CI Proxy Library - Browser simulator (REST, Cookies and Proxy Support, PURE PHP) [NEW] HTTP Authorization

[eluser]toopay[/eluser]
Hi all.

It's been a while, since i use CI in my (commercial) project. What i like from it, beside other framework (i use several framework, i have 2 favorite : Kohana and CI ), its had enough helper and library to make developer life easier.

When i develop some project, i need to have a simple way to call another controller, or even an external resource (ex. api from google, or iptocountry@sourceforge). I create a simple cURL class to do that. This is the (almost) stable version of that, this enough for my needs. I’ll take it to higher level, as soon as i had a long holiday from my work.

It have several features, which might help you during development (or even in live web) process.

Here's the way you use it

Code:

// Simple way to use this library
// In any controller, put this line...
...
$this->load->library('proxy');
$this->proxy->site('http://codeigniter.com',TRUE);
...

// Above example will give you rendered page of CodeIgniter site, if you didn't want
// to render it directly, or it was a json which you want to save to var, simply do this...
...
$this->load->library('proxy');
$json_var = $this->proxy->site('http://ip2country.sourceforge.net/ip2c.php?ip='.$ip.'&format=JSON');
...

// To call your controller (maybe in some situation, you need it instead use redirect() function)
...
$this->load->library('proxy');
$this->proxy->controller('your_target_controller_name/target_function/some_id',TRUE);
...

Grab it! And i'll be here, if you had some feedback, bug report or some trouble with it.

Cheers!
Toopay

[eluser]toopay[/eluser]
Well, because this library's main purpose is just to fetch/send simple http get request, one feature from this library, which distinguish this library with (*sigh) other CI curl library is : It can maintain image url from targeted site (see the different, in attachment image below. For comparing, i use phil's curl library to generate failed rendering page). It maintain anchor tag and form url too.

[eluser]toopay[/eluser]
Uh-Oh! Updated to V 1.0.1

MORE FEATURES(All old feature still there):

1. Get Full HTTP Header.
2. Set Proxy Call.
3. Set Delay HTTP Call.
4. Set user agent.
5. Internal cache (using gzip).
6. Persistent call (processing redirect, either from header or meta)
7. NO CURL OR OTHER FANCY STUFF DEPEDENCIES! PURE PHP.
8. Cookie support.
9. Log and error flag.

Code:

$this->load->library('proxy');
//Set proxy call
$this->proxy->set_proxy('proxy_host',80,'username','password');
$this->proxy->site('twitter.com',TRUE);
//Get rendered HTTP header, for use it as variable,
// put FALSE on second passed param or simple let it blank
$this->proxy->head('codeigniter.com',TRUE);
//Set user agent and delay
$this->proxy->set_delay(5);
$this->proxy->set_useragent($this->session->userdata('user_agent'));
$this->proxy->site('twitter.com',TRUE);

Test or play with it, download Proxy Library on wiki

[eluser]ClaudioX[/eluser]
Very, very nice!
When you call another controller (in the same site/server), the delay is big?

[eluser]toopay[/eluser]
No! It fast(test it).

The delay is optional. You can use set_delay() function, if you made more than one site(chaining call).

[eluser]toopay[/eluser]
V 1.0.2
cleaning (some debug's crap) and fixing structure also adding ReadMe.txt

[eluser]toopay[/eluser]
[UPDATED]V 1.0.3

MORE FEATURES(All old feature still there):

1. Get Crawled Web Content.
2. Optimize maintained rendered html : css and js path.

Code:

// Wanna see how CodeIgniter Forums looks like in searching engine's spider eyes ?
$this->load->library('proxy');
$this->proxy->crawl('ellislab.com/forums',TRUE);

Download Proxy Library on wiki

[eluser]toopay[/eluser]
No one intend to play with this one yet?

[eluser]tieungao[/eluser]
Hi i have some question about this lib :

1. After i crawl and get webpage return with this, how i can get special information from this page?

Before i have used HTML Dom Parse to get info from site.

2. Anything different from browser by this lib and real user browser

[eluser]toopay[/eluser]
[quote author="tieungao" date="1303459111"]
1. After i crawl and get webpage return with this, how i can get special information from this page?

Before i have used HTML Dom Parse to get info from site.
[/quote]
crawl() function will generates an informative array coressponding with url. Including meta tags and anchor(link). If you want to get full response, use site() or head(). Its have (optional) render functionality, which maintain image, anchor/link, js and css.

[quote author="tieungao" date="1303459111"]
2. Anything different from browser by this lib and real user browser[/quote]
Basicly, It simulates a real web browser, only that you use it with lines of code rather than mouse and keyboard.