daohualiuxiang
/
node
mirror of https://github.com/nodejs/node.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
2,750 Commits
53 Branches
866 Tags
2 GiB
Commit Graph

769 Commits (cd921c2c62cc5ab4d79ed403f8f646566b8ac526)

Author SHA1 Message Date
Ryan Dahl bbbcd1fee0 Safe constructors for fs.ReadStream and fs.WriteStream 2010-09-16 19:04:09 -07:00
Paul Querna c8f9728de7 Move dns.isIP to net.isIP 
Add tests and docs.
 2010-09-16 17:27:10 -07:00
Ryan Dahl 251d03197f Remove deprecation warnings in net module 2010-09-16 15:02:12 -07:00
Ryan Dahl f47ad10db7 Remove a few old deprecation warnings 2010-09-15 16:18:54 -07:00
Sam Stephenson 6560ab9073 Send multiple header lines in ServerResponse#writeHead when header values are arrays 2010-09-15 15:55:30 -07:00
Ryan Dahl feea1330cc Make a list of known globals 
And fix missing var!

It would be good to get this script running at the end of every test, so we
know that modules aren't leaking either - but it will require a lot
modification of the tests so that they themselves aren't leaking globals.
 2010-09-14 23:03:29 -07:00
Ryan Dahl 77fc61d539 Default value for second arg of Buffer#slice 2010-09-14 15:39:27 -07:00
Mark Nottingham 4fe3007a1a Support for outgoing HTTP trailing headers 2010-09-14 15:20:45 -07:00
Ryan Dahl 83ff473d30 Use child_process.exec rather than sys.exec 2010-09-14 10:29:56 -07:00
isaacs 686d9f9cc6 Bug in realpath with symlinks to absolute folder paths which have children. 
Found by Cliffano Subagio
http://groups.google.com/group/nodejs/browse_thread/thread/f46f093938265ac0/387e14da08c7dd7b?
 2010-09-14 10:27:57 -07:00
Peter Griess 422d3c93bc Get rid of PHP/Rails style parameter munging. 
- Handle only the most basic of query string parsing and construction.
  Leave the rest (e.g. Rails/PHP behaviors) to modules higher up the
  stack, like Express.
 2010-09-13 12:39:30 -07:00
Tony Metzidis debf389490 Catch Exceptions thrown when openssl is disabled 2010-09-13 12:04:31 -07:00
Paul Querna 9a701b0a70 Expose fingerproint from getPeerCertificate 
Expose the SHA1 digest of the certificate as the fingerprint attribute in
the object returned by getPeerCertificate()
 2010-09-09 23:35:17 -07:00
Ryan Dahl 3a9570386a Fix fs.realpathSync('/') 2010-09-09 18:49:28 -07:00
isaacs 85fb47c11c Better temporary directory handling for tests. 
Add a setUp and tearDown function to the test case class, and use it to
create and remove the test/tmp directory for each test.

TODO: amend other tests.
 2010-09-09 16:43:18 -07:00
Ryan Dahl 7628905a9b Don't use empty.js - breaks module test 2010-09-09 16:22:57 -07:00
Felix Geisendörfer f870240dcf Simple benchmark for node's startup time 2010-09-09 15:51:17 -07:00
Ryan Dahl efc723787a Fix big string bug 2010-09-09 11:03:50 -07:00
Ryan Dahl 3e9f636b64 Improve long buffer test 2010-09-09 11:03:50 -07:00
Ryan Dahl 8078ed1f86 Expose fast-buffer bug in tests 2010-09-09 11:03:50 -07:00
Ryan Dahl 7b772f3f68 More fast buffer work 2010-09-09 11:03:48 -07:00
Ryan Dahl d5bdda74c1 fast buffer bounds checking in copy() 2010-09-09 11:03:48 -07:00
Felix Geisendörfer ef54777fa5 Fix: uncaughtException was broken for main module 
See: 635986e433
 2010-09-09 10:47:30 -07:00
Ryan Dahl 17f3ffa633 Call Tick() after coming out of select() 
Previously we would only call it before going into select(). This is needed
to fix test/simple/test-next-tick-ordering2.js.
 2010-09-09 10:30:43 -07:00
isaacs ba0c32e2e1 Fix issue #262. Allow fs.realpath to traverse above the current working directory. 2010-09-08 17:35:58 -07:00
Tobie Langel 512016fd74 Test for ReadStream typo 2010-09-08 12:09:13 -07:00
Tobie Langel ccf4afa256 Do not emit WriteStream's drain event before ws.write has been called. 2010-09-08 11:58:54 -07:00
Tobie Langel f5e4047064 Avoid closing a WriteStream before it has been opened. 2010-09-08 11:51:29 -07:00
Herbert Vojčík 6789ab16d8 Modifying test-global to accomodate v8 inter-context 'global' protection. 2010-09-08 10:37:26 -07:00
Herbert Vojčík 2a03d5802b Removing test-global-between-modules. 
Using "global" to push data to require()d modules
not supported under NODE_MODULE_CONTEXTS=1.
 2010-09-08 10:36:45 -07:00
Felix Geisendörfer cf4b5fc52a Test case showing a bug in nextTick ordering 
nextTick should fire before setTimeout in this test, but it doesn't.
 2010-09-07 22:37:14 -07:00
Ryan Dahl 1d0fb850d9 test-http-parser should not use private API 2010-09-07 17:17:12 -07:00
Ryan Dahl 4fe5e86a7e Don't refer to private API in test-repl 2010-09-07 17:16:24 -07:00
Benjamin Thomas cda1a38426 Fix bug in process._tickCallback where callbacks can get abandoned. 
Change process._tickCallback so that if a callback throws an error but
there are other callbacks after it, we indicate that
process._tickCallback needs to be ran again.

Currently, if a callback in process._tickCallback throws an error, and
that error is caught by an uncaughtException handler and
process.nextTick is never called again, then any other callbacks already
added to the nextTickQueue won't be called again.

Updated the next-tick-errors test to catch this scenario.
 2010-09-02 09:24:39 -07:00
isaacs 0e311717b5 Treat "//some_path" as pathname rather than hostname by default. 
Note that "//" is still a special indicator for the hostname, and this does
not change the parsing of mailto: and other "slashless" url schemes.  It
does however remove some oddness in url.parse(req.url) which is the most
common use-case for the url.parse function.
 2010-09-02 09:24:21 -07:00
Ryan Dahl 7347fb3e2c Make sure setInterval(cb, 0) loops infinitely 2010-08-31 16:45:17 -07:00
Ryan Dahl a055153cda Fix style in test/simple/test-dgram-unix.js 2010-08-30 14:17:27 -07:00
Ryan Dahl f188b9d6d7 Remove timer from test/simple/test-dgram-unix.js 
Test running already has a timeout mechanism.
 2010-08-30 14:15:30 -07:00
Marco Rogers af9aa93e0c fix for fs.readFile to return string when encoding specified on zero length read 2010-08-30 14:08:18 -07:00
Marco Rogers 9c7c6e93e1 Fixed async fs writes with length 0, it should fire the callback 2010-08-30 13:58:50 -07:00
Ryan Dahl 635986e433 Add failing uncaughtException test 
FIXME
 2010-08-30 12:02:22 -07:00
Johan Euphrosine e49be4768b add readline support for meta-d 2010-08-30 10:26:51 -07:00
Russell Haering 37b6e10684 Pass an error to the sys.pump callback if one occurs 
- Add test case for pumping from unreadable stream.
- Document the sys.pump error handling behavior
 2010-08-30 10:23:48 -07:00
Benjamin Thomas 57642e2349 Fix process.nextTick so thrown errors don't confuse it. 
If the function for a process.nextTick throws an error, then the
splice() never removes that function from the nextTickQueue array.  This
makes sure the functions that have been run in _tickCallback get removed
regardless of errors.

Also add a test for this.
 2010-08-27 22:13:50 -07:00
Ryan Dahl 4fdebab005 Safe constructor: net.Server, net.Stream 2010-08-27 15:38:46 -07:00
Ryan Dahl f86ad1693f Safe Constructor: Buffer 2010-08-27 15:23:51 -07:00
Ryan Dahl 44a287987e Safe constructors: http.Server and http.Client 2010-08-27 13:58:52 -07:00
Johan Euphrosine 51224f12ca add test for readline putty support 2010-08-23 10:25:55 -07:00
Johan Euphrosine 748469c71c add readline support for meta-f and meta-b 2010-08-23 01:47:41 -07:00
Johan Euphrosine fd3e84499e add home/end support in rxvt and readline tests 2010-08-23 01:46:14 -07:00
Johan Euphrosine bedca2e7a9 add tests for console.log arguments handling 2010-08-22 12:34:00 -07:00
Ryan Dahl 2982e75a0b set-cookies header is an array of values. always 2010-08-21 16:34:38 -07:00
Ryan Dahl 60b93cca6d Fix race conditions in test-dgram-multicast.js 2010-08-19 18:49:27 -07:00
Herbert Vojčík 28b21d15c3 Fix undefined_reference_in_new_context test for own-context module loader. 2010-08-18 14:06:30 -07:00
Herbert Vojčík 2b126da395 Tests for behaviour of 'global'. 2010-08-18 12:08:39 -07:00
Herbert Vojčík 9253333850 Fix registerExtension for NODE_MODULE_CONTEXTS 
Fix of registerExtension-produced non-string module loading with own
context. Plus finishing touches to the test.
 2010-08-18 11:46:37 -07:00
Herbert Vojčík cf2b206a8e More changes to tests so they really work under context module loader. 
Plus, getting rid of test/common.js defining things in global.
 2010-08-17 08:41:05 -07:00
Herbert Vojčík 32e309440f Fix of 'sys.common.debug' calls in tests. 2010-08-17 08:08:36 -07:00
Marco Rogers 6744e59e46 detect 0 length fs writes with tests 2010-08-16 23:57:01 -07:00
Ryan Dahl 4593c04959 Implement net.Server.maxConnections 
Simplify EMFILE behavior.
 2010-08-15 14:01:55 -07:00
Ryan Dahl a8c2bb41af Fix tests 2010-08-13 08:55:42 -07:00
Ryan Dahl 03123435e0 Remove platform specific asserts 2010-08-13 08:13:54 -07:00
Bert Belder 0e8d858aba One more test with a race condition fixed 2010-08-13 07:18:38 -07:00
Rasmus Andersson 758f12f34c added read and write support for process.title Darwin 
This will only manipulate the OS X-level process name, not the title shown
in e.g. ps.
 2010-08-12 16:23:41 -07:00
Rasmus Andersson 8d5e05668b Added support for multiple listeners to DNS multicast datagrams. 
Some platforms require SO_REUSEPORT to be set for the socket.
 2010-08-11 19:02:20 -07:00
Bert Belder 7059be19ec Fix http and net tests failing due to race condition 
Plus some minor cosmetic corrections
 2010-08-11 17:27:44 -07:00
Ryan Dahl 9395786d11 Fix race conditions in tests 2010-08-10 17:48:37 -07:00
Bert Belder aaa1f451e6 Expose cwd option to child_process.exec() 2010-08-06 13:38:41 -07:00
Bert Belder 94914135df New api for child_process.spawn; ability to set cwd for spawn()ed process 
Tests for child_process.spawn() use new API

Test for deprecated child_process.spawn() API
 2010-08-06 13:37:30 -07:00
Ryan Dahl d408de87fc Remove module.js - put code into src/node.js 2010-08-06 12:34:02 -07:00
cloudhead fe3e419687 querystring.parse: handle undefined value properly 2010-08-05 10:46:28 -07:00
Nick Stenning 0a3eff8021 Standardise module load order for native and registered file extensions. 
This patch standardises the load order for modules. Highest priority is trying to load exactly the file the user specified, followed by native extensions, followed by registered extra extensions, etc.

In full, if we require('foo') having registered '.coffee' as an alternative extension, we try and load the following files in order:

    foo
    foo.js
    foo.node
    foo.coffee
    foo/index.js
    foo/index.node
    foo/index.coffee
 2010-08-04 16:03:08 -07:00
Nick Stenning 78520ba482 Don't attempt to load a directory. 
This patch replaces the path.exists check for module loading with a call to
fs.statSync (or fs.stat for require.async) which ensures that it's not trying
to load a directory.
 2010-08-04 15:55:47 -07:00
Ryan Dahl 7db5c8a10d Fix toString('base64') bug 
Thanks to Stepan Stolyarov for the test case.
 2010-08-04 11:38:48 -07:00
Ryan Dahl adec544fdd Revert "Expose the V8 debug object process.debug" 
This reverts commit d9fbb8a580.
 2010-08-04 10:38:19 -07:00
Ryan Dahl 31b59400f8 Allow null context in Script 2010-08-04 10:20:40 -07:00
Ryan Dahl d9fbb8a580 Expose the V8 debug object process.debug 
Add one duplicate test from V8, just to make sure it works.
 2010-08-02 00:46:09 -07:00
Mikeal Rogers 3214116be6 Implement keep-alive for http.Client 
Send the 'Connection: keep-alive' header in your request to enable.
 2010-08-01 22:34:17 -07:00
isaacs 65037eeb32 Don't let path.normalize get above the root. 
Any path.join or path.normalize that starts with a / will not go "above" that after normalization.  This is important because /../foo is almost *always* some sort of error, and doesn't match the corollary in sh: `cd $p; pwd`

At the worse, this can be a vector for exploits, since a static file server might do path.join(docroot, path.normalize("/"+req)) to get the file.  If the normalized request path could be something like "/../../../etc/passwd" then bad things could happen.
 2010-08-01 20:20:17 -07:00
Ben Noordhuis f72ac17c89 Buffer: graciously handle padding in base64-encoded input. 2010-07-30 10:49:00 -07:00
Ryan Dahl f291fbc8a7 Fix ending \0 in base64 decoding 2010-07-29 14:52:40 -07:00
Ben Noordhuis 95638c9b0d Buffer: adjust buffer size so the base64-decoded input fits snugly. 
Stops Valgrind from complaining about uninitialized memory access.
 2010-07-28 11:37:23 -07:00
Dmitry Baranovskiy 8ab3c3e7c3 Added ability to pass offset to buffer write and toString methods as a string, i.e. '2' and encoding as anything 2010-07-25 20:16:37 -07:00
Ryan Dahl 2c1ca4078d Implement buffer.write for base64 
There might be an off-by-one on the returned value.
 2010-07-23 16:36:52 -07:00
Ryan Dahl 528015e0d8 Implement buffer.toString('base64') 2010-07-23 13:52:44 -07:00
isaacs f0f247d7e5 Fix dirname so that dirname('/a/b/') -> '/a', like sh's does. 
Before there was this comment:
  Can't strip trailing slashes since module.js incorrectly
  thinks dirname('/a/b/') should yield '/a/b' instead of '/a'.
But now, such thinking is corrected.
 2010-07-23 09:08:49 -07:00
Chandra Sekar S f5f7cb9264 Support for reading byte ranges from files using fs.createReadStream. 2010-07-21 11:33:35 -07:00
Peter Griess 56f200af5d Fix Buffer.toString() on 0-length slices. 
- Buffer.toString('ascii', 0, 0) incorrectly returns the entire contents
  of the buffer. Fix this.
- Provide similar behavior to Buffer.write() and Buffer.copy() when
  dealing with 0-length in valid and invalid byte ranges.
 2010-07-20 15:38:10 -07:00
isaacs d75b63bc3c Support including modules that don't have an extension. 
This way, require("/foo") will work if there is a "foo.js", or a file named
simply "foo" with no extension.
 2010-07-20 10:26:24 -07:00
Ryan Dahl 7067a7155f Specify env differently in execFile 
Callbacks should always be the last argument.
 2010-07-19 20:08:35 -07:00
isaacs 49e0f14a2f Cache modules based on filename rather than ID 
This is ever so slightly less efficient than caching based on ID, since the
filename has to be looked up before we can check the cache.  However, it's
the most minimal approach possible to get this change in place.  Since
require() is a blocking startup-time operation anyway, a bit of slowness is
not a huge problem.

A test involving require.paths modification and absolute loading. Here's the
gist of it.

Files: /p1/foo.js /p2/foo.js

  1. Add "/p1" to require.paths.
  2. foo1 = require("foo")
  3. assert foo1 === require("/p1/foo") (fail)
  4. Remove /p1 from require.paths.
  5. Add /p2 to require.paths.
  6. foo2 = require("foo")
  7. assert foo1 !== foo2 (fail)
  8. assert foo2 === require("/p2/foo") (fail)

It's an edge case, but it affects how dependencies are mapped by npm.
If your module requires foo-1.2.3, and my module requires foo-2.3.4,
then you should expect to have require("foo") give you foo-1.2.3, and
I should expect require("foo") to give me foo-2.3.4.  However, with
module ID based caching, if your code loads *first*, then your "foo"
is THE "foo", so I'll get your version instead of mine.

It hasn't yet been a problem, but only because there are so few
modules, and everyone pretty much uses the latest version all the
time.  But as things start to get to the 1.x and 2.x versions, it'll
be an issue, I'm sure.  Dependency hell isn't fun, so this is a way to
avoid it before it strikes.
 2010-07-19 14:17:22 -07:00
Micheil Smith 299671b731 querystring: Removing typecasting of numeric strings to numbers 
The tests did not accurately test for a strict equality, meaning that the
number == to the string.
 2010-07-19 11:07:08 -07:00
Jan Kassens f18d9d8466 querystring.js: JSLint warnings fixed 2010-07-19 10:54:50 -07:00
Jan Kassens bb2acd5e75 querystring.stringify is now more solid 
* handles NaN and Infinity
* works with arrays from other contexts
 2010-07-19 10:53:32 -07:00
Benjamin Fritsch 0ed3532699 added test for Issue #214 2010-07-19 10:16:31 -07:00
Ryan Dahl 458bbb4687 fs.writeSync should return bytes written 2010-07-16 19:05:28 -07:00
Сергей Крыжановский 078a48a97b added env to child_process.exec 2010-07-15 17:17:03 -07:00
Ryan Dahl c926ab7746 Fix test-executable-path.js 
Now that process.version doesn't say if the build is debug or not.
 2010-07-15 14:21:32 -07:00
Ryan Dahl 8921bf454d Remove race conditions and simplify test-http-exceptions.js 2010-07-15 14:21:32 -07:00
Ryan Dahl 4c6f4dabf5 Adjust tests for error reporting on JSON.parse(undefined) 
The V8 upgrade changed the stacktrace reports (?)
 2010-07-15 14:21:32 -07:00
Ryan Dahl 9fd5e3c89c Update tests to work with module contexts 2010-07-15 14:21:31 -07:00
Ryan Dahl 5f30377bbc Load modules in individual contexts 
Add NODE_MODULE_CONTEXTS env var

Only one test was modified to check that this works. NEED to go through all
tests and modify them so that

  NODE_MODULE_CONTEXTS=1 make test

passes.
 2010-07-15 10:34:39 -07:00
Matt Ranney 4e50197e53 Datagram socket refactor. Add tests and documentation. 
Support setTTL() and setBroadcast() socket options.
 2010-07-15 10:27:44 -07:00
Peter Griess e7c4f8cdaa Buffer.copy() should liberally allow empty copies. 2010-07-15 10:20:42 -07:00
David Siegel 2085909aeb fix corner-case bug in Module 
and added a test to expose it
 2010-07-13 17:15:10 -07:00
Dmitriy Shalashov 70a8fb3763 Fix headers with empty value. 2010-07-07 18:46:24 +02:00
Ryan Dahl 515dc2161b Experimental: 'on' as alias to 'addListener' 2010-07-03 09:30:32 +02:00
Ryan Dahl 82ce3481fe Fix undefined_reference_in_new_context test 2010-07-03 08:16:36 +02:00
Ryan Dahl 8a52fb7aeb Revert "Fix 'uncaughtException' for top level exceptions" 
This reverts commit 8f8dcf8ed6.
 2010-07-01 11:10:22 -07:00
Ryan Dahl 8f8dcf8ed6 Fix 'uncaughtException' for top level exceptions 
Done by not evaluating the code in the first tick.

This breaks one test in test-error-reporting.js but I believe this to be a
V8 error and I have reported it in
http://code.google.com/p/v8/issues/detail?id=764
 2010-06-30 15:04:40 -07:00
Blake Mizerany 8c8534046c fix whitespace errors 2010-06-29 23:59:24 -07:00
Dmitry Baranovskiy f8ca6b383c Rewrote QueryString.parse to make it smaller and more effective. 
Also added ability to parse foo.bar=4 equal to foo[bar]=4
Added tests for this as well
 2010-06-29 23:52:42 -07:00
Ryan Dahl 5e86d01385 Revert "Buffer.copy should copy through sourceEnd, as specified." 
This reverts commit a2f70da4c9.

Keep tests modifies a few edge checks on Copy()
 2010-06-29 19:40:20 -07:00
Ruben Rodriguez 242161bef2 Added new API to Script, and implemented it in the REPL 2010-06-28 15:16:26 -07:00
Matt Ranney a2f70da4c9 Buffer.copy should copy through sourceEnd, as specified. 
Improve test-buffer.js to cover all copy error cases.

Fix off by one error in string_decoder.
 2010-06-25 09:10:49 -07:00
Ryan Dahl ba792ea202 :%s/sys.puts/console.log/g 
and there was much rejoicing
 2010-06-23 20:05:29 -07:00
Ryan Dahl f541043618 Remove ini.js 2010-06-22 19:37:29 -07:00
Mikeal Rogers f62979da6b Add sys.pump 2010-06-22 17:58:09 -07:00
Marshall Culpepper ca35ba640a Initial implementation of process.execPath 
Darwin, Linux, Solaris. FreeBSD still needs testing.

TODO: Amend the tests where we use spawn with argv[0].
 2010-06-22 02:02:25 -07:00
Peter Griess 51bd1b4483 Only concatenate some incoming HTTP headers. 
- Concatenate 'accept', 'accept-charset', 'accept-encoding',
  'accept-language', 'connection', 'cookie', and 'x-*' headers.
- For all others, drop duplicates.
 2010-06-21 12:01:50 -07:00
Ryan Dahl cb79614f73 Don't encourage strange ideas about req race conditions 2010-06-20 11:56:48 -07:00
Ryan Dahl eda21cccb4 Fix race condition in test-http-exceptions.js 2010-06-19 23:13:28 -07:00
Ryan Dahl b93faf9afa Comment out lines with full paths in message test. 
Note that * in the .out files is a special wildcard.
 2010-06-19 22:51:40 -07:00
Ryan Dahl f1a4f17897 script.runIn*Context not throwing errors properly. 2010-06-19 19:31:09 -07:00
Ryan Dahl 7d0252eae2 Fix message tests 2010-06-19 17:41:34 -07:00
Ryan Dahl 125b868e06 Add message tests 2010-06-18 18:14:06 -07:00
Micheil Smith 70c3e661b8 test-child-process-custom-fds to cleanup after running 2010-06-18 10:39:17 -07:00
Micheil Smith e0d30b545a Renaming tcp tests to net tests 2010-06-18 10:34:56 -07:00
Micheil Smith 3f48276bf9 Coding style changes, added tests in for the md5, sha256, etc methods 2010-06-18 10:21:38 -07:00
Peter Griess 187f191558 Support 'upgrade' event in HTTP client. 
- Add a unit test for client HTTP upgrade.
- Move around unit tests for server HTTP upgrade.
 2010-06-16 11:27:56 -07:00
Ryan Dahl cbf2a2233e Add callback to dgramSocket.send() 2010-06-16 11:03:34 -07:00
Ryan Dahl cf86527a28 Fix writeFile(buffer) test 2010-06-16 10:28:46 -07:00
Micheil Smith b05daaaf69 Adding path.existsSync (with tests.) 2010-06-16 10:26:11 -07:00
Aaron Heckmann 1d088fb906 fs.writeFile accepts Buffers 2010-06-15 23:58:32 -07:00
isaacs 4ce100fa62 A replacement for decodeURIComponent that doesn't throw. 
And add a few more tests.
 2010-06-15 19:28:09 -07:00
Ryan Dahl 6bed15e074 Refactor: Utf8Decoder -> StringDecoder 
Instead of just decoding Utf8, this will proxy requests to buffer.toString()
for other encodings. This makes for a simpler interface.
 2010-06-15 18:19:27 -07:00
Ryan Dahl 14414f81f3 Soft deprecation of 'listening' event. 
Add callback param to listen() instead
 2010-06-15 12:38:25 -07:00
Peter Griess c9dde726c6 Test case for net.Server.listenFD() 2010-06-15 11:44:33 -07:00
Ryan Dahl 009b4569b3 Merge Hash.init() function into JS constructor 2010-06-14 13:10:23 -07:00
Felix Geisendörfer 5f3906b65e Fix: fs.writeFile could not handle utf8 2010-06-14 10:58:17 -07:00
Paul Querna 02da5ed4a1 Implement datagram sockets 
- Adds new dgram module, for all data-gram type transports
- Supports both UDP client and servers
- Supports Unix Daemon sockets in DGRAM mode too (think syslog)
- Uses a shared Buffer and slices that as needed to be reasonably
  performant.
- One supplied test program so far, test-dgram-pingpong
- Passes test cases on osx 10.6 and ubuntu 9.10u
 2010-06-12 02:41:45 -07:00
Ryan Dahl fe89caba2e Improve fs.write tests 2010-06-10 15:17:46 -07:00
Ryan Dahl a6942b345d Fix test-repl 2010-06-07 17:39:52 -07:00
Peter Griess 8f0b4e9111 Send and receive file descriptors through net.Stream. 
a) create a layer of indirection in net.Stream to allow swapping in
different read/write implementations and

b) emit an 'fd' event when file descriptors are received over a UNIX pipe,
as finally as a tangential benefit

c) remove a bunch of conditionals from the primary codepaths for
ease-of-reading.
 2010-06-02 18:16:53 -07:00
Ryan Dahl 55a6f01732 Server responses should have a body if HEAD request 2010-06-02 15:07:15 -07:00
Orlando Vazquez 92da636b97 Add a parameter to spawn() that sets the child's stdio file descriptors. 
After the child is forked, these file descriptors will get dup2()'d to STDIN,
STDIO, and STDERR.

(API may be changed.)
 2010-06-02 10:41:24 -07:00
Ryan Dahl 1a5acd9850 API: readFileSync without encoding argument now returns a Buffer 
Correctly load utf8 data; add a test test-fs-read-file-sync.js
 2010-05-29 13:38:00 -07:00
Ryan Dahl 80a8e71fe0 Insert some hot paths into HTTP 2010-05-27 20:43:56 -07:00
Ryan Dahl 74b7fa29a1 Refactor HTTP 
Allow throttling from outgoing messages.
 2010-05-27 20:41:57 -07:00
Ryan Dahl 0e713e7482 :%s/8000/PORT/g for a few tests 2010-05-27 20:41:57 -07:00
Ryan Dahl 8840ce340d Bugfix: fs.ReadStream.setEncoding 
Improve test-fs-read-stream test.
 2010-05-26 18:33:38 -07:00
Sam Hughes 895f89d62a Avoided sending empty chunkedEncoding chunks in the middle of http responses 2010-05-25 22:58:17 -07:00
Ryan Dahl 15ec99ec59 Fix issue 89, parsing responses to HEAD requests 
Test from Mark Hansen (mark at markhansen.co.nz)
 2010-05-25 19:25:40 -07:00
Ryan Dahl 19f475c573 fs.ReadStream should emit Buffers 
And do proper utf8 encoding.
 2010-05-24 15:47:42 -07:00
Aapo Laitinen c4876d01a1 Change the behavior of path.extname 
Make path.extname return an empty string also if:
- The last dot is not in the last path component
- The last dot starts the last path component
 2010-05-24 09:56:28 -07:00
Mikeal Rogers fe3d8f2411 Add assert.ifError 2010-05-21 12:06:12 -07:00
Felix Geisendörfer c93e0aaf06 Deprecate string interface for fs.read() 
This patch makes buffers the preferred output for fs.read() and
fs.readSync(). The old string interface is still supported by
converting buffers to strings dynamically. This allows to remove the
C++ code for string handling which is also part of this patch.
 2010-05-20 16:31:18 -07:00
Ryan Dahl e84395ff8c Revert "Deprecate string interface for fs.read()" 
This reverts commit cbbf9e43d1.
 2010-05-20 16:11:33 -07:00
Felix Geisendörfer cbbf9e43d1 Deprecate string interface for fs.read() 
This patch makes buffers the preferred output for fs.read() and
fs.readSync(). The old string interface is still supported by
converting buffers to strings dynamically. This allows to remove the
C++ code for string handling which is also part of this patch.
 2010-05-20 15:25:49 -07:00
Ryan Dahl cbd2c3945b Throwing in a callback should kill the process 
There is a difference between errors which happen to a socket - like
receiving EPIPE - an exceptional situation but ultimately okay and the
situation where code throws in a callback - which is not okay.

Fixes test/simple/test-http-exceptions.js

TODO: explain this in docs.
 2010-05-20 15:23:36 -07:00
rentzsch 36a45c4e0d FIX path.dirname('/tmp') => '/'. 
Previously path.dirname('/tmp') incorrectly returned '.'.

Unfortunately module.js incorrectly thinks dirname('/a/b/') should
yield '/a/b', so I can't strip trailing slashes yet. Once module.js
is fixed, then the commented-out code should be activated and a test
written for it.
 2010-05-20 10:16:09 -07:00
Ryan Dahl d8af9d8edf Minor style changes for fs.write code 2010-05-19 14:59:03 -07:00
Felix Geisendörfer c46cbe0de4 Deprecate string interface for fs.write() 
This patch makes buffers the preferred input for fs.write() and
fs.writeSync(). The old string interface is still supported by
converting strings to buffers dynamically. This allows to remove the
C++ code for string handling which is also part of this patch.
 2010-05-19 12:53:43 -07:00
Peter Griess 2420f07e94 Allow process.setuid() and process.setgid() to accept string names in lieu of numeric IDs 2010-05-18 17:41:10 -07:00
Felix Geisendörfer caba9c70c3 Implemented Utf8Decoder module 
Allows to safely decode a utf8 stream into strings without breaking
on multibyte characters.
 2010-05-17 22:12:30 -07:00
Ryan Dahl e232f09d38 fs.WriteStream.write should support buffer 
Also re-adding the callback parameter.
 2010-05-17 19:44:14 -07:00
Felix Geisendörfer 0a539865dd Support arrays and strings in buffer constructor 
This is be very useful for testing code that deals with buffers.
 2010-05-17 11:19:14 -07:00
visionmedia 45948e054d Added better error messages for async and sync fs calls with paths 2010-05-14 19:46:16 -07:00
Jonas Pfenniger fea6f829bf FIX: EventEmitter#removeListener logic 
In the case of one defined listener, the function should only remove it
if it is the same as the passed listener.
 2010-05-14 19:13:50 -07:00
Jonas Pfenniger 4274e6b7d0 FIX: signal-unregister bug 
We now restore the default signal behavior when all listeners are
removed for a given signal.

Re-enabled test-signal-unregister
 2010-05-14 19:05:19 -07:00
Andrew Johnston 9599607065 Added posix fsync and fdatasync to fs module 2010-05-14 18:42:22 -07:00
Matt Ranney cd4f94e71a Add test for exceptions in HTTP parser event handlers. 2010-05-14 10:16:55 -07:00
Elijah Insua 976926376d Handle empty files with fs.readFile 2010-05-13 12:42:17 -07:00
Ben Noordhuis c6c77d535f Support for fstat(2) through fs.fstat() and fs.fstatSync(). Added test cases, updated documentation. 2010-05-12 14:20:43 -07:00
Ryan Dahl 67eeedd602 Disable test-signal-unregister.js as functionality doesn't exist yet 2010-05-12 12:45:53 -07:00
Ryan Dahl 7a2e6d674a Default to 2 second timeout for http servers 
Taking a performance hit on 'hello world' benchmark by enabling this by
default, but I think it's worth it. Hopefully we can improve performance by
resetting the timeout less often - ideally a 'hello world' benchmark would
only touch the one timer once - if it runs in less than 2 seconds. The rest
should be just link list manipulations.
 2010-05-12 11:49:28 -07:00
Ryan Dahl d2cff34fa3 Idle timeout changes 
- setTimeout should active the timeout too. (test-net-set-timeout tests
  this.)

- 'timeout' event is not automatically followed by an 'error' event. That
  is the user is now responsible for destroying the stream if there is an
  idle timeout.
 2010-05-12 10:06:13 -07:00
Ryan Dahl 15bb392afd Disable test-remote-module-loading 
In ab068db9b1 this test was broken because (I
think) compile/run errors are set to crash the program instead of being
passed back.

Error reporting is more important than remote loading. Disabling until there
is a fix
 2010-05-09 18:46:04 -07:00
Ryan Dahl 1ecb11f944 Add failing test - http responses not all complete 
Report and test by soderblom.peter@gmail.com
 2010-05-09 15:08:54 -07:00
Ryan Dahl ab068db9b1 Improve error reporting 
- No more single line "node.js:176:9" errors
- No more strange output when error happens on first line due to
  module wrapper function.
- A few tests to check these things
 2010-05-09 13:55:42 -07:00
Ryan Dahl d7762dffe3 Forgot to add child_process_should_emit_error.js 2010-05-08 23:24:24 -07:00
Ryan Dahl 264e540d00 Fix error reporting in child_process callbacks 
Issue 120, test case by Nathan Ostgard
 2010-05-08 22:13:34 -07:00
Tim Caswell c07c601c3b fs.readFile uses a buffer internally and will return the raw buffer if no encoding is specified. 2010-05-06 22:22:14 -07:00
Thomas Lee 3674563dd5 Fix a bug in http.Client where parsers may be prematurely released back to the free pool. 2010-05-06 00:44:07 -07:00
Ryan Dahl 1415650463 Use test port instead of 8000 in test-http-304 2010-05-04 22:06:46 -07:00
Rhys Jones 1a317135c5 Additional OpenSSL tests 2010-05-03 16:15:06 -07:00
Ryan Dahl b5bdf9434b Add another http upgrade test 2010-05-03 11:23:36 -07:00
Micheil Smith aee226b243 Fixing the 1 byte off error in http.Server's upgradeHead 
(And fix up some style issues)
 2010-05-01 18:05:53 -07:00
Micheil Smith f990f24ad3 Implementing the event emit for Upgrade requests, with tests. 2010-05-01 11:52:40 -07:00
Ryan Dahl 301b44d45d Chunk strings together on Stream buffer 2010-04-29 12:00:37 -07:00
Thomas Lee a63ce5cd72 Allow callers to disable PHP/Rails style parameter mungeing in querystring.stringify 2010-04-28 16:18:48 -07:00
Felix Geisendörfer f8a3cf980f Properly handle child process exit codes 
The child process 'exit' was returning the status of the process, rather than
the exit code. This patch properly deconstructs the status into the exit code
and the term signal a process may have received.

See:
http://pod.tst.eu/http://cvs.schmorp.de/libev/ev.pod#Watcher_Specific_Functions_and_Data_-5
and waitpid(2)
 2010-04-28 13:54:17 -07:00
Felix Geisendörfer c143cba590 Get rid of coupling for stdin 2010-04-27 18:02:30 -07:00
Felix Geisendörfer 76f98c27de Get rid of coupling for stdout 2010-04-27 17:27:34 -07:00
Ryan Dahl f618551694 Remove callback argument to FileWriteStream.prototype.write 2010-04-27 17:26:10 -07:00
Ryan Dahl c7b3c1ecbc Add more asserts to test-http-big-proxy-responses 2010-04-26 20:26:36 -07:00
Jérémy Lal 0802326fc9 test-dns so that NXDOMAIN does not rise errors 2010-04-26 15:09:03 -07:00
Jérémy Lal 5caf53a053 Add support for MX queries in dns module, works with c-ares upstream patch. 2010-04-26 15:09:03 -07:00
Jonas Pfenniger df07a713bd New test-case: signal unregistration 
The default signal-handler should be restored if no handlers are
assigned.
 2010-04-25 23:14:07 -07:00
Ryan Dahl 6abdf051d3 Unhandled 'error' events kill program 
By default 'error' throws. addListener to prevent this.
 2010-04-22 17:22:06 -07:00