ("test: enable addons
test to pass with debug build") enabled the addons tests to pass when
the build type is of type debug (configure --debug).
test/addons/node-module-version/test.js was recently added and expects
the the build type to be of type Release (like most of the others until
recently). This commit allows this test to pass when the build type if
of type debug.
PR-URL: https://github.com/nodejs/node/pull/9093
Reviewed-By: Santiago Gimeno <santiago.gimeno@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
The config.gypi target has a recipe that uses the control function error
to report if the config.gypi file is missing or if it is stale (the
configure file was updated which is a prerequisite of this rule).
GNU make has two phases, immediate and deferred. During the first phase
it will expand any variables or functions as the makefile is parsed.
The recipe in this case is a shell if statement, which is a deferred
construct. But the control function $(error) is an immediate construct
which will cause the makefile processing to stop during the first phase
of the Make process.
If I understand this correctly the only possible outcome of this rule is
the "Stale config.gypi, please re-run ./configure" message which will
be done in the first phase and then exit. The shell condition will not
be considered. So it will never report that the config.gypi is missing.
bnoordhuis suggested that we simply change this into a single error
message:
"Missing or stale config.gypi, please run configure"
PR-URL: https://github.com/nodejs/node/pull/9053
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
This commit attempts to fix one of the items in
https://github.com/nodejs/node/issues/4641, which was to remove a
TODO the UDPWrap::OnSend function and share the code in that method with
StreamWrap::AfterWrite.
While looking into this addaleax pointed out that the implementations
for these two functions have diverged since the original comment
was added:
$ git log --pretty=short -u -L 357,357:src/udp_wrap.cc
$ git show cbd4033619cc45abdf878285c412bac9c3f36e4e:src/udp_wrap.cc |
grep -1 -A26 'UDPWrap::OnSend'
git show cbd4033619cc45abdf878285c412bac9c3f36e4e:src/stream_wrap.cc |
grep -A27 'void StreamWrap::AfterWrite'
Removing the TODO comment seems appropriate in this case.
PR-URL: https://github.com/nodejs/node/pull/9000
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Currently when running configure with the --debug option in combination
with the tests (./configure --debug && make -j8 test) there are a few
addon tests that fail with error messages similar to this:
=== release test ===
Path: addons/load-long-path/test
fs.js:558
return binding.open(pathModule._makeLong(path), stringToFlags(flags),
mode);
^
Error: ENOENT: no such file or directory, open
'/nodejs/node/test/addons/load-long-path/build/Release/binding.node'
at Object.fs.openSync (fs.js:558:18)
at Object.fs.readFileSync (fs.js:468:33)
at Object.<anonymous>
(/nodejs/node/test/addons/load-long-path/test.js:28:19)
at Module._compile (module.js:560:32)
at Object.Module._extensions..js (module.js:569:10)
at Module.load (module.js:477:32)
at tryModuleLoad (module.js:436:12)
at Function.Module._load (module.js:428:3)
at Module.runMain (module.js:594:10)
at run (bootstrap_node.js:382:7)
Command: out/Release/node
/nodejs/node/test/addons/load-long-path/test.js
This commit allows for the tests to pass even if the configured build
type is of type debug.
PR-URL: https://github.com/nodejs/node/pull/8836
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Ilkka Myller <ilkka.myller@nodefield.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Currently the readable and writable arguments are not specified in the
req.oncomplete method. Adding and asserting that they are always true
(which is always the case for TCP). This might seem unnecessary but it
can't hurt to have them to pickup any breaking modifications made to
ConnectionWrap::AfterConnect in the future.
PR-URL: https://github.com/nodejs/node/pull/8815
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Currently, there are a few places where macro functions passed to the
PER_ISOLATE_PRIVATE_SYMBOL_PROPERTIES and
PER_ISOLATE_STRING_PROPERTIES macros, don't use the StringValue parameter.
This commit removes the StringValue parameter where it is not used.
PR-URL: https://github.com/nodejs/node/pull/7905
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
The node.js script was renamed to bootstrap_node.js as part of
81b6882e51. Several comments were
missed in src/node.cc that referred to the old file name.
This commit updates the comments to refer to bootstrap_node.js and
correct the path to this file where used.
It also moves a comment that seems to have drifted in the file.
PR-URL: https://github.com/nodejs/node/pull/8092
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Brian White <mscdex@mscdex.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Currently the build-addons target is called as part of the
test target, and it has test/addons/.buildstamp as a dependency.
When running ./configure --debug and later switching/
updating branches you can be in a situation where the config.gypi
files in the addons build directories are using an incorrect
value for default_configuration. Currently this can happen and you
will get test errors as there are a few test that depend on the
value of default_configuration to be Release.
Ben Noordhuis provided the solution for this by adding a dependency to
config.gypi, which is generated when running ./configure, and will
correct this situation.
This commit is related to #7860. Please see the dicussion in that
issue regarding the test using hard-coded paths.
PR-URL: https://github.com/nodejs/node/pull/7893
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This commit attempts to address one of the items in #4641 which is
related to src/pipe_wrap.cc and src/tcp_wrap.cc. Currently both
pipe_wrap.cc and tcp_wrap.cc contain a class that are almost
identical. This commit extracts these parts into a separate class
that both can share.
PR-URL: https://github.com/nodejs/node/pull/7501
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
One of the issues in #4641 concerns OnConnection in pipe_wrap and
tcp_wrap which are very similar with some minor difference in how
they are coded. This commit extracts OnConnection so both these
classes can share the same implementation.
PR-URL: https://github.com/nodejs/node/pull/7547
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
This commit updates the node.js script name to reflect its
actual name, which is now bootstrap_node.js. This commit also
fixes the requisite message tests, and relocates a comment
which seems to have drifted.
PR-URL: https://github.com/nodejs/node/pull/7277
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Brian White <mscdex@mscdex.net>
Daniel Bevenius