Re: Bug found: tetrinet_sendfield seems to send a change when it should send a complete update

[James Antill <james-gnome-gtetrinet and org>]

Dani Carbonell <bocata panete net> writes:

> Hi Ian
> 
> We've added a short check function to compare the output of the old
> sendfield code with the new sendfield code. Eventually, the new code
> sends a change, when the old one sends a complete update of the field.
> Something like this:
> 
> ** (gtetrinet:3681): WARNING **: sendfield ... non compat. data
>         BEG: sendfield data
>          old=<4
> 00000000000000003000022000000000022000000002022000000000022100030000022005030044022055332044223055331044225055310144455052311144452022001332222122533332202132532120532451q442225501105014133250141455034333314503211334203121355325031553321545121310335124533101532244>
>          new=<4
> !75:7>87<5>6><?:@=A4B6B>B9C7D4E3F8G7H">77;6<8<7=>>8 4A>A3B7B9B5C4D:D;D6E8E5F;F3G7G<G6H8H#<4=4:67:;:4===3>:>;><>7 9@8A:A<B9D3E7E=E:F4G=G;H<H$74685:=:8>9>6?7?3@<@:B;B<C=C>C3D8D5E9E4F8F6G:G4H5H%99:9;<>@6A7A8B4C6C;C5D>D=F>G=H&493:=;<<5?;@3A<A5B=B7C8C6D:E;E>E6F7F<F;G3H9H.5A>
>         END: sendfield data
> 
> Strings have a leading '<' and a trailing '>'. The difference is quite
> obvious. I hope this helps you to fix the algorithm :)

 Well I've found one problem (the old version counted the '!'+i as
part of the update size, and the new one didn't), so that _might_ be
it ... however it seems weird that changing from a patch update to a
full update does anything but make it be a bit bigger over the
network.
 Anyway, everyone "cvs update" and try again (or just fix stuff if you
see it :).

-- 
# James Antill -- james and org
:0:
* ^From:  *james and\ org
/dev/null