Bug in Session_Component for v0.0.4

Around line 41 in packages/session/component/session.php the following parameter passed to session_set_cookie_params is incorrect: time() + Registry::$config['session']['lifetime'].

It should instead be set to: Registry::$config['session']['lifetime'] This will be patched in version 0.0.5



Replies

Patching this issue...deploying soon

Fixed bug in Session_Component, also fixed a few typos in parameter delcarations within code. Pushing in a few moments.


Similar Streams

Bug in Session_Component for v0.0.4

Around line 41 in packages/session/component/session.php the following parameter passed to session_set_cookie_params is incorrect: time() + Registry::$config['session']['lifetime'].

It should instead be set to: Registry::$config['session']['lifetime']

This will be patched in version 0.0.5

v0.0.5-codename-Vanilla available for immediate download

http://negative11.com/downloads

This patches the bug discussed here: http://strem.in/stream/show/150
It is recommended that you update to this version if you have not already patched the bug, as session handling will not behave as expected otherwise.

It also updates a few templates, fixes/adds links to unit tests and increments the version number in Core.

Setting Subversion svn:externals from contents of file

Somebody deleted the extremely helpful file that remind me how to do this from one of the dev servers. I need somewhere new for it to live.

This is how you can set the svn:externals property from the contents of a file named svn.externals. This allows you to keep your svn externals list in an easy to read/modify file. Then you only have to remember this complex command instead of the other complex command lol:

On the CLI enter:

svn propset svn:externals -F svn.externals .

Carefully note the "." is needed to set to cwd.

negative11-chocolate pushed!

Just committed the #negative11 version 2 (Choclate) files to git.

https://github.com/negative11/negative11-chocolate

Updated the README with basic installation instructions.

More updates to follow.
More wiki docs to follow.

This version is extremely lightweight.
Key improvements: slimmed down packages. Provides only the barebone MVC.
Namespace autoloading.
Improved core, patched bugs from previous version.
Moved to github!

Patched routing issue

A constant has been added to system/core/kickstart.php that allows override of default $_SERVER variable. In shared hosting environments, it may be necessary to set this to a different value than the default of 'PHP_SELF'. Alternative options are 'PATH_INFO' and 'REQUEST_URI'. This may allow users in shared hosting environments to use the framework when PHP_SELF is not set to the appropriate value. This path is not the most optimal solution, and the issue may be revisited in the future.

Inbox Overload Begone: Taskforce Exits Beta, Goes Pro With Paid Version

Email is an essential part of our daily communication, but it can also be a real pain in the ass. Or, going one step further, as my colleague MG Siegler recently put it, "email is the absolute devil". This fact even prompted him to channel Peter Gibbons and quit email altogether. I, personally, applaud this bold move but, instead of taking a vow of abstinence, am turning to other tools to help find a way to funnel the fire hose. Of course, many s

http://jbx.in/22wq #science #technology

call_user_func_array issues when migrating to PHP 5.3

Before version 5.3, supplying a scalar as the second parameter to call_user_func_array() resulted in the parameter being passed as the first parameter to the called function. As of version 5.3, this generates a warning: "call_user_func_array() expects parameter 2 to be array".

When migrating, ensure that you use:
call_user_func_array(array('class', 'method'), array(scalar));

instead of:

call_user_func_array(array('class', 'method'), scalar);

otherwise, you should use call_user_func() instead

Working on improved search mechanism.

One strem.in item that suffered a bit with the migration to NoSQL was our search. As I navigate the site, I notice that our similar streams and search page just don't return solid results.

Our previous mechanism, MySQL offers decent full-text search capabilities (http://dev.mysql.com/doc/refman/5.5/en/fulltext-search.html) that I have not been able to attain using Mongo, even with its regex functionality. I initially considered keeping MySQL around for this task, but thought better of it, as I was really just using full-text indexes/queries as an out-of-the-box, it's already done feature of MySQL. When I think about it, the search functionality wasn't really that great.

Instead, I'm playing around with some map-reduce and pattern-matching logic to improve the searchability of our growing pool of streams. Later today, I will update the site to use some preliminary versions of these tools.