[FW]: This code performs the rather esoteric function of converting an image file to a raw HTML page, using a tag with a cell for each pixel. The output is, of course, huge, and will crash some older browsers or systems. The only use I can think of this is if you had a hosting service which had tons of bandwidth but didn't support images, and ''really'' needed one. Yes, I realize this isn't quite standard HTML with doctypes and all that, but for something this useless I don't think it's worth being precise ;) proc htmlize {image_file save_to} { set page [open $save_to w] set source [image create photo -file $image_file] puts -nonewline $page {

} foreach row [$source data] { puts -nonewline $page {} foreach color $row { puts -nonewline $page "" } puts -nonewline $page {} } puts -nonewline $page {

} close $page } '''Feedback: meh''': Found it quite useful in [CGI] scripts for human verification. Stops them just matching the image filename to a set of corresponding characters - with some modification, random parts of the image can be modified to further deter non-human usage. Good work :) [FW]: Thanks. It happens that the more accepted way to achieve the first effect is by using a second script that sends through the appropriate image file (or generates one). To alter random parts or otherwise custom generate [CAPTCHA]s you can use the image tools directly. [meh]: Oh, it was more a temporary solution until I figured out how to use TclMagick :p [Sarnold]: You may gain bandwidth (I guess that would help) by generating all this HTML code with [Javascript]. It may be something like : function line(row) { var i; // increment by steps 6 because the color is stored in 6 characters for (i=0;i"); } } function nextline() {document.write("");} then, in your HTML body, just append:

---- [[ [Category Graphics] - [Category Image Processing] ]]