Re: Update shouldn't nuke XF86Config-4

2002-04-26 Thread Paul Sargent
On Wed, Apr 24, 2002 at 01:28:04PM -0400, Joey Hess wrote: > What, you mean config files? Debconf does not ever, ever, write to > config files. Ill-designed maintainer scripts might. I wish people could > get their terminolgy right; accusing debconf of writing to a config > file is akin to accusing

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Osamu Aoki
On Wed, Apr 24, 2002 at 01:28:04PM -0400, Joey Hess wrote: > Paul Sargent wrote: > > I'm just glad backups are always made before debconf writes to files. > > Yes, /var/cache/debconf/*-old are handy. > > What, you mean config files? Debconf does not ever, ever, write to > config files. Ill-desig

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Joey Hess
Paul Sargent wrote: > I'm just glad backups are always made before debconf writes to files. Yes, /var/cache/debconf/*-old are handy. What, you mean config files? Debconf does not ever, ever, write to config files. Ill-designed maintainer scripts might. I wish people could get their terminolgy ri

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Gary Hennigan
"Paul Sargent" <[EMAIL PROTECTED]> writes: > On Tue, Apr 23, 2002 at 05:45:08PM -0500, Shyamal Prasad wrote: > > "Michael" == Michael D Crawford <[EMAIL PROTECTED]> writes: > > > > Michael> That's why, when you update the xserver-xfree86, you > > Michael> shouldn't overwrite the settin

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Mark Janssen
On Wed, 2002-04-24 at 18:58, Paul Sargent wrote: > The XF86Config file says that "bits outside of ### DEBCONF START and END > will not be touched" (I'm paraphrasing), so I moved the DEBCONF_START and > DEBCONF_END so they were together at the end of the file, but really I want > to set debconf to n

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Sridhar M.A.
On Tue, Apr 23, 2002 at 05:45:08PM -0500, Shyamal Prasad wrote: > > If you set up xserver-xfree86 using debconf (the dialog thing you get > when you install it), then you should not edit it by hand. If you do > edit it by hand, follow the instructions in the file that warn about > d

Re: Update shouldn't nuke XF86Config-4

2002-04-24 Thread Paul Sargent
On Tue, Apr 23, 2002 at 05:45:08PM -0500, Shyamal Prasad wrote: > "Michael" == Michael D Crawford <[EMAIL PROTECTED]> writes: > > Michael> That's why, when you update the xserver-xfree86, you > Michael> shouldn't overwrite the settings the user has laboured to > Michael> create. T

Re: Update shouldn't nuke XF86Config-4

2002-04-23 Thread Shyamal Prasad
"Michael" == Michael D Crawford <[EMAIL PROTECTED]> writes: Michael> That's why, when you update the xserver-xfree86, you Michael> shouldn't overwrite the settings the user has laboured to Michael> create. They are precious. Hi Michael, If you set up xserver-xfree86 using debcon

Re: Update shouldn't nuke XF86Config-4

2002-04-23 Thread Michael D. Crawford
I'm not sure, but I think maybe the change to my XF86Config-4 file was done by the GPM update. I think one is supposed to be able to get X and GPM to run on the same machine, but I couldn't get it to work. What you are supposed to do is read /dev/gpmdata as the mouse device - but this didn't

Re: Update shouldn't nuke XF86Config-4

2002-04-23 Thread Ed Lawson
Michael D. Crawford wrote: I looked into my XF86Config-4 file, and the protocol had been changed to PS/2. But the update didn't take me through any manual config process, it just silently changed it! H. I believe it asks if you want to redo the config file and isn't the default to not

Update shouldn't nuke XF86Config-4

2002-04-23 Thread Michael D. Crawford
Friends, One of the most difficult things for someone to get right on a new Linux installation is the settings of the XF86Config-4 file. Either the video doesn't work, or the mouse doesn't work, or both. Often it is troublesome for users to get it configured right, particularly novice users