home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.hp
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!ames!saimiri.primate.wisc.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!hpscit.sc.hp.com!scd.hp.com!hpscdm!cupnews0.cup.hp.com!wk
- From: wk@cup.hp.com (Wayne Krone)
- Subject: Re: HPUX 9.0 SIGWINCH on 800 series? -- was Re: hpterm resizable
- Sender: news@cupnews0.cup.hp.com
- Message-ID: <BxvzwE.4D0@cup.hp.com>
- Date: Wed, 18 Nov 1992 00:55:25 GMT
- References: <FRANL.92Nov16211347@draco.centerline.com>
- Organization: HP-UX Kernel Lab, Cupertino
- X-Newsreader: TIN [version 1.1.4 PL6]
- Lines: 24
-
- Fran Litterio (franl@centerline.com) wrote:
- : Yes, but will applications (such as vi) at 9.0 defer to the window
- : size as known by the device (accessed via ioctl) over the size as
- : specified in LINES and COLUMNS when the two disagree? Try this on a
-
- No, the user specification takes precedence (at least for the two
- commands I checked: vi and ksh).
-
- : If the same data (window size) is accessible via two different
- : mechanisms (environment variables and device ioctl()s), then
- : applications should use the mechanism that is least under the
- : influence of the user (ioctl()s). This is just standard defensive
- : programming.
-
- That's one philosophy. Another is that if a user goes to the trouble of
- expressly stating a desired value for some parameter then the application
- should use it in preference to any default or system provided value.
- Defensive programming is good but perhaps it should be limited to cases
- of protecting application or system integrity.
-
- Note I didn't implement any part of this functionality and the above
- philosophy probably has nothing to do with why vi acts the way it does.
- I believe this issue was discussed here previously and the implementers
- did post their reasoning but I've forgotten what they said.
-