[Users] Suggested change in Carpet output file names

Erik Schnetter schnetter at cct.lsu.edu
Thu May 16 22:13:14 CDT 2013


I want to add that this change would be implemented in the same way as
previous changes: There would be a parameter that chooses which file name
format to use, and the default setting for this parameter would (at least
initially) be backward compatible, i.e. one would explicitly need to set
this parameter to see this change.

Regarding the new file name, I prefer a single dot "." instead of the
double colon "::". A single dot is commonly used to access fields in a
structure or member in a class, so that thorn.variable or thorn.group make
sense. I don't see a reason to use two separating characters. Cactus chose
:: because this is what C++ uses for namespaces; if we want to follow Java
or Python instead, we would use a single dot instead.

-erik



On Thu, May 16, 2013 at 9:25 PM, Frank Loeffler <knarf at cct.lsu.edu> wrote:

> Hi,
>
> Several authors debated the plan to change the colons "::" in especially
> Carpet output file names to something else. One stated reason is the
> incompatibility of this and Windows, Samba and related systems, another
> is inconveniences these cause for MacOS users.
>
> After considering several options we seem to converge on a replacement
> of "::" by "--" (double minus signs) as delimiter (see
> https://trac.einsteintoolkit.org/ticket/1338 for the full discussion).
>
> This would, for example change a filename from
>
>   hydrobase::w_lorentz.maximum.asc
>
>  to
>
>   hydrobase--w_lorentz.maximum.asc
>
> However small this change is, it would affect the vast majority of
> users. Thus, we would like to get feedback before actually going
> forward. So: if you like to have your opinion heard, now is the best
> time to accomplish that before it is "too late" (whatever direction it
> might be).
>
> Frank
>
>
> _______________________________________________
> Users mailing list
> Users at cactuscode.org
> http://cactuscode.org/mailman/listinfo/users
>
>


-- 
Erik Schnetter <schnetter at cct.lsu.edu>
http://www.perimeterinstitute.ca/personal/eschnetter/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.einsteintoolkit.org/pipermail/users/attachments/20130516/8930763f/attachment.html 


More information about the Users mailing list