[Grub-dev] C# Grubng 0.1 and few questionsHi everyone I create today first alpha version of graphic Grub client. It have all options of first release - minimal GUI, unstable program and ugly source code ; ) But i think it can be useful for test protocol. He works in two modes: - Semi-automatic:

[jer]

Very cool Bartek, really moving fast :)

A few answers on the questions below:

> 	I try to find this information on wiki or in archive
>  but maybe I'm blind ;) I'm interested how many different answers  
> can go from
>  server ("Content accepted", any other?), how he create links for  
> client -
>  sometimes he send invalid URLs (for example with HTML code)

The workunit generator had some bugs and was indeed making a few bad  
URLs, I've just fixed that and re-generated the workunits.

> and how long
>  client must wait for answer on server (sometimes server dont  
> answer even
>  if all workunit is send)?

Might be good to give it a minute or so?

> 	If this is (invalid URLs) are normal - how client
>  must react on it? Fix it or report as a invalid link? (for now it  
> use this
>  second option)

If the client detects anything bad about an individual request it  
should just insert a 500 error in the generated ARC file for that entry.

> And one thing - maybe funny, maybe weird - but wiki projects  
> (wikipedia,
> wikinews, wiktionary, wikibooks) have banned Grub client in robots.txt

The URLs in the workunits are obeying and have been validated against  
any robots.txt, but we're also using a different user agent string.   
I also fixed this in the new batch of workunits, it's now properly:
	User-Agent: GrubNG 20080128

Any sites that are blocking because of the old Grub, we'll just have  
to explain that this is an entirely new and improved and fully obeys  
robots.

Jer