I am trying to create a social network application using elgg.Since i am pretty new to elgg i like know whether i can define url routing like in cakePHP
I need the url like this
mydomain.com/username instead of mydomin.com/pg/profile/username
Is there any way that i can avoid /pg/ and /mod/ from the urls??
I am using elgg version 1.7.8.
I am not interested in url rewriting with .htaccess.
Thanks in advance
elgg itself is using .htaccess file to redirect /pg, /action etc. So, I think there is no other way than using .htaccess.
Elgg's page handling is pretty bad but you can register handlers as follows.
for /mypage:
elgg_register_page_handler('mypage', function($pages){
//content here
//the $pages parameter is an array. so /mypage/a/b with return
// array('a','b');
});
Matt Beckett has written a profile URL plugin (https://community.elgg.org/plugins/1091233) that, though for 1.8 and above, is only 35 lines of code and does exactly what you ask. You should be able to adapt this quite easily for 1.7.8.
He has also written a fuller page handler hijack plugin that may help if you have other use cases apart from user profiles - see https://community.elgg.org/plugins/854839
It is also only for 1.8 and above, however, and may take more effort to backport than the first.
Related
I have the same issue described here: Magento URL Rewrite Management preserve GET parameters
Which kind of changes should I do to Rewrite.php? Is there any other way to make it work in general without changing rewrite.php?
Build the urls like:
$this->getUrl("*/controller/action", array('param1'=>value, '_current'=>true));
The value '_current'=>true will keep your current params in the new URL
currently I am trying to register a Doctrine-Eventlistener for every request in my FLOW3-Package. Some research pointed me to the Package.php, but unfortunately the ObjectManager is not available when the boot()-Method is called.
I searched the whole FLOW3-Documentation http://flow3.typo3.org/documentation/guide/partiii/bootstrapping.html without luck
Any hints on which is the right place to do package-wide setup with access to the object manager?
thanks and best regards
Your are bit to early in the bootstrap to get every object, I have a problem, maybe related to yours, you can check my bug report on http://forge.typo3.org/issues/33838
Why do you need a Doctrine Eventlistener, maybe you can use AOP to have this kind of feature ?
I have a website set on a specific domain which is completely separated from my couchdb url through rewrites and virtual hosts, and I got to a point where I need to add some user authentication using _sessions API but I'm afraid I can't do it with rewrites:
{
"from": "auth",
"to": "../../../_session"
}
gives me:
{"error":"insecure_rewrite_rule","reason":"too many ../.. segments"}
which is acceptable, but now I'm wondering how would I get the session authentication to work from my domain without exposing couchdb url, and also, the session seems to be related to the domain so if I login through couchdb.example.com it won't work when using mywebsite.com as the public interface?
Thanks
PS. I've just found this post where there's an alternative by disabling secure_rewrites on the httpd config file, which seems to work, although, I was wondering that perhaps might be not a good approach and if is there something else which is ideal for this kind of problem.
I recommend to set secure_rewrites=false and don't worry about it.
We had a great discussion about CouchDB rewrites and security in the Iris Couch forum. Also see my post later about using Audit CouchDB. These are the highlights:
The secure_rewrites option is not the ultimate source of security for your data. At best, it is one layer in a multi-layer solution
The ultimate source of security is the _security object in the database. So that is where you should focus your attention
The Audit CouchDB tool scans every detail about your couch and it will tell you if any red-flags are present. It is implemented in Javascript so if you have NodeJS, you can run it; or simply reading the source code gives you an idea of what it is looking for.
If you are using vhost, than /_session handler is available at the vhost root without any rewrite rules (by default).
See the section [httpd] of default.ini:
vhost_global_handlers = _utils, _uuids, _session, _oauth, _users
could any one guide me on how can I update status on linkedin using codeigniter?
Any help appreciated thanks!
I implemented this library into CI to make sure a user could log in with his LinkedIn profile. Maybe you can also update the status of a user.
http://code.google.com/p/simple-linkedinphp/
Implementing it in CI is pretty easy if you know your way around in CI.
I really like it when people write libraries for others to use and the code is commented/well thought out/well structured and is easy to follow.
I tried the other examples, but settled on this one.
http://blog.justni.com/posting-to-linkedin-from-php-and-codeigniter/
Edit:
Having banged my head for a long time on the POS code of Sudhir and the code above. Yes I can change my opinion, because when I started to work on it, it just doesn't work!.
Although I am new to CI, I managed to do the same as P.T and implemented the lib into CI.
The simple-linkedinphp class works like a charm.
I can now authorise an app, authenticate, get profile details and even post to linkedin.
I even posted a comment on Sundir's blog that it doesn't work and he deleted the comment! Stay away, well away if you don't want any headaches!
I'm trying to load some external content using jQuery load function to div on my page. load method works ok, with local content, but if you want something out of your domain, it won't work.
$("#result").load("http://extrnal.com/page.htm #data);
(it actually works in IE with security warning, but refuses to work in Chrome at all). jQuery documentation says that it is right, because cross-domain content is restricted because of security reasons. Same warning I get if use .getJSON method.
OK, after a googling a bit I found very interesting approach of using YQL for loading content, I've tried some examples, like this:
var request = "http://query.yahooapis.com/v1/public/yql?q=select%20*%20from%20html%20where%20url%3D%22http%3A%2F%2Ffinance.yahoo.com%2Fq%3Fs%3Dyhoo%22&format=json&diagnostics=true&callback=?";
$.getJSON(request, function (json) {
alert(json);
}
);
And it really works!
What I dont understand now is that http://query.yahooapis.com is also cross-domain resouce but browser (both IE and Chrome) works OK with that?
Whats the difference? What am I missing?
Thank you
The results you are getting back from YQL are in JSON format which is permitted for cross site AJAX calls like this. Its the same mechanism that allows you to communicate with web services for external sites via JSON (Ie. the twitter API).
Details here - http://www.wait-till-i.com/2010/01/10/loading-external-content-with-ajax-using-jquery-and-yql/
you can make on external site JSON like this:
callback({key:value,etc:1})
and define
function callback(json) {
..here is processing..
}
Thanks for your answers, but unfortunately both of them do not answer my orginal question..
I've checked out related questions on stackoverflow (i know i need to do that first) and found the reason of such behavior.
First code snipset uses AJAX/JSON to retrive the data and it is permitted because of Same Origin Policy. But request to YQL uses JSONP instead, that is OK.
The JSONP was something that I don't know about, that's why I didn't undrestand the behaviour.
Introduction info on JSONP could be found here:
http://ajaxian.com/archives/jsonp-json-with-padding