nsp audit

• npm install -g nsp
• nsp audit-package

The node security project maintains a database of security vulnerabilities in modules on npm. The nsp module will audit your package.json (or npm-shrinkwrap.json) file to check if any of your current dependencies have known vulnerabilities.

Vulnerabilities tend to get patched up pretty quick. Knowing about them gives you the choice to upgrade, or find an alternative dependency to keep your code secure.

npm-check

• npm install -g npm-check
• npm-check

This is the missing module for maintainers. npm-check uses semver to output how far out-of-date your dependencies are, and also reports on modules that seem to be unused.

If you've reviewed your dependencies and decided it's time to do some updates you can run npm-check -u to get an interactive upgrade experience! Mark the modules you want to upgrade with a space, then hit enter and npm-check will fix up your package.json, as well as run npm install AND finally, it reminds you to run your test suite at the end. Pure awesome.

Bonus tips...

Back in February, I watched Dan Hough's talk on npm module maintenance best practice at LNUG. The talk was also distilled into a handy blog post. There were two key take-aways that I wanted to reiterate:

Don't overlook CONTRIBUTING.md

GitHub's CONTRIBUTING.md file is a much under-used feature. It's the first thing I look for when I visit a new repo and usually I'm hunting for info on the project's preferred medium for questions, as every project is different. It's always worth taking two minutes to include one, even if it just says 'Please contribute'. It really helps to give new would-be-contributors like me confidence to get involved.

shrinkwrap for releases

The importance and usefulness of npm-shrinkwrap cannot be understated. I recently started shipping an npm-shrinkwrap.json file with the released versions of Ghost, which helps to ensure everyone gets the same experience. Most importantly, it eliminates tricky bugs caused by dependencies getting accidental breaking changes.

Admittedly, we still don't use it in the master branch because managing the checked-in file is a bit clunky, especially with a fast-moving codebase. However, we'll likely review this as the project matures.

Blogging elsewhere

Truth is I've been blogging on Ghost since day one. I've had 3 or 4 different Ghost blogs about different subjects hosted by various mechanisms, the main one being dev.ghost.org. However, my old blog with its legacy content about PHP stuff that I was doing 3/4 years ago and various pages with who-knows-what lurking on them was a minefield I didn't want to face. It desperately needed a refresh, I wasn't using it, and I have had 100 million more important things on my todo list than sorting out my old website!

At the very beginning of 2011 I moved from doing predominantly PHP & WordPress-based agency work, to working for MOO.COM where my daily tasks had me coding in PHP, Java, ActionScript, JavaScript and even Python. My years at MOO lead me to favour JavaScript, but the enormous commute to London left me little time for blogging. My old site became like a dusty old diary lost under the bed, a shadow of a former me - interesting, perhaps even useful to some, but not really relevant to the developer I am now.

A new lease of life

Today, I am bringing my old site to life again in a new home and under my real name Hannah Wolfe (hannah.wf). I've merged my old posts with posts from various Ghost blogs I've been keeping & set it all up in a new home on Ghost's Hosted Service.

It's cleaner, it's more grown up, and it's totally focused on blogging. I hope to continue to keep this blog in the same style as the old - a kind of 'notes to self' (and anyone listenting) diary of interesting things I discover. Expect the content to start leaning heavily towards JavaScript & Open source though - my PHP days are (mostly) over.

On 13th Feb I gave a talk about Ghost, being open source, and making node work for the masses at London JS Conf.

My slides for the talk can be found on speakerdeck.

If you want to make sense of the slides, Francis Rowland did an awesome set of sketch notes, and there is also a lovely write-up of the whole London JS Conf which has been published via a Ghost blog :)

Sign up for an account on Joyent, set up your key, and then hit create instance. It'll ask you to choose an image, I went for the first one:

And all you need is the smallest package:

Give your instance a name, hit 'Create Instance' and wait for it to start. Once it's up and running ssh into the box and run the following commands...

Install unzip, gcc47 and gmake packages:

pkgin install unzip gcc47 gmake

Download the latest Ghost package:

wget http://ghost.org/zip/ghost-latest.zip --no-check-certificate

Unzip Ghost into a directory called ghost:

unzip ghost-0.3.3.zip -d ghost

Change into your new directory

cd ghost

Install all of Ghost's dependencies:

npm install --production

Create a config file from the example:

cp config.example.js config.js

Open the config file for editing:

vim config.js

Change the IP address and port for development to 0.0.0.0 and 80 respectively. Save, and exit vim.

Finally start up ghost by typing

npm start

And then visit your public IP in a browser to see Ghost running.

Navigate to http://your-ip/ghost/ to create an account and start writing posts.

I'll likely be keeping this as the new home of my personal blog, more soon.

Tem um bom dia!

The build clones Ghost from GitHub and uses a grunt task that we already had setup thanks to Jacob Gable to pre-compile sass & handlebars, clean up some config files and get the raw codebase into a state ready for deploy. Grunt is turning into one of my favourite things... it's just so useful for getting tricky things done.

Ghost is already making pretty heavy use of Grunt: for compiling sass & handlebars, for linting, for running tests, for doing builds and soon for building documentation. I have a feeling Grunt & Ghost are going to be best of buddies for a long while.

What does this fix?

This is a really small update / bugfix to add support for https:// in the URL checks.

All credit goes to bteryk

It's been a while...

It's been far too long! This is one small step towards getting Migrate! back into active development. There are many more features & fixes to come, but first I had to get my head around github, pull requests etc again. All that is done now and I'm hoping to push several additional updates in the coming weeks.

Does this mean it still works?

I still use Migrate! on a regular basis and always on the most recent version of WordPress. So far it is standing the test of time. There has been an issue reported with Migrate! not migrating the URLs in the new menu feature, but this will only affect you if you hard code absolute URLs in the menu. You can keep up to date with any issues on GitHub

Warning, it is quite long! But it does cover installing the tools, setting up a LAMP stack, creating a virtual host and installing WordPress.

Step 1: Install the tools

Now for our virtual machine, if you want to run Ubuntu like me, then you can download an Ubuntu image from here: Ubuntu Server Images for VMWare. I used "32-bit: ubuntu-server-12.04-i386.zip" even though my computer is 64bit. I just didn't see any reason that my web server needed to be a 64 bit machine, but this should probably match whatever you use for hosting.

Next, unzip the Ubuntu image and put it somewhere safe & easily accessible on your hard drive. This will end up containing all your work, so really do make sure it's safe!

Now for the fun part, run VMWare player. Choose "Open a Virtual Machine" and navigate to where you saved your Ubuntu image & choose to open it.

You'll now have the options to "Play virtual machine" or "Edit virtual machine settings". You can give your new VM a better name by choosing edit, and then going to the options tab.

Hit "Play virtual machine". Important: you will be prompted to choose between "I moved it" and "I copied it", always choose "I moved it" which is NOT the default option. Failure to do this will cause you problems later. You will be prompted to log in, if you downloaded the suggested Ubuntu image then use the provided username and password from the thought police site, probably something like notroot & thoughtpolice.

Run the following commands to get any updates. It will take a few minutes, but then you will have Ubuntu installed, up-to-date and ready to start playing with.

sudo apt-get update
sudo apt-get upgrade

Note: throughout this tutorial, whenever you are prompted with a [Y/n] always type Y.
Note: if you've not used VMWare player before - to focus the VM you click on it as you would expect, but to get out of it and get your mouse cursor back you need to press CTRL + ALT.

Step 2: Install apache

Next, lets turn our Ubuntu install into a working web server, by installing Apache HTTP Server:

sudo apt-get install apache2

We need to edit the config:

sudo vim /etc/apache2/httpd.conf

Add the following line to the file, which will be blank. This defines our web server as being called localhost which is the norm for local development environments.

ServerName localhost

Note: If you've not used vim on the command line to edit a file like this before, you'll need to press i to enter insert mode and start adding new content to the file. When you're done editing press esc to leave insert mode, and then type : followed by w and then q. The : brings up a command line and the wq means "write and quit". Press enter and the file will be saved and closed.

Finally, we need to restart the server for our changes to take effect.

sudo /etc/init.d/apache2 restart

We now have a working local web server, therefore we have files on the Ubuntu VM which are addressable via IP addresses / URLs such that they can be opened and run in a browser. Apache has created these web accessable files a new folder that will be located at /var/www. Let's verify that this is true.

Type the following into your VM:

ifconfig

You should see something similar to the image below, you are looking for the inet address assigned to eth0, which will be the IP address you can use to access your webserver.

Open up your favourite browser, and enter http://#IP address here#, for the example above it would be http://192.168.178.129. You should see the "It works!" default file from apache much like the one shown below:

You can verify that this is indeed fetching files from your virtual machine by making a change. Navigate to your www folder:

cd /var/www/

Change the owner of the default index file and then open it for editing:

sudo chown www-data.www-data index.html
sudo vim index.html

Make any change you like for testing.

Don't forget: i for insert mode, type esc + :wq to save and quit

Once you've edited and saved the file, refresh the page in the browser to see your changes take effect.

Editing the Windows hosts file

In XP and before, this was merely a case of opening the file, but from Vista onward it's a little trickier as you have to be administrator.

• Browse to `C:/Windows/System32` or do a search for `notepad.exe`
• Right click on `notepad.exe` and choose `"Run as administrator"`
• Go to `File > Open`
• Navigate to `C:/Windows/System32/drivers/etc`
• Change file types `to All Files (*.*)`
• Open `hosts`

Localhost

It may be the case that you would prefer to refer to your web server by http://localhost, instead of having to remember the IP address. If this is the case then you need to edit your Windows hosts file to tell Windows about the name you want to use for your IP address. See the block to the right for details of how to open up your Windows hosts file.

Once you've got it open, you need to add a line to it to map the IP address to localhost or whatever name you might like to use:

#IP address here# localhost

So for my example I would put

192.168.178.129 localhost

Once you've edited the file you will need to save it. It may take a few seconds to update, but you will then be able to enter http://localhost into your browser and get to the default page. If it doesn't seem to be updating, you may need to flush the DNS cache. This is done in Windows by running ipconfig /flushdns either via the "Search programs & files" box in the start menu, or in a command prompt.

Step 3: Installing PHP & MySQL

We've got our LA part of LAMP, now it's time for the M & P! This bit is super easy.

To install PHP type the following in Ubuntu:

sudo apt-get install php5 libapache2-mod-php5
sudo /etc/init.d/apache2 restart

Let's check it works, by adding a PHP file:

cd /var/www/
sudo vim test.php

Add the following code to the file and save it:

<?php
echo "Hello World";

You can access this file to test it via either http://192.168.178.129/test.php or if you set it up: http://localhost/test.php.

To get MySQL installed:

sudo apt-get install mysql-server

You will be prompted to set and confirm the password for your MySQL admin user.

sudo apt-get install php5-mysql

You can test that MySQL is installed correctly by playing around with the MySQL admin (you'll be prompted for the password you just setup):

mysql -u root -p

Exit mysql by typing "quit".

Step 4: install and setup samba

Samba is what makes using a Linux VM as a development environment really simple. It's a little fiddly to setup, but once it is done you will be able to edit the files in your /var/www/ folder directly through Windows, using any IDE you choose. Let's get started...

sudo apt-get install samba samba-common

Add your user to samba (if you used the image suggested, this will be "notroot"):

sudo smbpasswd -a #your user#

Enter your password for your user, then we need to backup and edit our config for Samba so it's safe to edit:

cd /etc/samba
sudo cp smb.conf smb.conf.bak

Open the file for editing

sudo vim smb.conf

You need to make the following changes (use the arrow keys to scroll):

In the "Authentication section, uncomment (remove the hash / semi-colon) from the line

security = user

Then at the very bottom of the file, add the following and save the file:

[www]
comment = Web Root Dir
path = /var/www/
browsable = yes
read only = no
writable = yes
valid users = notroot, www-data
write list = notroot, www-data
read list = notroot, www-data
force user = www-data
force group = www-data
public = no
writable = yes
printable = no

Note that the user list should be your user (notroot if you are using the recommended Ubuntu image) and www-data.

Finally, restart samba:

sudo /etc/init.d/smbd restart

Now we should be able to connect via Windows. Open up a Windows Explorer Instance and navigate to Computer. At the top you should see a button which reads Map network drive. Select this and in the dialog that appears, choose a letter to map to, and then enter either localhost or the ip address of your web server as you found it earlier in the following format:

\\localhost\www\ or \\#IP address#\www\

Press "Finish", you'll be prompted for your username and password again, enter it and you should see the files index.html and test.php that are in your /var/www/ folder.

If you would like to be able to login via PuTTY for a slightly nicer terminal, run the following.

sudo apt-get install openssh-server

Now in PuTTY you should be able to connect to your web server's IP address and login with your username and password.

Step 5: Installing WordPress on a virtual host

Now that we have our LAMP stack all setup, and we have access to edit our files, we're going to use it to do something interesting.

First we need to set up a folder structure which the WordPress files will live in, which should look like /var/www/wordpress/public_html. You can either create these on the command line:

cd /var/www/
sudo mkdir wordpress
cd wordpress
sudo mkdir public_html
cd ../
sudo chown -R www-data.www-data ./

Or now that Samba is setup and working, just create the folders in Windows.

With the folders created, setup a virtual host by creating a new vhost file:

sudo vim /etc/apache2/sites-available/wordpress

Add the following to the file:

<virtualhost *:80>
  ServerName wordpress.dev
  DocumentRoot /var/www/wordpress/public_html
  DirectoryIndex index.php
</virtualhost>

Save the file. We need to enable our new vhost and then restart apache:

sudo a2ensite wordpress
sudo /etc/init.d/apache2 restart

So that we can access it in the browser, we need to add the new ServerName as a host to the Windows hosts file (see the localhost section above).

If you added localhost to it earlier, then change the line to look like:

#IP address# localhost wordpress.dev

Otherwise just add the line:

#IP address# wordpress.dev

Go to http://wordpress.dev in your favourite browser, and you should get an "It works!" page again.

Note: I like to use the fake TLD .dev to remind myself when I'm on the development version of a site, other people prefer to use a subdomain of the correct domain.

Now for WordPress. Download WordPress from wordpress.org and extract the files.
Open the folders until you get to the base of WordPress where you can see the wp-admin folder then copy all of the present files & folders into the /var/www/wordpress/public_html folder using windows explorer.

Our final piece of preparation is to create a database via the MysQL admin client. In VMWare type the following:

> mysql -u root -p
CREATE DATABASE wordpressdb;
GRANT ALL PRIVILEGES ON wordpressdb.* TO "wordpressdb"@"localhost" IDENTIFIED BY "wordpressdbuser";

Now we have everything we need, go back to http://wordpress.dev in your browser and you should see the install screen for WordPress.

Follow the on-screen install instructions providing the MySQL database details you setup above.

Finally, we're done! We've got a working web server which is running a virtual host with a WordPress install. Setting up a virtual host for any other purpose is much the same as for WordPress - you need the folders first, then a vhost file, you enable your vhost,restart apache and finally add the new name to your Windows hosts file.

Hopefully this has been useful, it only covers the basics of getting from A to B so there are many, many other configuration options to play with and packages to install depending on what you want to do. It's also not been secured, I do not recommend using these settings for configuring a production server.

Additional resources

• HTTPD - ubuntu help
• How to install LAMP and PHPMyAdmin on Ubuntu - Unixmen
• File sharing with samba - SitePoint
• How to set up apache virtual hosts on ubuntu 12 04 lts - digitalocean
• Installing WordPress using the MySQL Client - WordPress Codex

I'm just filing this away here for posterity :)

[caption id="attachment_2384" align="alignleft" width="528" caption="From philipchircop.com, click to see full post"][/caption]

I know, we're almost a week into 2012 already, it's a bit late for contemplating achievements and setting resolutions isn't it? Every year, the "January diet" gets further reinforced by our populist media - all the adverts are suddenly for exercise DVDs, clever cookbooks and healthy food - and nearly everyone I know is "on a diet". We're a nation gone mad, and yet we'll all be overweight again by Easter. We all strive to be better, and I'm tired of failing... so I've been trying to figure out how to do it right.

Out with the old and in with the new.

Why is New Year synonymous with thoughts of past & future personal achievement? As our calendars tick over to a brand new year, most folk feel drawn to it as a fresh start, a simple changing of the date makes us feel like we're starting over. Combined with a break from the weekly work-eat-sleep routine and copious amounts of time with friends & relatives, thoughts turn to work-life balance, evaluating what we've done with our year & reflecting on what was missing.

Then there's the tradition of setting a new years resolution. Usually something spur of the moment, arbitrary to measure and impossible to achieve: 'lose weight', 'quit smoking', 'do more exercise', 'drink less', 'spend more time with family', 'work less' & so on. These day most of us can count at least two of those on our todo lists, and most of us will try really hard for the whole of January only to inevitably forget, give up, or just downright fail.

Conceived to fail

Why is it so damned hard to keep up a resolution? Whilst the Christmas & New Year breaks give us all plenty of time to think & have our self-realisations, epiphanies & moments of sworn change, the truth is that unless you make drastic changes to your various environments, it doesn't take long before the old routines take over again, and you remember WHY you smoke/drink/never have time to cook a decent meal/have to stay late at work...again.

Cause and effect

I have come to believe, that if you really do want to change something about yourself, you have to look at the cause of your unwanted behaviour and importantly, WHY you want to change it. Habits are usually a coping mechanism of some sort, so either you need to remove the need to cope, or change the way you do it. But when it comes to the why, i'll bet the answer is something along the lines of social pressure from the very same society which gave life to your bad habits in the first place (cue vicious circle and nothing ever changing).

However, once you've figured out the causes and decided whether YOU really want to change (as opposed to feeling like you should do) you'll discover you can set yourself a more realistic goal. Hopefully something along the lines of learning a new skill, or challenging yourself to make a small change. That, or you'll realise you have a much more fundamental problem to solve.

This year, I'm not setting myself a new years resolution. Instead, I'm going to spend some time looking at my list of potentials & really understanding why those things are on that mental list of mine. Then perhaps I'll be able to action some real, long-lasting change.

Happy New Year Everyone. I hope you do something GREAT with it.

** Disclaimer *** I am well aware that Wikipedia is not the best source of information... but you know... it's just.. there!


A very dear friend of mine just posted a comment on Facebook that has prompted me to write this "Thinking out loud" post about my sleeping habits, which is largely an excuse for exploring wikipedia and looking at pictures of cute sleeping kitties. Anyway, here's what I wrote, and what my friend replied with:

Holiday-fatigue is something I regularly battle. I always have massive plans for what I can achieve with a long weekend or a week off, but by the time I've spent a day chilling out I'm in relaxed mode and never have the impetus to do any of it. People who do a full working week and still churn out side projects at a rate of knots in their evenings and weekends have both my envy and my awe.

However, my friend's comment made me stop and think because she believes I'm suffering sleep deprivation - this isn't something I recall ever having even hinted to her. Sure, getting up at 5:30 every morning to do a 2 hour commute means I live on the edge of getting enough sleep, but as far as I'm concerned this was something I adjusted to after about 2 months. The idea that it takes more than a few long sleeps to cancel out is also intriguing - so off I went to Wikipedia to read their article on Sleep Deprivation.

Apparently, your brain keeps a close account of how much sleep you owe it (that's kinda freaky), and if you continue to accrue sleep debt (apparently that's a thing) you eventually have to pay it back - sometimes your body will steal it back through unavoidable microsleeps or tiny naps (who hasn't had one of those at their desk after a big lunch?!). By this point, I've fallen into a Wikipedia Trap so also had to read the article on the Hypnic Jerk (tee hee).

All this information on sleep is fascinating: apparently the falling dream that often accompanies a Hypnic Jerk is actually a hallucination (weird!) and also, apparently, you can determine just how sleep deprived you are by timing how long it takes you to fall asleep (or your "Sleep Latency"). Shockingly (to me at least) according to the sleep latency test, if you fall asleep in less than 15 minutes, you are suffering a level of sleep deprivation?!? It usually takes me 10-15 minutes if I'm comfy in my bed in a quiet, dark room, regardless of how tired I feel or how much sleep I have. Husband on the other hand struggles to sleep without a TV on - so am I not just a good go-to-sleep-er?

The average adult requires 8 hours sleep (watch it, that link is NOT from Wikipedia haha) so if I'm "the" average adult, and I get roughly 7.5 hours sleep a night, then each week I accrue a sleep debt of 2.5 hours. I usually also get at least one train-nap or lie in a week, gaining me back another 30 minutes and I always have a lie in (at least 10 hours) at the weekend. Therefore, if I'm "the" average adult - I sleep enough (although husband will tell you that I always used to be a 9 hours kinda gal).

I remember a conversation I had before I started my new job where I was saying I couldn't remember the last time I didn't feel like I could fall asleep if you just gave me a pillow and some peace and quiet (this is still true). Back then I put it down to the stresses of my old job because I was commuting 10 minutes and easily getting 9 hours sleep a night. Nowadays I've a spring in my step and a smile on my face - I might not sleep but I'm much less stressed!

All this is purely academic anyway. We're all entirely different and have different needs. I'm certian that energy levels have as much to do with diet and levels of physical & mental exertion as they do your sleep debt.  Who knows what we really know or understand about sleep? We know naff-all about dreams. Most adults I know complain of little other than the weather and feeling tired.

So there we go, my random Wikipedia & thought trail is over. I have a cup of coffee in hand and feel more awake than I did before I wrote this... so what about you? Got any weird sleep habits?!