Google has just added a new feature to their biggest product: Google Search. "Google plus your World" shows you search results not only backed by Google Search but also by things your friends seem to like and have shared on Google+.
Personally I don't like what has happened here. While I agree on most topics with my friends, I don't want them to mess up my search results on a daily basis.
Google lets you turn off the "Google plus your world" feature with a simple click on the right side:
However, when coming back again to Google and doing a search the feature gets turned on again. On the right top corner you can click on the gear symbol and select "Search Settings". You can then select to disable personal results in your search results.
Don't forget to save your settings.
Happy browsing without messed up search results!
When looking at the various methods by which you can write server-side scripts, there are quite a few options, though it's easy to say that PHP is definitely one of the best. Like any langauge, it does have its limitations (unless you code past them) and its flaws, but there are a lot of advantages to choosing it as your server-side language of choice.
Having slowly become one of the most popular coding languages since its inception in the 1990s, PHP is open source and extremely adaptable to whatever purpose you're thinking of using it for. It's straightforward enough to learn if you're already familiar with code in general, though mastering it takes a great degree of focus and intense reading and writing, but if it means you end up coding the O2 servers, then it's worth it!
The best thing about PHP is just how easy it is to use, and if youre a coder it takes an afternoon to get started in earnest. Once you're up and running, you'll notice a few things about the language that'll make it one of your favourites. It's great for security measures, given its many customisable levels of security. It's also never a resource hog, and is easily connected to anything from image databases to encrpytion.
You've also got the major advantage of the colossal community surrounding PHP, as it's open source and this tends to foster a strong bond between people who want to work hard to further the code and help it improve and evolve over time. This means you have access to a lot of tech support and a fun, friendly bunch of people who will always be around to help you with the odd bit of code or any particular query you may have. PHP could just be the best language there is for server-side processes, so check it out and if you love it, get stuck in.
Some days ago a friend asked me how to extract a single database from a complete database dump. He created the dump using mysqldump and only wanted to restore a single database, without having go through >80k lines.
After a little thinking I came up with a little script that does the work:
I used grep to find the line of where the database is created and cut to extract the line number. Sed can be used to extract lines or a range of lines from a text file.
sed -n 1,10p file
This extracts lines 1 to 10 from a file. It's pretty amazing what sed can be used for..
As the number of virtual hosts on my vserver increased massively I searched for an alternative to cronolog. I used cronolog, a program for log rotation, in combination with lighttpd for every vhost. The result was that the more vhosts I added, the more cronolog processes had to be started. This consumed a lot of memory and cpu time.
The solution to this problem is simple: I need a program that automatically creates a log directory for every vhost and does rotation.
A program that serves this purpose is vlogger, which is written in Perl and allows tracking traffic per vhost to a database based on the weblogs. While this is a nice feature, vlogger did not completely fit my needs as it requires a dedicated logging directory and therefore can't be integrated into my existing directory structure. I forked the project and implemented logging into subdirectories and logging for "known-hosts" only (which basically means that if the logging directory does not exists for a vhost its logs get written to "default").
Shortly after commiting my version of vlogger to github I found rvlogger, an incomplete port of the original vlogger to ruby. As I'm a little addicted to Ruby I took rvlogger, and started implementing the missing features. My version of rvlogger is not compatible to vlogger because of a different database structure and file-based configuration, but it currently works very smoothly on my production server.
I'm using MySQL as database backend for rvlogger and lighttpd 1.4.26 to serve my vhosts. One instance of rvlogger replaced more than 80 cronolog processes for me.
Links:
« | 1 | 2 | 3 | 4 | 5 | 6 | »
