Re: [PATCH] itrafmon.c: fix window change functionality

[Augusto Mecking Caringi]

On Fri, Jul 11, 2014 at 5:52 AM, Vitezslav Samel <vitezslav(a)samel.cz&gt; wrote:

>    See below.
>
>  On Fri, Jul 11, 2014 at 05:16:53AM -0300, Augusto Mecking Caringi wrote:
>  > ---
>  >  src/itrafmon.c |    9 ++++-----
>  >  1 file changed, 4 insertions(+), 5 deletions(-)
>  >
>  > diff --git a/src/itrafmon.c b/src/itrafmon.c
>  > index f19eb61..75975f2 100644
>  > --- a/src/itrafmon.c
>  > +++ b/src/itrafmon.c
>  > @@ -535,11 +535,12 @@ static int checkrvnamed(void)
>  >       return 1;
>  >  }
>  >
>  > -static void ipmon_process_key(int ch, int curwin, struct tcptable
>  *table, struct othptable *othptbl)
>  > +static void ipmon_process_key(int ch, struct tcptable *table, struct
>  othptable *othptbl)
>  >  {
>  >       static int keymode = 0;
>  >       static WINDOW *sortwin;
>  >       static PANEL *sortpanel;
>  > +     static int curwin = 0;
>
>  This patch adds a bug when the ipmon() is run twice or more:
>
>    1) run ipmon()
>    2) change active window (this sets curwin to 1)
>    3) exit ipmon()
>    4) run ipmon() (in the same iptraf-ng instance)
>    5) bug: now the curwin is still 1, but markactive(0, ...) just before
>       the mainloop assumes the curwin is 0!
>
>    You need to reset the curwin to 0 at every ipmon() innvocation!
>
>    So: NAK to this patch!
>
>    Was this patch supposed to fix something (according to Subject)?
>  
Hi Vita,

    Now I see... Sorry for that.

    I was trying to solve a bug that I found...

    After the keypress handling was moved into a separate function, the
switch of active window became broken.

    Best regards!

-- 
Augusto Mecking Caringi