Subject: Re: CVS commit: basesrc/bin/ksh
To: NetBSD Userlevel Technical Discussion List <>
From: Greg A. Woods <>
List: tech-userlevel
Date: 09/29/2002 15:24:14
[ On Sunday, September 29, 2002 at 11:54:31 (+0200), der Mouse wrote: ]
> Subject: Re: CVS commit: basesrc/bin/ksh
> Or perhaps there's some other standard that specifies sequences
> generated by keystrokes.  (Or perhaps I'm wrong about X3.64, even.)

Hmmm.... I can only find one other reference online suggesting that ANSI
X3.64 does not imply that cursor keys should generate the same sequences
that would be interpreted as cursor movement commands on input.

Unfortunately Global Engineering Documents ( say that
ANSI X3.64 has been cancelled with the explanation "W/D NO S/S",
whatever that means.  They also offter to sell a hard-copy and an
electronic copy for the same price of $36[usd], which is almost triple
what it was as recently as 15 years ago!  :-)

I suppose X3.64 has in effect been superseded by "ISO 6429: ISO 7-bit
and 8-bit coded character sets -- additional control functions for
character-imaging devices".  Unfortunately that's apparently not one of
the ISO standards available through the CSA.

However it seems ECMA-48 is essentially the same as ISO-6429, and it is
freely available for download from  It does say:

	This Standard does not deal with the input component.

However it also talks about "Originating devices" transmitting control
functions conforming to the standard, and strictly I would interpret the
above sentence as meaning that the standard does not describe any
features or functions of the physical input components, and that it
still expects the manipulation of those components to ultimately
cause the device to act as an "originating device".

Indeed it is not unreasonable to consider that two pairs of conforming
terminals should be able to facilitate communication between their users
with only an interconnecting communications link between them, and as
such when one operates as an originating device then it is logical to
assume that there must be some correspondence between the control
sequences sent by the various cursor control keys and the action they
have on the display (or "presentation component") of the receiving
device.  So I'm going to assume, just as almost every terminal
manufacturer and terminal emulation programmer has that cursor keys on a
standards conforming terminal will generate the squences which
correspond to the related cursor controls.  :-)

								Greg A. Woods

+1 416 218-0098;            <>;           <>
Planix, Inc. <>; VE3TCP; Secrets of the Weird <>