On Tue, Jan 03, 2006 at 05:23:36PM -0600, Manoj Srivastava wrote: > On Tue, 3 Jan 2006 17:37:05 +0100, Marc Haber > <[EMAIL PROTECTED]> said: > > Reopening the bug to keep it from getting archived. postinst scripts > > generating output need to be supported. > > Umm, they shall not be supported, no.
Ah, it's even documented now. > Why do you think they > must be supported? I think they should be supported since they have been supported for a long time. >If you can figure out a way of supporting debconf, > and other hook scripts that use debconf, and also support scripts > that write to stdout, well, we'll talk. One possibility would for example be to buffer hook script's output fd's until all hook scripts have been executed and thus it is guaranteed that debconf interaction can be stopped. After issueing the perl equivalent to db_stop, it should be safe to dump out the buffer contents. You have, however, already indicated that you are reluctant to implement a buffer scheme like that. Please at least offer the possibility to pass parameters to hook scripts (so that update-grub could get a --silent option), or - even better - to use a shell for hook script invocation so that constructs like postinst_hook = /sbin/update-grub > /dev/null 2>&1 become possible by forcibly silencing the hook script until the hook script has been adapted. Thanks for your consideration. Greetings Marc -- ----------------------------------------------------------------------------- Marc Haber | "I don't trust Computers. They | Mailadresse im Header Mannheim, Germany | lose things." Winona Ryder | Fon: *49 621 72739834 Nordisch by Nature | How to make an American Quilt | Fax: *49 621 72739835 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
