Thanks for your answer.

> Too bad, that's HOW IT WORKS.

Yeah, I understand you maybe need to have temporary files to store data that
you need to  parse, but for each file/URL visited, it's like to dump each file
on the hard disk. (Too bad for just a spider option).

> Your WANTs are irrelevant. It does what it does.
Yeah, it does what it does. That just more easy to have simple readable file
than a kind of log with new.txt/new.log. Maybe my request is too selfish, but
why not dump the requests into a csv format for example?

Well that's just my point of view.
Httrack is great tool with an impressive work of Xavier (really impressive),
but with I find... some missing features that could be useful. 

I'm not the only one to just want a list of URLs for example ;)

But sorry to bother you, take it easy :)