KenFallon.com

It was an ordinary day.
A day like any other when suddenly…

As I was transferring my ogg casts to my ogg player I realised that there was a new JakAttack episode. The show is by Jon Watson and his wife Kelly Penguin Girl. It resulted from an earlier show that Jon had that was restricted to technology topics. Their network wouldn’t let them change to a more relaxed “tech ‘n talk” format so they left and started JakAttack on their own.

They were going great guns for a while but when they moved back to Nova Scotia and bought a old house to renovate as a B&B, their podcasting time was limited. While others had given up all hope of ever seeing a new episode (even Dan put them in the graveyard section of http://www.thelinuxlink.net/) I knew they wouldn’t let us down.

So here we are – over a year has passed since the last episode and out of the blue Jon and Kelly were back. Back and at top form. Find the next cast (whenever that will be) at:
http://thejakattack.libsyn.com/

I listen to so many tech podcasts that I forgot that my ogg player can actually play music. Then while listening to the Linux Outlaws I heard that Dan Lynch had started a music podcast called Rat Hole Radio. So I set out to modify bashpodder not to speed up all my podcasts. In doing so I went down a rathole of my own where a simple change turned into a major project. Those that have seen Hal’s attempt to fix a light bulb in the opening sequence of “Malcolm in the Middle” Season 3 Episode 6 “Health Scare” will know what I mean.

Once I was back up and running I went looking for some Irish podcasts and came across Spud Show which features mostly unsigned female artists from Ireland and around the world. The show has been going for four years and although the presenter Brendan Bolger started out speaking English, he switched to Irish for a while. That turned out to be a long while as the show is still in Irish but don’t let that put you off. The show comes out a few times a week and it’s always something different, fresh and new. He does a great job of linking to the musicians on the website and always makes a point of spelling out the artist’s URL’s (in the English alphabet).

I had a happy few months going through the back catalogue and I can recommend the following shows (of course these are songs that I liked and more importantly were able to get the repeating sound of bob the builder out of my head.) :

• Catherine Hol (ep 9)
• Sheri Miller (ep 229)
• Kyoshi (ep 254)
• Dala (ep 356)

In short, give spudshow a listen. Now what was I doing again – ho yea – adding Rat Hole Radio to my feed . . .

Every so often I come across a PDF document that has been protected to prevent copying. You probably won’t notice this for a while until your copy and paste fails a few times. Finally you will see that the Edit -> Copy menu in Adobe Reader has been disabled/grayed out. Other readers like Okular show a more meaningful “Copying forbidden by DRM” message to alert you to what is going on.

The logic behind this feature is to prevent copying but is completely pointless because of the “if you can see it you can copy it” principle more commonly referred to as the analogue hole. A determined thief can take photographs of all the pages effectively bypassing all the content protection completely. Sure it’s not a digital copy of the text but they still have a digital copy of the image of the text and more importantly they have the confidential information you were trying to protect.

So all this protection achieves is forcing legitimate users to type text extracts instead of copying them. This will of course lead to errors in transcription and which will cause confusion down the line when your customers tries to order part number XI20458-AA7 when it is listed as X12O4513B-AA7 in the document. Can you spot the three differences ?

For the moment the solution is to convert  the PDF to text using the pdftotext tool from the poppler-utils package on kubuntu.

$ aptitude show poppler-utils
Package: poppler-utils
[snip]
Description: PDF utilitites (based on libpoppler)
 This package contains pdftops (PDF to PostScript converter), pdfinfo (PDF
document information extractor), pdfimages (PDF image extractor), pdftohtml
(PDF to HTML converter), pdftotext (PDF to text
 converter), and pdffonts (PDF font analyzer).

To install on GNU/Debian based systems:

$ sudo apt-get install poppler-utils

Once this is installed you can simply run the command

$ pdftotext bla.pdf

This will output a file called bla.txt with all the text from the document. Oddly enough running

pdftohtml bla.pdf

results in the error “Error: Copying of text from this document is not allowed.”, so it remains to be seen how long it will be before pdftotext will be patched to remove this feature.

I needed to convert some existing word documents to the DokuWiki format. After a quick search I realized that the best option was to:

1. open the word document in OpenOffice.Org Writer and select File -> save as -> HTML
2. use the program Html2DokuWiki.exe to convert the HTML to DokuWiki format.

This gave me a text file and a lot of graphics files of the embedded images.

In my case I wanted to be able to split this large file up into separate DokuWiki pages so that the document would be more manageable. This would allow several people to update different sections at the same time without running into locking issues.

After looking through the document I noticed that I wanted to split the document every time I came across the string “Section Title:”. I first thought of doing this in perl but then I found a bash one liner here that uses awk to split the document into smaller files.

$ awk '/Section Title:/ \r
{if (n) close(output); \r
output= f n++} n {print >> output }' \r
f=output.txt input.txt

This produces a series of files that are titled output.txt0, output.txt1, …, output.txtn. The first line of each is now something like this:

\ \ | **Section Title:** | Example_1: Example section| \ ||

After that I was able to rename the files to example_1.txt using the following script:

for i in output.txt*;do \r
 mv ${i} \r
`head -1 ${i} | \r
awk -F '|' '{print $3}' | \r
awk -F ':' '{print $1}' | \r
sed 's[^ [[g' | \r
tr "[:upper:]" "[:lower:]"`.txt;\r
done

July 31, 2009 is the 10th Annual System Administrator Appreciation Day. A day set aside for you to thank the people that keep the data flowing through the veins of your building. Having been a system administrator for a long time I appreciate the work that goes into keeping your computer, phone, pda, printer, tablet, or whatever connected and working. It’s a job where success is often gauged by how many people don’t notice changes.

On System Administrator Appreciation Day you get a chance to say thank you to the IT Crowd. It might be as simple as sending the help desk a email to comment on how well the systems are working. I know some people give gifts and others just grab them some coffee.

Personally, I like to give cake.

Happy 10th Annual System Administrator Appreciation Day !