Experimental Physics and
Industrial Control System

"Johnson, Andrew N. via Core-talk" <core-talk at aps.anl.gov> · Wed, 14 Apr 2021 18:45:45 +0000

On Apr 14, 2021, at 12:32 PM, Ralph Lange via Core-talk <core-talk at aps.anl.gov> wrote:
> 
>> Do we need some syntax in macros to prevent them from escaping back-slashes when expanding values?  I’m not sure how easy it would be to implement that, but it might be possible. Any other suggestions on what might be changed?
> 
> Looking at the macLib code, I was getting the expression that the original intention might have been to do it right.
>     /* discard quotes and escapes if level is > 0 (i.e. if these aren't
>        the user's quotes and escapes) */
> 
> Or not? (Shouldn't quotes and escapes only be discarded when they are the user's, and kept through replacements?)
> Well, at least there was an intention.

Read the macLibNOTES file in the libCom/macLib directory. Unfortunately it’s a bit self-referential in places as it was originally meant to be run on itself so not always immediately obvious.

> What if using a variable wouldn't change its content? After the second line, I would expect PP and QQ to be the same.
> epics> epicsEnvSet PP a\\b\\c
> epics> epicsEnvSet QQ $(PP)
> epics> epicsEnvShow
> [...]
> PP=a\b\c
> QQ=abc
> 
> We could maybe add another bracket character pair for verbatim usage.
> epics> echo $[eeed]
> 
> I am afraid that changing the current behaviour - even if it was "correcting" it - would add a pretty high risk of breaking things for users.

I agree, I wouldn’t do that; I was thinking of something more on the lines of ${\PP} or ${'PP'} which both currently return the same as ${PP}. I’d rather not add more start/end character pairs to the macro syntax. What would be simplest depends on whether I can detect the modifiers in the right place inside the code, the leading back-slash might be simplest to implement. We'd have to be careful about how to expand things like ${\${PP}=${Q}} though.

- Andrew

-- 
Complexity comes for free, simplicity you have to work for.

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System