On 2021-05-31, Hans Bachner <hans@bachner.priv.at> wrote:
> Chris Townley schrieb am 31.05.2021 um 16:51:
>> It was sometimes useful to add an incorrect character to avoid a symbol
>> definition - for example deletes
>
> It still is useful. If you work at various customers' systems from time
> to time, with setups varying from customer to customer, it is convenient
> to type "$ direx /qual1 /qual2" to get displayed exactly the file
> attributes you want for the files you want to see.
>
> I never found a customer system with "direx" defined as a symbol, and if
> it would be defined it wouldn't do much harm. On the other side, most
> customers have some dir* symbols defined, with and without abbreviaton.
>
> Similar for other commands like copy etc.
>

That's one hell of an ugly hack and there should be a far better way to
handle this other than resorting to that.

Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
allow the use of a leading underscore to skip any symbol translations ?

Something like:

$ _dir filename.dat

would get you the "real" dir command without having to worry about
symbols.

I wonder if we could have something like this for VMS DCL ?

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

In article <s938qc$7ed$1@dont-email.me>, Simon Clubley
<clubley@remove_me.eisner.decus.org-Earth.UFP> writes:

> I wonder if we could have something like this for VMS DCL ?

I've known about a trailing underscore for decades.

On 2021-05-31 19:56:32 +0000, Phillip Helbig (undress to reply said:

> In article <s938qc$7ed$1@dont-email.me>, Simon Clubley
> <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>
>> I wonder if we could have something like this for VMS DCL ?
>
> I've known about a trailing underscore for decades.

There's no trailing-underscore syntax in DCL—any more than there's
actually really any leading underscore syntax in logical name support
with V4.0 and later—there's just stuff appended to the command that
gets ignored by the parser but that also suppresses symbol matches.

--
Pure Personal Opinion | HoffmanLabs LLC

On 5/31/2021 2:10 PM, Simon Clubley wrote:
> On 2021-05-31, Hans Bachner <hans@bachner.priv.at> wrote:
>> Chris Townley schrieb am 31.05.2021 um 16:51:
>>> It was sometimes useful to add an incorrect character to avoid a symbol
>>> definition - for example deletes
>>
>> It still is useful. If you work at various customers' systems from time
>> to time, with setups varying from customer to customer, it is convenient
>> to type "$ direx /qual1 /qual2" to get displayed exactly the file
>> attributes you want for the files you want to see.
>>
>> I never found a customer system with "direx" defined as a symbol, and if
>> it would be defined it wouldn't do much harm. On the other side, most
>> customers have some dir* symbols defined, with and without abbreviaton.
>>
>> Similar for other commands like copy etc.
>
> That's one hell of an ugly hack and there should be a far better way to
> handle this other than resorting to that.
>
> Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
> allow the use of a leading underscore to skip any symbol translations ?
>
> Something like:
>
> $ _dir filename.dat
>
> would get you the "real" dir command without having to worry about
> symbols.
>
> I wonder if we could have something like this for VMS DCL ?

DCL does have:

$ set symbol/scope=(nolocal,noglobal)

Arne

In article <s93h3h$571$1@dont-email.me>, Stephen Hoffman
<seaohveh@hoffmanlabs.invalid> writes:

> On 2021-05-31 19:56:32 +0000, Phillip Helbig (undress to reply said:
>
> > In article <s938qc$7ed$1@dont-email.me>, Simon Clubley
> > <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
> >
> >> I wonder if we could have something like this for VMS DCL ?
> >
> > I've known about a trailing underscore for decades.
>
> There's no trailing-underscore syntax in DCL—any more than there's
> actually really any leading underscore syntax in logical name support
> with V4.0 and later—there's just stuff appended to the command that
> gets ignored by the parser but that also suppresses symbol matches.

Right. But it works. :-)

On 2021-05-31, Phillip Helbig (undress to reply) <helbig@asclothestro.multivax.de> wrote:
> In article <s938qc$7ed$1@dont-email.me>, Simon Clubley
><clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>
>> I wonder if we could have something like this for VMS DCL ?
>
> I've known about a trailing underscore for decades.
>

Very much _NOT_ the same thing.

Using a trailing underscore in that way is no different than typing
a trailing letter and is as big a hack.

You can't use a trailing underscore after only 3 characters and you have
to type 4 characters before you can append the underscore, just as you do
when appending a letter.

I wonder how much effort it would be to add support for a leading underscore.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 2021-05-31, Arne Vajhøj <arne@vajhoej.dk> wrote:
> On 5/31/2021 2:10 PM, Simon Clubley wrote:
>>
>> Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
>> allow the use of a leading underscore to skip any symbol translations ?
>>
>> Something like:
>>
>> $ _dir filename.dat
>>
>> would get you the "real" dir command without having to worry about
>> symbols.
>>
>> I wonder if we could have something like this for VMS DCL ?
>
> DCL does have:
>
> $ set symbol/scope=(nolocal,noglobal)
>

Not exactly viable for interactive use unfortunately. :-)

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 6/1/2021 8:11 AM, Simon Clubley wrote:
> On 2021-05-31, Arne Vajhøj <arne@vajhoej.dk> wrote:
>> On 5/31/2021 2:10 PM, Simon Clubley wrote:
>>>
>>> Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
>>> allow the use of a leading underscore to skip any symbol translations ?
>>>
>>> Something like:
>>>
>>> $ _dir filename.dat
>>>
>>> would get you the "real" dir command without having to worry about
>>> symbols.
>>>
>>> I wonder if we could have something like this for VMS DCL ?
>>
>> DCL does have:
>>
>> $ set symbol/scope=(nolocal,noglobal)
>
> Not exactly viable for interactive use unfortunately. :-)

So you do want some symbols.

:-)

$ set symbol/scope=notable

to disable symbols that shadows verbs in dcltables?

:-)

Arne

On 6/1/2021 8:42 AM, Arne Vajhøj wrote:
> On 6/1/2021 8:11 AM, Simon Clubley wrote:
>> On 2021-05-31, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>> On 5/31/2021 2:10 PM, Simon Clubley wrote:
>>>>
>>>> Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
>>>> allow the use of a leading underscore to skip any symbol translations ?
>>>>
>>>> Something like:
>>>>
>>>> $ _dir filename.dat
>>>>
>>>> would get you the "real" dir command without having to worry about
>>>> symbols.
>>>>
>>>> I wonder if we could have something like this for VMS DCL ?
>>>
>>> DCL does have:
>>>
>>> $ set symbol/scope=(nolocal,noglobal)
>>
>> Not exactly viable for interactive use unfortunately. :-)
>
> So you do want some symbols.
>
> :-)
>
> $ set symbol/scope=notable
>
> to disable symbols that shadows verbs in dcltables?
>
> :-)

No matter what then I don't see this as a major problem
worth much effort by VSI.

Arne

In article <s938qc$7ed$1@dont-email.me>, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>On 2021-05-31, Hans Bachner <hans@bachner.priv.at> wrote:
>> Chris Townley schrieb am 31.05.2021 um 16:51:
>>> It was sometimes useful to add an incorrect character to avoid a symbol
>>> definition - for example deletes
>>
>> It still is useful. If you work at various customers' systems from time
>> to time, with setups varying from customer to customer, it is convenient
>> to type "$ direx /qual1 /qual2" to get displayed exactly the file
>> attributes you want for the files you want to see.
>>
>> I never found a customer system with "direx" defined as a symbol, and if
>> it would be defined it wouldn't do much harm. On the other side, most
>> customers have some dir* symbols defined, with and without abbreviaton.
>>
>> Similar for other commands like copy etc.
>>
>
>That's one hell of an ugly hack and there should be a far better way to
>handle this other than resorting to that.
>
>Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
>allow the use of a leading underscore to skip any symbol translations ?
>
>Something like:
>
>$ _dir filename.dat
>
>would get you the "real" dir command without having to worry about
>symbols.
>
>I wonder if we could have something like this for VMS DCL ?

$ SET SYMBOL/VERB/SCOPE

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

In article <s9584n$rps$2@dont-email.me>, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>On 2021-05-31, Arne Vajh�j <arne@vajhoej.dk> wrote:
>> On 5/31/2021 2:10 PM, Simon Clubley wrote:
>>>
>>> Didn't one of the other DCL implementations (maybe RSTS/E V9 onwards)
>>> allow the use of a leading underscore to skip any symbol translations ?
>>>
>>> Something like:
>>>
>>> $ _dir filename.dat
>>>
>>> would get you the "real" dir command without having to worry about
>>> symbols.
>>>
>>> I wonder if we could have something like this for VMS DCL ?
>>
>> DCL does have:
>>
>> $ set symbol/scope=(nolocal,noglobal)
>>
>
>Not exactly viable for interactive use unfortunately. :-)

It works interactively. Try it.

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.

On 2021-06-01, VAXman- @SendSpamHere.ORG <VAXman-@SendSpamHere.ORG> wrote:
>
> $ SET SYMBOL/VERB/SCOPE
>

That works Brian even though it says it's only for use in command
procedures.

The annoying thing is that it appears to be global so you can't
override a DIR command while leaving a /CONFIRM on DELETE in place.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 2021-06-01, VAXman- @SendSpamHere.ORG <VAXman-@SendSpamHere.ORG> wrote:
> In article <s9584n$rps$2@dont-email.me>, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>>On 2021-05-31, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>>
>>> DCL does have:
>>>
>>> $ set symbol/scope=(nolocal,noglobal)
>>>
>>
>>Not exactly viable for interactive use unfortunately. :-)
>
> It works interactively. Try it.
>

It was more a comment on having to issue and then reverse the command
potentially multiple times during an interactive session. Using /VERB
does make it more viable thanks but that's still not perfect.

I wonder how many other people didn't know about this either given
the number of comments here about still placing a character at the
end of a command ?

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 6/1/2021 1:38 PM, Simon Clubley wrote:
> On 2021-06-01, VAXman- @SendSpamHere.ORG <VAXman-@SendSpamHere.ORG> wrote:
>> In article <s9584n$rps$2@dont-email.me>, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>>> On 2021-05-31, Arne Vajhøj <arne@vajhoej.dk> wrote:
>>>>
>>>> DCL does have:
>>>>
>>>> $ set symbol/scope=(nolocal,noglobal)
>>>>
>>>
>>> Not exactly viable for interactive use unfortunately. :-)
>>
>> It works interactively. Try it.
>
> It was more a comment on having to issue and then reverse the command
> potentially multiple times during an interactive session. Using /VERB
> does make it more viable thanks but that's still not perfect.

/VERB seems useful for COM files.

But I do not see a big difference interactively. It must be
extremely rare to use symbols in parameters/qualifier interactive.

Arne

On 2021-06-01 17:38:14 +0000, Simon Clubley said:

> I wonder how many other people didn't know about this either given the
> number of comments here about still placing a character at the end of a
> command ?

Local preference would be tools that better handle symbol scope, which
DCL does not do at all well.

I've used SET SYMBOL mostly defensively, though prefer to append a character.

Particularly prefer it interactively, as SET SYMBOL will invariably
blow out some other needed symbol.

DCL is a creature of the 1970s, with ~forty years of baggage.

--
Pure Personal Opinion | HoffmanLabs LLC

In article <s95r0p$v7i$2@dont-email.me>, Simon Clubley <clubley@remove_me.eisner.decus.org-Earth.UFP> writes:
>On 2021-06-01, VAXman- @SendSpamHere.ORG <VAXman-@SendSpamHere.ORG> wrote:
>>
>> $ SET SYMBOL/VERB/SCOPE
>>
>
>That works Brian even though it says it's only for use in command
>procedures.
>
>The annoying thing is that it appears to be global so you can't
>override a DIR command while leaving a /CONFIRM on DELETE in place.

Redefine the DELETE command and make /CONFIRM a default qualifier.

--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG

I speak to machines with the voice of humanity.