Hi.

The Mango project states that it can serve HTML pages without allocating memory on the heap all by using slices. Are these techniques documented anywhere? A good tutorial on how to parse text without memory allocation would be appreciated. A guide to the guts of the Mango server would be nice too.

This been the kind of techniques that separates the D language should be taught to newbies like me from day one.

Thanks.

== Quote from bleep (bleep1024@mailinator.com)'s article
> Hi.
> The Mango project states that it can serve HTML pages without allocating
> memory on the heap all by using slices. Are these techniques documented
> anywhere? A good tutorial on how to parse text without memory allocation would
> be appreciated. A guide to the guts of the Mango server would be nice too.
> This been the kind of techniques that separates the D language should be
> taught to newbies like me from day one.
> Thanks.

I think the Mango forum would be a good place to post this question: http://www.dsource.org/forums/viewforum.php?f=5

bleep wrote:
> Hi.
> 
> The Mango project states that it can serve HTML pages without allocating
> memory on the heap all by using slices. Are these techniques documented
> anywhere? A good tutorial on how to parse text without memory allocation would
> be appreciated. A guide to the guts of the Mango server would be nice too.
> 

Yeah, that's a good idea. Walter had suggested doing something like that before also (for an article) but life has been a bit hectic recently :)

Slices help tremendously, but there's other techniques involved also: per-thread data, preallocated buffers, loads of caching, and so on. The design works well for a certain type of workload, but is not ideal for all.

> This been the kind of techniques that separates the D language 

I agree. Much easier to do this kind of thing in D