Experimental Physics and
Industrial Control System

Benjamin Franksen <[email protected]> · Thu, 24 Sep 2015 12:20:31 +0200

On 09/23/2015 09:37 PM, Ralph Lange wrote:
>> I'm starting to think that being explicit about which files go into
>> the output file might be a better approach as Torsten suggested.
>> The IOC developer then has the final say as to what gets run, and
>> nobody else can just add a file to the global snippets directory to
>> have their code run on all IOCs — that could be really bad. Of
>> course you can still use wildcards in the list if desired for your
>> site.
> 
> Using directories (or wildcards) is the whole point of my use case. 
> If a new version of my global support module needs an additional line
> in the st.cmd, I want its RPM to drop it in the global snippets dir,
> rebuild my ~500 IOCs and be done with it. I do *not* want to update
> 500 Makefiles. I want my configuration database exporter to populate
> a single directory with all generated st.cmd content (from the
> database) for a specific IOC, and have it merged with the developer's
> special snippets (under version control) without having to change the
> version controlled Makefile.

I find this argument pretty convincing.

Here at BESSY we create st.cmd files together with their binaries and
dbd files from a common set of templates (for st.cmd files) and
hierarchical information we keep in data structures in perl modules.
There is one of these per IOC. The scenario Ralph is alluding to we
solve by allowing the perl modules to be based on other perl modules of
the same sort. Thus we have e.g. one MultiCAN.pm which contains every
piece of information that is common to all IOCs that use CAN; the IOC.pm
import them and the generic mechanism takes care to merge the
information from all imported modules before fleshing out the
template(s) and creating the Makefile snippets for the libraries etc.

Our mechanism is quite specifically tailored to the task, and thus not
as general as the one Ralph wants to create. I wonder if it could be
adapted to make use of Ralph's more general solution...

Cheers
Ben
-- 
"Make it so they have to reboot after every typo." ― Scott Adams

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System