Re: [patch] LyXSocket cleanup

2004-07-26 Thread Andre Poenitz
On Sun, Jul 18, 2004 at 04:18:13PM +0200, Lars Gullik Bjønnes wrote: > | You've tested this carefully to make *sure* it works? > > No :-) > > it is more "this should obviously work, else qt is braindead"... but > we already know the answer to that I guess The problem is that Qt _is_ braindead.

Re: [patch] LyXSocket cleanup

2004-07-20 Thread Lars Gullik Bjønnes
Joao Luis Meloni Assirati <[EMAIL PROTECTED]> writes: | Lars wrote: > >> | Index: lyxsocket.C >> | - fd_(lyx::support::socktools::listen(addr, MAX_CLIENTS)), >> | + fd_(lyx::support::socktools::listen(addr, 3)), > >> The use of MAX_CLIENTS here is not doing what the original auth

Re: [patch] LyXSocket cleanup

2004-07-19 Thread Joao Luis Meloni Assirati
Lars wrote: > | Index: lyxsocket.C > | - fd_(lyx::support::socktools::listen(addr, MAX_CLIENTS)), > | + fd_(lyx::support::socktools::listen(addr, 3)), > The use of MAX_CLIENTS here is not doing what the original author > thinks it does. The backlog parameter is just how many _o

Re: [patch] LyXSocket cleanup

2004-07-18 Thread John Levon
On Sun, Jul 18, 2004 at 04:18:13PM +0200, Lars Gullik Bj?nnes wrote: > it is more "this should obviously work, else qt is braindead"... but > we already know the answer to that I guess Indeed. It's best to not touch Qt /at all/ unless you're fully prepared to test every changed line... > Is it

Re: [patch] LyXSocket cleanup

2004-07-18 Thread Lars Gullik Bjønnes
John Levon <[EMAIL PROTECTED]> writes: | On Sun, Jul 18, 2004 at 01:38:33PM +0200, Lars Gullik Bj?nnes wrote: > >> Here is an updated patch as well, gets rid fo the data_recieved >> function in qt socket_callback. > | You've tested this carefully to make *sure* it works? No :-) it is more "this

Re: [patch] LyXSocket cleanup

2004-07-18 Thread John Levon
On Sun, Jul 18, 2004 at 01:38:33PM +0200, Lars Gullik Bj?nnes wrote: > Here is an updated patch as well, gets rid fo the data_recieved > function in qt socket_callback. You've tested this carefully to make *sure* it works? connect() does NO compile time checking. In fact, I'd be very surprised

Re: [patch] LyXSocket cleanup

2004-07-18 Thread Lars Gullik Bjønnes
Angus Leeming <[EMAIL PROTECTED]> writes: | Lars Gullik Bjønnes wrote: >> So no substantial objections to this patch then? > | Q (Angus): Do things continue to work after your patch is applied? | A (Lars): Of course, or else I would not be suggesting applying this patch. So you are saying that I

Re: [patch] LyXSocket cleanup

2004-07-18 Thread Angus Leeming
Lars Gullik Bjønnes wrote: > So no substantial objections to this patch then? Q (Angus): Do things continue to work after your patch is applied? A (Lars): Of course, or else I would not be suggesting applying this patch. What objection could I make, therefore, given that a feature freeze is meani

Re: [patch] LyXSocket cleanup

2004-07-18 Thread Lars Gullik Bjønnes
Angus Leeming <[EMAIL PROTECTED]> writes: | Lars Gullik Bjønnes wrote: > >> At least I find it a bit nicer. Comments/Objections? > | You go on to use 'MAX_CLIENTS' below, so don't cull it here: > | Index: lyxsocket.C | - fd_(lyx::support::socktools::listen(addr, MAX_CLIENTS)), | +

Re: [patch] LyXSocket cleanup

2004-07-18 Thread Angus Leeming
Lars Gullik Bjønnes wrote: > At least I find it a bit nicer. Comments/Objections? You go on to use 'MAX_CLIENTS' below, so don't cull it here: Index: lyxsocket.C - fd_(lyx::support::socktools::listen(addr, MAX_CLIENTS)), + fd_(lyx::support::socktools::listen(addr, 3)), In Engl

Re: [patch] LyXSocket cleanup

2004-07-17 Thread Lars Gullik Bjønnes
[EMAIL PROTECTED] (Lars Gullik Bjønnes) writes: | At least I find it a bit nicer. Comments/Objections? > | I have not fixed up qt and gtk to use the same scheme yet. This patch supposedly fixes up qt and gtk as well. (I am not even compiled it...) Index: lyxsocket.h =

[patch] LyXSocket cleanup

2004-07-17 Thread Lars Gullik Bjønnes
At least I find it a bit nicer. Comments/Objections? I have not fixed up qt and gtk to use the same scheme yet. (And I you wonder what I am doing: I am combing my way through the code and looking close at stuff that went in when I was not paying close attention. And trying to understand all the