HTML

Printomi Maps

Well since we have made the decision to discontinue Printomi I have been backing up the databases and downloading the 90GB+ of images that users have uploaded.

Well it wouldn’t be like me if I didnt start thinking about what cool things I do with all those pixels. I remember seeing those cool Mincraft maps that use the Google Maps API to explore the Minecraft servers and it got me thinking if it could be possible to do something like that but for the Printomi images.

Well it turns out yes, you can: http://www.mikecann.co.uk/projects/PrintomiMaps/

Its only 1,024 of the total 27,497 images that were uploaded but even this small ratio results in a total map size of 115,200 x 86,400 pixels. The Google Maps obviously cant handle an image that large and it would take forever to download so you must split it up into many tiles. To achieve the zooming and performance you must also provide the map at various sizes.

With a map of 32×32 image with each image at 3600×2700 pixels and each tile at 450×337 this results in 87,000 tiles at 8 zoom levels! Generating and uploading all that data takes several days. But the result is you can pan and zoom around what seems like one huge 115k x 84k image.

If you are interested I have uploaded the source code I wrote to generate the 87k tiles at the various zoom levels from the 1024 3600×2700 input images. You can find it over on Github: https://github.com/mikecann/PrintomiMaps

If I had enough disk space on my web server I would love to do the whole 37,000 images in this way. If I did 128×128 that would use 16,384 of the images. This would result in 10 zoom levels and a map 460,800 x 345,600 pixels large. I have no idea or how long it would take to generate and upload all those tiles :P

Chrome Crawler – A web-crawler written in Javascript

Depending on your level of geekness you may or may not enjoy this one.

I proudly present Chrome Crawler, my latest Google Chrome extension:

The idea is simple really. You just give it a URL, it then goes off and finds all the links on that page then follows them to more pages then gets all the links and follows them and so on and so on.

Along the way it checks each page to see if there are any ‘interesting’ files linked there, if it finds an interesting link it will flag it for you so you can check it out.

Theres an options page that lets you customise the way it all works:

If you are still confused check out the video below:

So why did I make this? Well to be frank, I made it mostly “just ’cause I can”!

Also having learned from my last Chrome Extension project PostToTumblr I realised the Chrome API allowed you to do some things that you wouldn’t normally be allowed to do on a website (nameley the Cross-Origin XHR) and I wanted to do something to take advantage of it.

It didnt take me long to knock out this project, one lazy Saturday for the majority of the code and today for a quick fix or two and to write this post and make the video. As such I expect there to be many bugs and problems so if you encounter one drop me an email (my address is found in the options page).

Oh finally, I wouldnt try using this on a google page as you will likely end up seeing this quite often:

Anyways you can grab it over on the Chrome extensions gallery here. If you enjoy it please leave me a review / comment, much love!

The Case of The Transitional Doctype

I took a little break from my ‘top secret’ project (more on this coming soon) this evening to do some much needed repair work on one of my flash games portals www.worldsbestflashgames.com.

Top of my list was the fact that the index page doesn’t render correctly in IE7. What makes it odd is the fact that the category pages which are essentially identical to the index page rendered fine.This is what it looked like:

ScreenHunter_01 Dec. 10 19.28

Normally when you hear that there is an issue with a site on IE and not on FF or other browsers you automatically assume its CSS, and so did I. After stripping the site down to its bear bones however, comparing the index page against the category page I will still getting this oddness.

To cut a long story short I worked my way up from the bottom of the page to the top until I reaced the very top line:

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">

This line was the only line that differed in the index to the category page, which read:

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN">

So I took out the “Transitional” and low and behold it worked!

ScreenHunter_02 Dec. 10 19.41

No idea what Transitional does, but I hope this helps someone else out in the future!

 Scroll to top