[Info-vax] C99 updates to CRTL

[Stephen Hoffman]

On 2019-07-29 17:50:54 +0000, John Reagan said:

> On Monday, July 29, 2019 at 12:21:28 PM UTC-4, Stephen Hoffman wrote:
>> On 2019-07-27 20:45:09 +0000, Simon Clubley said:
>> 
>>> Arne, are you _trying_ to trigger Stephen ? :-)
>>> 
>>> It's bad enough using logicals to alter behaviour at runtime, but at 
>>> compile time as well ? :-)
>> 
>> There's already at least one logical name around that effects the 
>> compile-time behavior.
>> 
>> One existing logical name effects the compile-time resolution of the 
>> external references within the compiler.
> 
> Do you have a reference?

The DEC/Compaq/HP/HPE/VSI C compiler opens up the C RTL and searches it 
for a list of entry points, and that processing can be redirected by 
logical name.  I don't recall off-hand if it was the DECC$SHR logical 
name, or if some other was used.

I've had a few occasions to do this redirection, but not in recent 
times. This redirection is most common when working cross-version, 
something which is somewhat unsupported.

It wouldn't surprise me that there are other logical names latent and 
lurking in the compilation process.

App management and configuration with logical names—other than for 
device redirection, and that usage is only marginally better than app 
config data—is among the more fetid coding practices and endemic on 
OpenVMS.

Logical names as configuration data storage?  Hacks all the way down. 
Isolation by convention. Parsing by assumption.  Architecture by 
accretion. Documentation by omission.

But I'm feeling polite today.  And yes, I have used logical names for 
configuration data. Having encountered better configuration data 
storage schemes else-platform, now usually whilst holding my nose, too.


-- 
Pure Personal Opinion | HoffmanLabs LLC