#1875: SetMask_SphericalSurface: retain mask -------------------------------------+-------------------------------------- Reporter: cott@… | Owner: Type: enhancement | Status: review Priority: unset | Milestone: Component: Other | Version: development version Resolution: | Keywords: -------------------------------------+-------------------------------------- Changes (by rhaas):

* status: new => review

#1875: SetMask_SphericalSurface: retain mask -------------------------------------+-------------------------------------- Reporter: cott@… | Owner: Type: enhancement | Status: review Priority: unset | Milestone: Component: Other | Version: development version Resolution: | Keywords: -------------------------------------+--------------------------------------

Comment (by knarf):

I think I used the following for the case you describe:

{{{ AHFinderdirect::reset_horizon_after_not_finding[0] = "no" }}}

Would that also work for you (haven't tried now it this still resets the spherical surface)?

Otherwise: it would be good to add this as a (steerable) parameter, so users can choose between the old and the new mechanism. I could see the "new" mechanics even be the default, as this is most likely what a user wants.

#1875: SetMask_SphericalSurface: retain mask -------------------------------------+-------------------------------------- Reporter: cott@… | Owner: rhaas Type: enhancement | Status: assigned Priority: unset | Milestone: Component: Other | Version: development version Resolution: | Keywords: -------------------------------------+--------------------------------------

Comment (by knarf):

I made a few changes, addressing Roland's comments. Roland (or any other dev): ready to go like this? I personally don't think a parameter for this is really necessary, unless there is actually a user who wants it otherwise.

Concerning Roland's comment "rather than always expecting the horizon to be found, it will use possibly old horizon data forever": I don't see how this could happen. The thorn always initializes the mask to 'normal', and only overwrites later with 'excises' if a surface is active. If there isn't a valid surface anymore, I would expect the current code to reset everything to 'normal'. Am I missing something?

#1875: SetMask_SphericalSurface: retain mask -------------------------------------+-------------------------------------- Reporter: cott@… | Owner: rhaas Type: enhancement | Status: review Priority: unset | Milestone: Component: Other | Version: development version Resolution: | Keywords: -------------------------------------+--------------------------------------

Comment (by rhaas):

The current code will do what you describe. The proposed code would use the last found surface information (since it ignores sf_valid which AHFinderDirect sets). AHFinderDirect behaves like this (src/driver/BH_diagnostics.cc) {{{ // only try to copy AH info if we've found AHs at this time level if (! AH_data.search_flag) { sf_valid[surface_number] = 0; return; } // did we actually *find* this horizon? if (! AH_data.found_flag) { sf_valid[surface_number] = -1; return; } sf_valid [surface_number] = 1; }}} ie it will set {{{sf_valid}}} to -1 if an AH was looked for but not found, so that current code that only accepts surfaces for which {{{sf_valid >= 0}}} will reject the surface. One *may* instead possibly use {{{sf_active[] && sf_valid[]}}} which should works about as well as the proposed change though it may differ in the time interval before the AH is found for the first time (but is already searched for).