[Moose-dev] Re: GLMTextualPresentation background color

Hi Andre, Looks cool. I would love to get these capabilities. To get something like this working, first you have to describe it at a higher level of abstraction: just like Mondrian is not a drawing engine, Glamour is not a UI engine. Glamour is a browser engine, and this means that the pieces should be at the browser level of abstraction. Ideally, this means that you do not mess around with variables like color, but rather with concepts that map on color. In your case, ideally we would have a model of text that can support highlighting of concerns and contextual actions. However, right now we do not have that, and even code highlighting is done with overusing display to pass in a formatted Text object (take a look at the senders of formattedSourceText). So, I have two suggestions: 1. There was a project led by Lukas called Safara that offered an object-oriented model of a text editor. I looked at it a while ago and it does not look too bad, but it's not quite finished. Maybe you can play with it and let us know: http://www.squeaksource.com/Safara.html 2. Perhaps you can take a shorter route and build a model of a simply text with zones that have styles and actions. Then we integrate this in the rendering. What do you think? Cheers, Doru On 1 Jul 2011, at 21:03, Andre Hora wrote: -- www.tudorgirba.com "When people care, great things can happen."