I vote that his project be named "bike shed". You could put it in npm as bshed.

Hi noders,

I'm the author of nodemock (https://github.com/arunoda/nodemock) project.
I would like to how many of you use it with your testing toolkit.
And any suggesstions for improvments also welcome here.

Thanks.

--
Arunoda Susiripala

Is it too late to rename it? Maybe "ropes." ;)
 
--Josh

How about noose . . . ?

- Jeff

Jeff Curtiss
(650) 762-5470
jeffrestore-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org

Group: http://groups.google.com/group/nodejs/topics

• js toolkit v1.1.0 released [1 Update]
• ANN: threads - enabling safe use of synchronous coding [10 Updates]
• New User Question - can't find module error [2 Updates]
• Node v0.5.5 [2 Updates]
• Chromium debugger in Eclipse [3 Updates]
• NPM install failure on FreeBSD [1 Update]
• OT: look-behind assertions [2 Updates]
• Has no method set Error [1 Update]
• Reading Client certificate - webid implementation [1 Update]
• node.js not compiled with openssl crypto support? [1 Update]

ollym <oliver.morgan <at> kohark.com> Aug 31 08:00PM -0700 ^

 
Just to mention i've released the next version of the toolkit with
better documentation and added function support.
 
Docs here: http://ollym.github.com/toolkit/
Github Page: https://github.com/ollym/toolkit
 
Feedback welcome as ever!


 

Isaac Schlueter <i <at> izs.me> Aug 31 01:44PM -0700 ^

 
>> already confusing enough.
 
> LOL, isn't this rather ironic, given it's built on child_process.fork(),
> which isn't fork(2) at all, and is really quite confusing?
 
Indeed. I recommended not calling it fork for that very reason.
Whatever, it's done :)
 
child_process.fork is a wrapper around child_process.spawn, with an
additional channel open to send and receive messages in a somewhat
webworker-like manner.
 
child_process.spawn uses vfork(2), which is quite similar to fork(2),
but uses virtual memory much more efficiently. Also, it is not safe
or advisable to use vfork(2) for any purpose other than an immediate
call to some form of exec(2).
 
Using fork+exec is not quite the same as merely forking and then
running some more code. Yes, it does create a thread, but in that
thread, it splits off a whole other child process, and that's all it
does.
 
So, yes, it *does* use fork, it just doesn't *only* use fork. The
thread itself is not exposed to JavaScript, and you never have two
bits of JavaScript code running in parallel in the same process. This
is the guarantee that Node makes which makes reentry not an issue we
have to even think about.
 
If you're writing C++ addons and using eio_custom or something, then
you do have to be aware of these issues.
 
>> it's absolutely not in any way "multithreaded".
 
> Computer scientists would probably disagree with you.
 
Well, I mean, of course it's using threads. You can't write software
on a CPU without "using threads", but it's not as if
child_process.fork results in a new thread in the same process that
can communicate with its parent by use of a shared memory space.
 
When *I* start complaining that you're being too hyperliteral and
robotic, you might have a problem. ;P


 

Mark Hahn <mark <at> boutiquing.com> Aug 31 02:08PM -0700 ^

 
Many more people than computer scientists could benefit from this and
it would be a shame if they didn't look into it because of its name.
 


 

Joshua Holbrook <josh.holbrook-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 31 02:19PM -0700 ^

 
Is it too late to rename it? Maybe "ropes." ;)
 
--Josh
 


 

rtweed <rob.tweed <at> gmail.com> Aug 31 02:33PM -0700 ^

 
I really dont care what you want me to call it. I spent a few minutes
looking for a name that was vaguely related to its architecture and
came up with threads and found the name wasn't registered in npm so
went with it. I mainly wanted to get it out as soon as possible
because I have specific plans for its use.
 
How about you guys vote for a name and I'll rename it accordingly
 
Rob
 


 

Joshua Holbrook <josh.holbrook-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 31 02:41PM -0700 ^

 

> How about you guys vote for a name and I'll rename it accordingly
 
> Rob
 
This sounds fun! Maybe we (you?) can collect a short list of possible
names, and then have an online poll to decide which wins. Maybe
running it from threads' issues page
(https://github.com/robtweed/threads/issues) would be the way to go.
 
--Josh


 

Mark Hahn <mark <at> boutiquing.com> Aug 31 03:37PM -0700 ^

 
It doesn't matter what it is called as long as it isn't confusing.
Marketing studies for product names have shown that names unrelated to
function perform better in the market, hence Apple, Google, etc.
 
I like the "ropes" moniker.
 


 

Marcel Laverdet <marcel <at> laverdet.com> Aug 31 06:16PM -0500 ^

 
But it's not a rope either.
 
http://en.wikipedia.org/wiki/Rope_(computer_science)
 


 

Diogo Resende <dresende <at> thinkdigital.pt> Sep 01 12:19AM +0100 ^

 
On Wed, 31 Aug 2011 15:37:47 -0700, Mark Hahn wrote:
> to
> function perform better in the market, hence Apple, Google, etc.
 
> I like the "ropes" moniker.
 
How about spoon? It's not a fork..
 
---
Diogo R.


 

Zachary Carter <zack.carter-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 31 07:47PM -0400 ^

 

> How about spoon? It's not a fork..
 
> ---
> Diogo R.
 
Or maybe spork, as a play on spawn/fork.
 
 
--
Zach Carter


 

Marcel Laverdet <marcel <at> laverdet.com> Aug 31 08:59PM -0500 ^

 
I vote that his project be named "bike shed". You could put it in npm as
bshed.
 


 

SteveCronin <steve.d.cronin-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 31 10:22AM -0700 ^

 
Hello Folks!
 
Just joined - I'm a Mac user and have been dinking with node.js
(0.4.11) for ~ a day..
 
I have it installed and just root installed using npm 'node-ffi'
I now see node-ffi at /usr/lib/node_modules/npm/node_modules/node-
ffi
 
.bashrc looks like: export NODE_PATH="/usr/lib/node_modules:/Users/
steve/Documents/Node"
 
I also am using 'NodObjC'.
 
I've written a proof-of-understanding script which calls into NodObjC
but I'm getting this error:
 
node.js:134
throw e; // process.nextTick error, or 'error' event on first
tick
^
Error: Cannot find module 'node-ffi'
at Function._resolveFilename (module.js:317:11)
at Function._load (module.js:262:25)
at require (module.js:346:19)
at Object.<anonymous> (/Users/steve/Documents/Node/NodObjC/core.js:
6:11)
at Module._compile (module.js:402:26)
at Object..js (module.js:408:10)
at Module.load (module.js:334:31)
at Function._load (module.js:293:12)
at require (module.js:346:19)
at Object.<anonymous> (/Users/steve/Documents/Node/NodObjC/
import.js:7:12)
 
What am I missing here?
Steve


 

Nathan Rajlich <nathan <at> tootallnate.net> Aug 31 04:17PM -0700 ^

 
Hey there Steve,
 
NodObjC is still very much under development, and the initial setup is
not very streamlined at this point. If you come over to #nodobjc on
the Freenode IRC servers then I can help walk you through what you
need to do to setup.
 
(That goes for everyone out there too, I've registered #nodobjc so
head over there if you wanna talk NodeJS + Cocoa)
 


 

EllisGL <xellisx <at> gmail.com> Aug 31 03:23PM -0700 ^

 
Downloaded that fix, now http.listen doesn't exist.
 


 

EllisGL <xellisx <at> gmail.com> Aug 31 03:25PM -0700 ^

 
Using the code from the "hello world" front page example.
 
Server running at http://127.0.0.1:1337/
 
node.js:205
throw e; // process.nextTick error, or 'error' event on first
tick
^
Error: listen UNKNOWN
at errnoException (net_uv.js:564:11)
at Array.<anonymous> (net_uv.js:630:26)
at EventEmitter._tickCallback (node.js:197:26)
 


 

hsimon <simon.hunanyan <at> googlemail.com> Aug 31 05:02AM -0700 ^

 
Hi,
 
I updated on Eclipse 3.6 (Helious) the Cromium JavaScript Remote
Debuger up to latest version (0.3.0.201108190405). Unfortunately it
wants not to work at all (java.lang.NullPointerException).The node.js
(0.4.9) is running and listen on port 5858.
I don't understand what is the reason. I even tried to install the
clean Eclipse and did the setup here - the same result:(
Before, I have used Cromium JavaScript Remote Debuger (version
0.2.2.20110...) and it worked perfectly. but after update I got
dramatic situation. Can somebody help me and advise where could be a
reason and how it could be resolve.
 
Thank you,
 
Simon


 

hsimon <simon.hunanyan <at> googlemail.com> Aug 31 05:26AM -0700 ^

 
I have found one interesting dependency - when I am installing the
Chromium JavaScript Remote Debugger alone - it working properly, but
when I add VJET JavaScript IDE (https://www.ebayopensource.org/wiki/
display/VJET) debugger wants not work!!!
 
Simon


 

telamohn <telamohn <at> gmail.com> Aug 31 03:04PM -0700 ^

 
I suspect is has to do with the fact that VJET has it's own JavaScript
engine, Whereas node uses Google V8-engine.
You cannot run java script that depends on native Node.JS function calls or
libraries.
I'm not quite sure what VJET does but i guess it's not entirely impossible
for a JS-engine collision to occur in Eclipse.


 

Daniel <dbchristopher <at> gmail.com> Aug 30 03:57PM -0700 ^

 
So I've been trying to install NodeJS + NPM on my FreeBSD server for a
while. With some help from a post on superuser.com (http://
superuser.com/questions/262385/getting-node-js-0-4-2-to-build-install-
on-freebsd-8-2-amd64) I managed to get NodeJS 0.4.11 running. But now
trying to install NPM I'm not getting anywhere.
 
With the easy installation I get to "fetching: http://registry.npmjs.org/npm/-/npm-1.0.27.tgz"
and then it just stops doing anything. Likewise when installing from
source it gets to "node cli.js install -g -f" and then it hangs there
and doesn't progress.
 
I'd love it if you had any ideas for me!
 
Thanks,
Daniel


 

tjholowaychuk <tjholowayhuk-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 30 08:46PM -0700 ^

 
you can kinda fake it with the callback
 


 

Diogo Resende <dresende <at> thinkdigital.pt> Aug 31 10:27AM +0100 ^

 
On Tue, 30 Aug 2011 21:54:18 -0500, Felipe Gasper wrote:
> Forgive the OT post, but does anyone know why no JS engine has yet
> implemented look-behind assertions in regular expressions?
 
> -FG
 
You're talking about negative look-behind/ahead right? I don't use
positive
ones, they are easy to emulate without look-*. I actually asked myself
that
a while ago.. :P
 
---
Diogo R.


 

Richard Marr <richard.marr-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> Aug 31 08:45AM +0100 ^

 
>   res.simpleBody(200, "Hello Connect");
 
Could be a version problem... I think the "simpleBody" method was
removed about a year ago.
 
I'm on Node 0.4.7 and Connect 1.4.0 and your code snippet works fine
if I replace res.simpleBody(200, "Hello Connect") with res.end("Hello
Connect")
 
 
 
 
 
 
--
Richard Marr


 

Thomas Fritz <fritztho <at> gmail.com> Aug 31 08:58AM +0200 ^

 
There is another issue / merge request which adds more properties to
the peer certificate in node. https://github.com/joyent/node/pull/1286
When fixed and tested it seems that it is in the next version of node
(0.5.6 ??) - so it is "foaf+ssl ready" (tm)
 
In the meantime i came to this project which solves it in another way:
http://lists.foaf-project.org/pipermail/foaf-protocols/2011-May/004942.html
 
Regards
 
 
 
---
Thomas FRITZ
web http://fritzthomas.com
twitter http://twitter.com/thomasf
 
 
 


 

tre <tre.theresia <at> gmail.com> Aug 31 04:18AM ^

 
thanks Nate, it works :D