In fact I did simply use
libraries: [path/to/dll_or_lib],
but it gave me the same error. In linux i was able to link it successfully
with .so file with the line: libraries: [path/to/dll_or_lib.so],
Don't know why this is not working in windows!
--
--
Job Board: http://jobs.nodejs.org/
That's great news. Thanks!
On Tuesday, February 25, 2014 6:49:15 PM UTC+1, Jimmy Guerrero wrote:
Hello,
Whether you are new to debugging Node apps or have used Node Inspector in
the past, it's time to check out what's new in the latest v0.7 release of
Node Inspector!
It's been a big
Also note that fs-ext (on npm) implements flock() which is required in many
circumstances too.
On Wed, Feb 26, 2014 at 12:20 AM, Andrew Kelley superjo...@gmail.comwrote:
Actually, there is a solution! fs.link to the new file. This will fail
with EEXIST if it exists. And then fs.unlink on the
If node is single threaded and I can only be making one database call at
any time, what is the point of having a pool of db connnections?
--
--
Job Board: http://jobs.nodejs.org/
Posting guidelines:
https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines
You received this message
JavaScript in Node is single thread. But async I/O is derived to LibUv,
that uses threads. Usually, a DB driver calls the database server, using
sockets. Then, that communication will be handle by LibUv, returning the
results in a callback. Your single JS threads is available for other tasks
makes perfect sense, thank you
Regards,
Reza
On Wed, Feb 26, 2014 at 9:16 AM, Angel Java Lopez ajlopez2...@gmail.comwrote:
JavaScript in Node is single thread. But async I/O is derived to LibUv,
that uses threads. Usually, a DB driver calls the database server, using
sockets. Then, that
I am writing a GPU database and looking at using javascript as the language
to query data and node.js looks ideal for sending and receiving binary
data. I have been writing a node addon as I have written it in C++. However
I have problem with my node.js addon as my c++ objects are not being
But unfortunately isn't how it works at all, unless the database driver in
question is using C libraries which use the thread pool (which most of them
don't these days).
It's all about the event loop. Here's how an event loop works:
http://baudehlo.com/2013/02/14/how-an-event-loop-works/
On
Ummm... You can write a driver in JS code that use require('net'). And in
that module, the socket I/O is, at the end, managed by LibUv. The same for
require('fs') for file I/O.
Am I right?
And yes, the driver could be implemented in C, and then, it should call
LibUv in some way.
Angel Java
Some more concrete example, to support my assertion, and to have a clear
picture of what can be used.
I just found:
https://github.com/felixge/node-mysql
It manage a pool:
https://github.com/felixge/node-mysql/blob/master/lib/Pool.js
A pool manage a list of available Connection
Hi all,
On NPM, I have had a few packages that I no longer maintain, and have
not updated for the last 3 years. These packages either no longer work,
or have been superseded by far better packages.
I have published a deprecation notice using `npm deprecate` on the
following packages:
Matt,
just so I understand, you are saying if a node package such as mongodb
provides connection pooling, they have to be implementing the package using
a pooling system that is implemented in C?
On Wednesday, February 26, 2014 11:50:18 AM UTC-8, ajlopez wrote:
Some more concrete example,
This sounds like a very good idea... I wonder, is there any downside to it? Performance maybe? 26.02.2014, 09:20, "Andrew Kelley" superjo...@gmail.com:Actually, there is a solution! fs.link to the new file. This will fail with EEXIST if it exists. And then fs.unlink on the old file. Everything
Some databases (MySQL for example) don't allow you to send next query
before receiving result of a previous one.
The only way to have queries run in parallel is have each query in separate
connection.
Even when query is very short it's a good idea to use connection pool to
overcome network
I am processing an incoming request that may or may not have post data.
I found that if I see a true value for req._readableState.ended, that the
request is ended and there is no post data.
However, this doesn’t seem like the best way to do it.
Is there a better way to determine if I have data
There is no performance downside in the failure mode. But obviously in the
success mode there's two function calls vs one. But realistically the overhead
of file system operations on spinning disks will far outweigh the extra API
calls.
On Feb 26, 2014, at 7:40 PM, Alex Kocharin
16 matches
Mail list logo
