[Info-vax] Last Call for (New) DEC VT Terminals

[Johnny Billquist]

On 2015-06-29 03:24, Arne Vajhøj wrote:
> On 6/28/2015 9:14 PM, Johnny Billquist wrote:
>> On 2015-06-29 02:15, Scott Dorsey wrote:
>>> Johnny Billquist  <bqt at softjar.se> wrote:
>>>>
>>>> For me, the most annoying PuTTY bug is how it behaves when writing and
>>>> backspacing at column 80. If you write at column 80 without automatic
>>>> wrapping, the character appears at column 80, and the cursor remains at
>>>> column 80. If you then backspace on a real VT100, your cursor will move
>>>> back to column 79. In PuTTY, the cursor remains at column 80, because
>>>> the PuTTY developers prefer this behavior, even though it is
>>>> incompatible. I talked with them many years ago about this, but they
>>>> refused to change their code. So I have a program that I've written,
>>>> which behaves correctly on a real terminal, but will exhibit a broken
>>>> behavior in PuTTY.
>>>
>>> PuTTY follows the ansi behaviour, the vt100 does not.
>>
>> So maybe then PuTTY should not claim to be VT100 compatible?
>>
>> But to be honest, I don't know if I've ever seen/read the ANSI standard
>> on this. Do you have an URL for that document, so I can read what ANSI
>> is saying?
>
> ECMA-48 / ANSI X3.64 / ISO 6429 is available here:
>
> http://www.ecma-international.org/publications/files/ECMA-ST/Ecma-048.pdf

Aha. Cool. Thanks!
And now I can say that PuTTY is no more conformant than a VT100, since 
the standard actually says that the behavior in that situation is 
undefined. (Section 6.1.6 page 16).
You could possibly argue that PuTTY is less compliant, as the issue 
actually is the backspace, which PuTTY gets a different behavior for 
which could be slightly questionable, as it carries over from the 
previous undefined behavior of the implicit movement from the previous 
character.

Oh well, I'll just continue to occasionally complain that PuTTY is 
broken in comparison to a VT100. :-)

	Johnny