Category Archives: tech

12 Nov
2010

Curl and Cookies

I noticed the other day that the cheat sheet I have on this site for curl doesn’t show how to use cookies, so I thought I’d remedy that omission, and quickly! Being able to use the command line to authenticate and then go on and use part of a site behind a login box can be really handy, and it is also super-simple.

-c to Save a Cookie

Pass the -c switch followed by a filename and curl will write the cookies to a file. This is the “cookie jar” and you can dip into it whenever you want to send the cookies back with a future request. For example:
curl -c cookies.txt http://www.lornajane.net
This writes a file named cookies.txt to the local directory. When I look in it, it contains:

# Netscape HTTP Cookie File
# http://curl.haxx.se/rfc/cookie_spec.html
# This file was generated by libcurl! Edit at your own risk.

www.lornajane.net FALSE / FALSE 0 s9y_4e071c5ccc553288993faf0369cb076c 539e01676501366ea0f04e2646b1a31d

-b to Send Cookies

All I do when I want to use the cookie on future requests is pass exactly the same command but with a -b switch; this will read the named file and send the cookies along. You can edit the cookies as you wish, at your own risk of course, and this makes the use of cookies and curl an absolutely invaluable technique for testing! It’s also common to use it on sites where you want to download a file directly to the server but the site requires login first.

20 Oct
2010

Is Enum Evil?

When I work on database designs, either on my own projects or as advisor to others, I often find people very reluctant to use an enum type for any columns. Now, I’m not about to advocate the gratuitous use of an enum column, they definitely have some pitfalls, but I think it is important to understand these rather than just shouting “enum evil!” if anyone should mention them.

There are cases where an enum is the correct choice for a particular type of data, so let’s look at what an enum type actually is and does.

Continue reading →

12 Oct
2010

Return on Investment: Example

I gave a talk at the weekend which talked in outline terms about Return on Investment or ROI. It was a keynote so I skated over the details, but I wanted to include a specific example to illustrate what I meant.

Imagine the scenario where, given 3 days to work on it, a developer can get the deployment time for their code down from 3 hours to 20 minutes. This company does, on average, 42 deployments per year (you can guess these numbers are totally imaginary).

So 3 days at 7.5 hours per day means we are investing 22.5 hours on this.

The return is the difference between the deployments, multiplied by the number of deployments that are needed. So 3 hours is 180 minutes – so we save (180 – 20) = 160 minutes with each deploy. We do that 42 times in a year so we’ve saved 6720 minutes (per year) which is 112 hours or 14.9 days.

Project managers might not like to lose 3 days from their schedule but how do they feel about having a spare 3 weeks each year?

05 Oct
2010

My Three Favourite Opera Features

I’m an Opera user, and one of the reasons I am is that I can control my browser entirely from the keyboard. I think everyone uses browser keyboard shortcuts, and in a lot of ways Opera has similar functionality to other browsers, but it also has keyboard spatial navigation, meaning I can jump to the next element in whichever direction I choose. Opera also has per-site preferences, which actually make the web usable for me.

Continue reading →

27 Sep
2010

Missing pcre.h when installing pecl_oauth

I was playing with pecl_oauth last week (more about that in a later post) and when I tried to install from PECL, it grabbed the files, ran the configure step but stopped with an error status during make. This is bad news for those of us who are ubuntu users rather than compile-happy linux users! Closer inspection showed this line around the point things started to go wrong:

Error: /usr/include/php5/ext/pcre/php_pcre.h:29:18: error: pcre.h: No such file or directory

I didn’t have the header files for pcre installed – in ubuntu the headers are in the -dev packages so I just installed what I needed:

sudo aptitude install libpcre3-dev

Re-attempting the pecl install, everything worked as expected. This is on Ubuntu 10.04 Lucid Lynx, and from reading around you’d want to install the same package in response to this error message, regardless of what you were doing to cause it. Hope this helps someone.

14 Sep
2010

Serendipity Template Update

If you read this site purely on a feed reader, this is probably irrelevant to you, but I recently made some changes to this blog and thought I’d post about how I worked with Serendipity to achieve them, mostly involving some template hacking and using a plugin to make a particular area of the site editable through the admin interface. Continue reading →
13 Sep
2010

Book Review: The Art of Community

This review has been in my drafts folder for 9 months, because I didn’t feel I was doing the book justice. It seems like I never will, so here it is – as it was written then, but hopefully still useful and relevant to someone

I bought this book last year when I was still working at Ibuildings, and my role changed a lot to include events and community representation. Before that I was doing entirely PHP development and it was around this time that I noticed myself saying “has everyone forgotten I’m actually a developer?” a lot! So I quickly decided that I needed a copy of The Art of Community, a book by Jono Bacon published by O’Reilly. Actually, I should thank O’Reilly at this point for publishing the book and even more so for sending Josette and her book stand to conferences – I was able to buy the book and it came with a pep talk :)
Continue reading →

06 Sep
2010

Indexes on Tables

Increasingly I find a very binary split between the professionals I come into contact with. One group of people are very database-aware and take the design of their storage quite seriously – with good results. The other group are more concerned with the functionality of their application, and have little regard for how it is stored other than considering it a keeping-place and making useful table and column names.

Too often though, they don’t think about how that data will be retrieved or what the implications are when it gets beyond the thousand records that were used for testing. This is where having an idea of how the data will be retrieved can really help application performance. (note: this article is aimed at users of traditional relational databases, and ignores all other possibilities). This post takes a look at the various index types and when to use them. Continue reading →

30 Aug
2010

Printable PDF Handouts from OpenOffice Impress

Last week I was preparing a training course for a client, and I wanted to print the slides nicely for the attendees to refer to and make notes on etc. The slides were done, I’d talked to my friendly printers (Mailboxes etc in Leeds) and all I needed to do was generate the handouts. Which was fine until I googled for help with doing that from OpenOffice, only to find that although it has this awesome “Export to PDF” functionality for documents, slides, etc, it wasn’t going to do it for handouts.

I’m an ubuntu user, and it turns out that there’s a clever package called cups-pdf which installs a pretend printer, and anything you could print, you can turn into a PDF. Brilliant. I installed it with aptitude and instantly I had a printer named “PDF” which printed to a /home/lorna/PDF directory.

Did I mention I love ubuntu?

I also wanted to add a cover page to my document, before I sent the whole thing to the printers in a PDF file for them to print and bind. For this I simply created an OpenOffice document and used the usual export to PDF. By the magic of twitter, I got some great advice from EmmaJane and installed the package PDFShuffler which enabled me to combine the two documents and save the result as a PDF.

DSCF2526.JPG

By the magic of open source, I have beautiful handouts :) Printing in Linux really has come a long way, I can’t thank the developers and maintainers of all those libraries enough – all I did was install two packages!

19 Aug
2010

One-Step Symlink Switch

This is a trick I use when deploying websites so I thought I’d post it here for posterity. Actually, technically I stole it from someone else but for now let’s pretend it’s mine (thanks @__kb!)

When I deploy an application, which is almost invariably a PHP application, I like to put a whole new version of the code alongside the existing one that is in use, and when everything is in place, simply switch between the two. As an added bonus, if the sky falls in when the new version goes live, the previous version is uploaded and ready to be put back into service. In order to be able to do this, I have my document root pointing at a symlink, let’s say it is called “current”. (disclaimer: I have no knowledge of non-linux operating systems, this post is linux-specific)

When it is time to deploy, I place the new code onto the server, and create two new symlinks, one called “previous” which points to the same location as the “current” symlink does (bear with me) and one called “next” which points to the location of the new code. To deploy, all I need is this:

mv -fT next current

The f forces mv to overwrite the target if needs be, and the T directs mv to consider the second argument as a normal file, rather than as a directory to copy in to. The neat thing about doing it this way is that it happens in a single move, no weird results for people who manage to hit your site while you are typing the new symlink command or during the code updating. It is also just as simple to roll back from this, since you have a symlink pointing to the previously used code version.

I thought I’d share this snippet as it is a handy inclusion in deployment scripts/strategies. What are your tips for managing code deployment?