[Info-vax] DCL, was: Re: Microkernel

[Paul Sture]

On Sun, 12 Aug 2012 10:02:24 -0700, John Wallace wrote:

> On Aug 12, 5:40 pm, Simon Clubley <clubley at remove_me.eisner.decus.org-
> Earth.UFP> wrote:
>> On 2012-08-12, ChrisQ <m... at devnull.com> wrote:
>>
>>
>>
>> > Another way might be to save the (just reloaded) recall buffer as a
>> > separate file on entry to a session, then you just have to compare
>> > that at any time later with the current recall buffer to know what's
>> > been added or changed. Something like this must have been done years
>> > ago though. Perhaps something on old decus tapes ?.
>>
>> I've never seen anything like this available for DCL. The first time I
>> heard of permanent command history recall was on Unix based systems.
>>
>> I've just had a look online to see if I could find anything and I could
>> not, although I do see that Hoff has a comparison between DCL and bash
>> on his website, and I also see I've been asking for this capability for
>> a number of years. :-)
>>
>> > Usually have 2 or 3 terminal sessions running bash here, and they all
>> > have unique history files. Not sure how that works though...
>>
>> Interesting. For me, everything ends up in ~/.bash_history.
>>
>> (I've just double checked to be sure. :-) This is on a RHEL 5.x clone.)
>>
>> Simon.
>>
>> --
>> Simon Clubley, clubley at remove_me.eisner.decus.org-Earth.UFP Microsoft:
>> Bringing you 1980s technology to a 21st century world
> 
> Had this discussion at work a while back re one global bash history vs
> separate bash histories per shell instance. This was on a not quite up
> to date Fedora, I forget which. Anyway it's a config option somewhere,
> and it's documented (ie it's on the web somewhere). Can't remember where
> though.

Here's the Gnu bash documentation:

http://www.gnu.org/software/bash/manual/bashref.html#Using-History-
Interactively

> http://serverfault.com/questions/337123/strange-bash-history-behaviour-
when-running-multiple-sessions
> looks potentially relevant although I'm not sure it is either necessary
> or sufficient.

That might explain some of the things I have seen.

-- 
Paul Sture