webdevRefinery Forum: Cyril - Viewing Profile - webdevRefinery Forum

Hey guys.
So, yesterday I pretty much decided to say "fuck it" and buy a gaming rig (parts ordered, receiving this weekend).

Part of that was 16GB of Kingston HyperX Beast (overkill, I know, I've been told, but eh). Now, I'm obviously not going to be using all of that; max is probably going to be half if I'm running a virtual machine or something. In any case, that leaves me with a ton of RAM, and I was thinking of using it as a RAM Drive.

For those who don't know what that means, it's mounting some of your RAM as a "fake" drive. It's basically like an SSD but faster. It's RAM, so it's most probably going to be for temporary files (did some research; most people redirect application caches to them to avoid writing to the SSD).

Opinions? Anybody done this before, or have suggestions?

Yes, I realize this is most probably overkill, not particularly useful and ya da da, but yeah.

<3

I have the habit of never commenting anything I write, so I was wondering, what's the standard / correct / best way to write documentation?
I'm referring to the comments you'd find before a function, which describes what it does, etc. Do you also have any specific habits or tips regarding this?

This is also in General Programming, so is there anything specific to languages? I mainly use PHP and JavaScript.

Some of you guys might remember a project I had some time ago; iceberg. Well, while it worked okay, the code was horrific, so I decided to rewrite the whole thing from scratch.

One thing I want to change is the hook system. The way I had it set up before was to have a folder containing PHP files, each one named after the hook it represented. For example, you'd have a file called

PreGenerateHook.php

which would be included and called by the script (hook autoloader thing).

I want to have something more flexible this time, but I'm a little lost on how to do it. I've looked into things such as the observer pattern, but to be honest, I'm not too sure how it'd work out. Here's essentially what I want it to have;

• Must be able to be called by the script. For example, I should be able to do something such as

  Hooks::run("pre-generate")

  and it should work.
  
• Naming of hooks. For example, I could have a "post-generate" hook called (unique name)

  upload-files

  .
  
• Multiple hooks set for the same "event" (let's say, I want to have

  upload-files

  and

  compress-images

  hooks, both on the

  post-generate

  event.
  
• Calling hooks by name should work as well (well, that's the point of naming them).

I've considered things such as scanning the folder at the start of the script, and load all the hooks, but that sounds pretty inefficient. Having it with the system I was using previously, with PHP files being loaded using the hook event won't enable flexibility with the names either.
Anybody got some other ideas?

Not much to say about it, but I'd like some feedback
I decided to go really simply with this one, and I'm pretty damn happy with the result.

http://cyril.so