Re: [nodejs] Re: TJ Holowaychuk is leaving the community and now

2014-07-19 Thread Phoscur
Aren't we just speeding on the work of the V8 team? NodeJS just leverages a low level API to that beast (and that's why there are callbacks, they are low level function blocks in javascript. So it's not about performance and Callbacks.. Where exactly are the usability leaks? TJ thank you for your

Re: [nodejs] Inconsistent method names?

2012-12-28 Thread Phoscur
What's the harm in adding this alias to node? I remember myself running into this (as not everyone knows posix calls by heart) and posting on this list. Am 27.12.2012 22:51, schrieb Isaac Schlueter: On Thu, Dec 27, 2012 at 1:03 PM, Jonathan Chayce Dickinson wrote: fs.prototype.readDir = func

[nodejs] Full HD Video Streaming

2012-10-31 Thread Phoscur
Hi Guys, I have a small hobby project to create a nice album for my films I have stored on my computer (and servers). Besides loading information from the internet (such as from IMDb) into a nice HTML UI, I'd like to be able stream the films. My question is now, is it even possible yet? Witho

Re: [nodejs] Re: Best practices for sharing code and data between the server and client

2012-08-13 Thread Phoscur
Server not found Am 13.08.2012 09:11, schrieb greelgorke: to the topic: take a look at derbyjs.org :) Am Samstag, 11. August 2012 18:21:20 UTC+2 schrieb shoshy: Nils, Thanks a lot for your quick reply. I'll test it, i did see the entry_validation, but didn't know if backbone l

Re: [nodejs] Re: nodeconf videos?

2012-07-04 Thread Phoscur
I haven't even seen the last nodeconf videos, because I forgot about them after a similar thread like this. Maybe anyone collected links? Am 04.07.2012 17:01, schrieb dhruvbird: :-( I was hoping to catch the node-mysql live coding - heard that it was a super-session! I hope they don't edit too

Re: [nodejs] Version 0.7.12 (last one before 0.8.0)

2012-06-19 Thread Phoscur
Haven't seen any discussions on domains lately, they are in .7.12, so they are going to be in 0.8? Anyone using them yet? Am 20.06.2012 01:31, schrieb Isaac Schlueter: 2012.06.19, Version 0.7.12 (unstable) This is the last release on the 0.7 branch. Version 0.8.0 will be released some time

Re: [nodejs] requireindex

2012-05-07 Thread Phoscur
Could this be made default behavior for requiring folders without index.js? Am 04.05.2012 04:19, schrieb Stephen Handley: New package to allow writing minimal node index.js/index.coffee files that require and export siblings by file basename. Would appreciate any feedback... I assumed there was

Re: [nodejs] OOP Sample

2012-03-30 Thread Phoscur
JavaScript is objectoriented and functional, node and its API are mostly functional. For a simple project, you wont need OOP at all (that would be development overhead). When you have a complex business logic and want expandability, for that part it might be a good idea to chose an objectorient

Re: [nodejs] Re: garbage collection, patterns: classical/prototypal inheritance and ValueObject

2012-03-11 Thread Phoscur
> I suggest you look at how util.inherits is implemented, then use that > for inheritance. I know that is a better idea, which keeps constructor and super_ properties, but it hides the fact that in JavaScript you use Object.create() even for classical inheritance. > With regards to your code, th

Re: [nodejs] Re: garbage collection, patterns: classical/prototypal inheritance and ValueObject

2012-03-10 Thread Phoscur
l the old ones (e.g. with > obj.__proto__.__proto__.__proto__), so since it's accessible, it will > not and must not be garbage collected, so, yes, effectively that will > leak. > > On Mar 10, 5:27 pm, Phoscur wrote: >> Considering that a long chain of prototypes is creat

Re: [nodejs] garbage collection, patterns: classical/prototypal inheritance and ValueObject

2012-03-10 Thread Phoscur
same point after they get "overshadowed" by the newer states, or would it become a memory leak? Am 10.03.2012 22:06, schrieb Marcel Laverdet: > If you can write it in JS, v8 can garbage collect it. > > On Sat, Mar 10, 2012 at 2:46 PM, Phoscur <mailto:phos...@pheelgood

[nodejs] garbage collection, patterns: classical/prototypal inheritance and ValueObject

2012-03-10 Thread Phoscur
Short Question: Do V8 and Jägermonkey garbage collect prototype chains at some point? Let code speak: https://gist.github.com/e83c353f7f16e14e4333 Actually, while writing this, I found the solution (line 69/70), but I would like to leave it without that, as forgetting it may produce a memory leak

Re: [nodejs] Will Node.js work for me?

2012-03-04 Thread Phoscur
> I'm trying to figure out if Node will work for our platform or not (or > even V8 in general). We're planning a rewrite. We use PHP. We need to > allow users to write server side plugins. For it to be secure in PHP > we'd have to write a "language" using flex/bison to eliminate dynamic > function

Re: [nodejs] use fs.watch instead

2012-03-03 Thread Phoscur
when you use fs.watchFile? Maybe it should be marked as deprecated. Am 03.03.2012 22:16, schrieb Phoscur: > This is not uncommon when you try to use watch functionality of some > module on windows. > Is there a simple workaround to get this without patching the module? I > mean fs.watc

[nodejs] use fs.watch instead

2012-03-03 Thread Phoscur
This is not uncommon when you try to use watch functionality of some module on windows. Is there a simple workaround to get this without patching the module? I mean fs.watch works fine on windows, but apparently the old linux notation is not supported. -- Job Board: http://jobs.nodejs.org/ Postin

Re: [nodejs] why this code memory leak

2012-02-29 Thread Phoscur
> 9 > 10 > 11 > 12 > 13 > 14 > 15 > 16 > > setInterval(function(){ > > someExpensiveOperation(); > }, 1); > What happens here if the operation takes longer than one millisecond? setInterval will continue to add you operation onto the eventloop while the execution lags behind -> eventloop i

Re: [nodejs] Re: enable/disable assertion

2012-02-28 Thread Phoscur
How is this better than: function debug(msg) { if (DEBUG_ON) { console.log(msg); } } ? Am 28.02.2012 23:55, schrieb JeanHuguesRobert: > de = false // true when debugging > de&&mand( cond ) // ie "demand" instead of "assert" > de&&bug( msg) // ie "debug" to output trace msg > > On 27

Re: [nodejs] Re: JSLint or not?

2012-02-20 Thread Phoscur
Crockford himself writes "JSLint will hurt your feelings". We don't need to discuss what style is the best, it's the same as discussing "the best programming language". You can't fight about anyone's taste. JSLint is especially good if you have a problem in you code and can't find it. JSLint will

Re: [nodejs] Re: Node version 0.6.11 (stable)

2012-02-18 Thread Phoscur
Am 18.02.2012 12:51, schrieb Jann Horn: > 2012/2/18 Phoscur : >> I think windows integration of node atm is actually better than on ubuntu! > You call "please compile it yourself" badly integrated? :D No, but that you have a to open a new console after installing som

Re: [nodejs] Re: Node version 0.6.11 (stable)

2012-02-17 Thread Phoscur
First I was a little disapointed by the installer, as it leaves you no options at all, you can only install to Program Files(x86). But as I had to install node on ubuntu yesterday and apt-get just didn't get me an up to date package of npm (v0.2) wtf? Then went to download the binaries and had to w

Re: [nodejs] Fork and patch npm modules

2012-02-14 Thread Phoscur
that - I think someone recently patched > it, not sure. > >> Am 14.02.2012 02:50 schrieb "Phoscur" > <mailto:phos...@pheelgood.net>>: >> >> Well it seems it's still somewhat unstable as the module I'm using >> does not work as expecte

Re: [nodejs] Fork and patch npm modules

2012-02-13 Thread Phoscur
Well it seems it's still somewhat unstable as the module I'm using does not work as expected (and not as without symlink). Could that be a problem in path.dirname or path. normalize? Am 14.02.2012 00:08, schrieb Phoscur: > Just updated my nodejs to be sure, from 6.8 to 6.10,

Re: [nodejs] Fork and patch npm modules

2012-02-13 Thread Phoscur
Just updated my nodejs to be sure, from 6.8 to 6.10, now it works.. Couldn't find an entry of this in the changelog.. Am 14.02.2012 00:00, schrieb Kevin Swiber: > On Mon, Feb 13, 2012 at 5:49 PM, Phoscur <mailto:phos...@pheelgood.net>> wrote: > > The object (even in t

Re: [nodejs] Fork and patch npm modules

2012-02-13 Thread Phoscur
The object (even in the REPL) I get is empty({}), it should be a function. It seems node does not follow windows symlinks. Am 13.02.2012 23:21, schrieb Kevin Swiber: > > On Mon, Feb 13, 2012 at 5:05 PM, Phoscur <mailto:phos...@pheelgood.net>> wrote: > > > &quo

Re: [nodejs] Fork and patch npm modules

2012-02-13 Thread Phoscur
> "npm link" >> How do I accomplish this on windows? > You're screwed, I guess. Oh man, but I want this!! Windows 7 is able to make symbolic links too "mklink", but it seems all I get back from a symlinked folder is an empty object..? Am 13.02.2012 21:11, s

[nodejs] Fork and patch npm modules

2012-02-13 Thread Phoscur
Hi, After reading this: http://www.debuggable.com/posts/how-to-fork-patch-npm-modules:4e2eb9f3-e584-44be-b1a9-3db7cbdd56cb I see there is a nice way to replace broken modules with your own patched ones. But this always involves git push and npm update commands when you change something. Isn't there

Re: [nodejs] Re: bundle javascript for reuse in the browser

2012-02-09 Thread Phoscur
ght dependency parsing (probably with help of UglifyJS) What does that mean? Am 10.02.2012 01:59, schrieb Phoscur: > Looks good! I'll give it a try! > > Am 10.02.2012 00:21, schrieb Mariusz Nowak: >> @Phoscur, give modules-webmake >> <https://github.com/medikoo/modules-web

Re: [nodejs] Re: bundle javascript for reuse in the browser

2012-02-09 Thread Phoscur
Looks good! I'll give it a try! Am 10.02.2012 00:21, schrieb Mariusz Nowak: > @Phoscur, give modules-webmake > <https://github.com/medikoo/modules-webmake> a try, with plain written > requires it will just work (no matter if it's same or external package > modul

Re: [nodejs] bundle javascript for reuse in the browser

2012-02-09 Thread Phoscur
hing like: > > require("someFile", function(someModule){ > > }); > > Browserify does not handle this use case at all. Yes, if you want to > concat and minify all your js into a single file and load that file > everywhere, sure, go right ahead, but you're doing it wrong

Re: [nodejs] bundle javascript for reuse in the browser

2012-02-09 Thread Phoscur
I think you haven't actually understood how browserify/browserbuild work: 1. load source files 2. wrap a function around them 3. concat all of them together and add a custom require() emulating node's require 4. instead of many script tags, insert one with the big file Now when moduleA requires mo

Re: [nodejs] bundle javascript for reuse in the browser

2012-02-09 Thread Phoscur
Then do not load code via ajax. Am 09.02.2012 22:18, schrieb Mark Hahn: > > What exactly the problem with using node's synchronous approach in > the browser? > > Node loads the module quickly from the disk. Loading over an ajax > request is much slower. > > -- > Job Board: http://jobs.nodejs.

Re: [nodejs] bundle javascript for reuse in the browser

2012-02-07 Thread Phoscur
d/event-key/event-key-min.js&yui3/build/dom-style/dom-style-min.js&yui3/build/node-style/node-style-min.js&yui3/build/transition/transition-min.js&yui3/build/cookie/cookie-min.js&yui3/build/intl/intl-min.js</a>>" > async > > > You could fairly easily do

[nodejs] bundle javascript for reuse in the browser

2012-02-06 Thread Phoscur
Splitting this from Build tool. What I want: -given entry.js, the tool has to resolve it's dependencies inspecting require, recursivly including all relevant files -conviently do jslint/hint checks as codestrings are already loaded -possibility to minify for production -filewatching and possibly c

Re: [nodejs] Re: Build tool

2012-02-05 Thread Phoscur
t; -- > Paolo Fragomeni > Co-founder, CTO > Nodejitsu, Inc. > www.twitter.com/hij1nx > www.github.com/hij1nx > > > > On Feb 2, 10:42 am, Phoscur wrote: >> Yes, both browserifyand browserbuild assemble a concatenated file with >> all files used

Re: [nodejs] [ANN] JavaScript Jabber Podcast

2012-02-05 Thread Phoscur
Hi, I just heard the cast, and I was quite happy that you are talking about a theme I'm still struggeling with: build tools (continue the discussion on this on this topic: https://groups.google.com/group/nodejs/browse_thread/thread/9814da6a8873db2f/fdfaefa8ca5db7aa ) First off: Please check the ca

Re: [nodejs] Re: Build tool

2012-02-02 Thread Phoscur
(so this is indeed the dependency resolving i ment). Am 02.02.2012 06:51, schrieb hij1nx: > do any of these do automatic dependency resolving? > > > On Feb 1, 12:23 am, Phoscur wrote: >> Great, an alternative to browserify! This is perfect. Now I can choose a >> build proc

Re: [nodejs] Re: multiple version of node on windows

2012-02-01 Thread Phoscur
sion? > > On Wed, Feb 1, 2012 at 7:27 AM, Bert Belder wrote: >> On Feb 1, 8:44 am, Phoscur wrote: >>> Some hours ago I installed node v0.7.1 because of the path.exists bug, >>> now I can't install another module because it's dependency explictly >>> doe

[nodejs] multiple version of node on windows

2012-01-31 Thread Phoscur
Some hours ago I installed node v0.7.1 because of the path.exists bug, now I can't install another module because it's dependency explictly doesn't support node >0.7. On *nix I know there is nvm to manage multiple node versions, but what can I use on windows? The installer won't even let me downgra

Re: [nodejs] Build tool

2012-01-31 Thread Phoscur
Great, an alternative to browserify! This is perfect. Now I can choose a build process to combine this with. Thanks a lot! So far I've seen cake, grunt and codesurgeon for the build process and enderjs, requirejs, pakmanager, browserify and browserbuild to port node modules to the browserside. Th

Re: [nodejs] Build tool

2012-01-31 Thread Phoscur
Looks like another solid build tool. It doesn't meet all my requirements though: > -browser-side require() replacement > -automatic dependency resolving and including files > and the ability to run on filechanges. 31.01.2012 23:28, Tim Branyen wrote: > Check out: > > http://github.com/cowboy/grun

Re: [nodejs] Build tool

2012-01-31 Thread Phoscur
> Your directory structure might look like this: > > - project > - browser > - server > Where do you put code, you use on both sides? I should be possible with just an entry point, say src/browser.js, to scan throw the whole source directories bundling all files, getting modules it can't find

Re: [nodejs] Build tool

2012-01-31 Thread Phoscur
; smart. > > RequireJS seems to be the most popular. I didn't use it because I > didn't understand AMD before I had written pakmanager. > > > UglifyJS and google closure can minify > > > Some of those other things - such as linting - you can just run in a > bash scr

Re: [nodejs] Re: Build tool

2012-01-31 Thread Phoscur
www.twitter.com/hij1nx > www.github.com/hij1nx > > > On Jan 31, 1:52 pm, deitch wrote: >> Been a long ongoing discussion about this >> athttp://groups.google.com/group/nodejs/browse_thread/thread/4f5270afb0... >> >> On Jan 31, 11:31 am, Phoscur wrote: >>

Re: [nodejs] Re: Build tool

2012-01-31 Thread Phoscur
eitch: > Been a long ongoing discussion about this at > http://groups.google.com/group/nodejs/browse_thread/thread/4f5270afb0de400f > > On Jan 31, 11:31 am, Phoscur wrote: >> So, this means, there are no other build tools out there? >> >> Am 30.01.2012 17:48, schrieb Phoscu