|
Home > Archive > Slony1 PostgreSQL Replication > July 2005 > PATCH: refactoring slon a tiny bit
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
PATCH: refactoring slon a tiny bit
|
|
| Magnus Hagander 2005-07-26, 1:24 pm |
| Hi!
Attached is a patch that refactors slon a tiny bit, by creating
functions SlonyMain() and SlonyWatchdog() for the two branches of the
fork() command. It makes the code a little bit more readable from an
#ifdef WIN32 perspective, but mainly it's a bit of groundwork for
upcoming service integration code.
It also contains a bundle of changes to correct a previous error in some
win32 code that was committed with invalid line endings (don't ask me
how that happened - Dave, perhaps you need to verify how your cvs client
is set up? I recall this happening once on the pginstaller project as
well, sometime in the beginning?). I'm not sure if that'll actually be
fixed through this patch (when I apply it to my linux box, patch happily
ignores the changes), but it needs to be fixed in the repo somehow. If
nothing else, the patch shows where the problem is :-)
(slon builds fine even with the bad line endings in there, but it screws
up line numbering in gcc - different line numbers in error messages than
in the actual file...)
//Magnus
| |
| Darcy Buskermolen 2005-07-26, 8:25 pm |
| On Tuesday 26 July 2005 11:20, Magnus Hagander wrote:
> Hi!
>
> Attached is a patch that refactors slon a tiny bit, by creating
> functions SlonyMain() and SlonyWatchdog() for the two branches of the
> fork() command. It makes the code a little bit more readable from an
> #ifdef WIN32 perspective, but mainly it's a bit of groundwork for
> upcoming service integration code.
I'll be adding *nix demonizing code shortly after you have the service
integration code complete though I haven't polled the list to see what the
default should be.
>
> It also contains a bundle of changes to correct a previous error in some
> win32 code that was committed with invalid line endings (don't ask me
> how that happened - Dave, perhaps you need to verify how your cvs client
> is set up? I recall this happening once on the pginstaller project as
> well, sometime in the beginning?). I'm not sure if that'll actually be
> fixed through this patch (when I apply it to my linux box, patch happily
> ignores the changes), but it needs to be fixed in the repo somehow. If
> nothing else, the patch shows where the problem is :-)
> (slon builds fine even with the bad line endings in there, but it screws
> up line numbering in gcc - different line numbers in error messages than
> in the actual file...)
>
>
> //Magnus
--
Darcy Buskermolen
Wavefire Technologies Corp.
http://www.wavefire.com
ph: 250.717.0200
fx: 250.763.1759
| |
| Dave Page 2005-07-27, 3:26 am |
| =20
> -----Original Message-----
> From: Magnus Hagander [mailto:mha- +PAb3xVejjn11pZQbBMD
iQ@public.gmane.org]=20
> Sent: 26 July 2005 19:21
> To: slony1-general- AuKwsB3Fm+ugFIWk8tvy
RWD2FQJk+8+b@public.gmane.org
> Cc: Dave Page
> Subject: PATCH: refactoring slon a tiny bit
>=20
> Hi!
>=20
> Attached is a patch that refactors slon a tiny bit, by creating
> functions SlonyMain() and SlonyWatchdog() for the two branches of the
> fork() command. It makes the code a little bit more readable from an
> #ifdef WIN32 perspective, but mainly it's a bit of groundwork for
> upcoming service integration code.
I'll apply this in a couple of days if there are no objections.
> It also contains a bundle of changes to correct a previous=20
> error in some
> win32 code that was committed with invalid line endings (don't ask me
> how that happened - Dave, perhaps you need to verify how your=20
> cvs client
> is set up? I recall this happening once on the pginstaller project as
> well, sometime in the beginning?). I'm not sure if that'll actually be
> fixed through this patch (when I apply it to my linux box,=20
> patch happily
> ignores the changes), but it needs to be fixed in the repo somehow. If
> nothing else, the patch shows where the problem is :-)
> (slon builds fine even with the bad line endings in there,=20
> but it screws
> up line numbering in gcc - different line numbers in error=20
> messages than
> in the actual file...)
Odd - last time it was because I had reinstalled my laptop, and WinCVS
had the wrong type of line endings selected. Nothing has changed in a
long time since (though I now use TortoiseCVS) so I don't know how that
happened.
I'll keep an eye on it...
/D
| |
|
|
|
|
|