[Yt-dev] HaloFinding thoughts

[Matthew Turk]

Hi Stephen,

I think adding a new field to be parsed, as well as a new column, is
worse than just adding a new column.  However, again I reiterate my
opposition -- the people who will be outputting the final field will
be in the vast minority, and I think really we should just output a
map file *as needed*.

When YT moves to HDF1.8, we can use the symlink capabilities of HDF5
1.8 to avoid the mapping file.  But I think a single map file,
catering to the specific problem of outputting multiple particle
files, is the way to go.  I think we are verging on bike shedding at
this point.

-Matt

On Mon, May 11, 2009 at 7:42 AM, Stephen Skory <stephenskory at yahoo.com> wrote:
>
> Hi Britton & Matt,
>
>> The HaloProfiler is wired to read in the HopAnalysis.out file in its current form.  The name isn't the most important thing, but I would still be in favor of leaving that alone and using new file names for new systems and functionality.
>
> The way I see it is it's one fewer file that needs to be kept track of, and it's relevant information to add to each halo line. The way I have it now is it adds a column to the end of each line with the h5 file name if and only if .write_out is called by .write_particle_lists. At the top of the file, it now looks like:
>
> # HALOS FOUND WITH (HOP/FOF). h5 FILENAME COLUMN: (True/False)
>
> A file created by calling .write_out, like is done now, is unchanged from before except for the line above. Because the h5 filename is the last column, reading in the halo values using .split() means that HaloProfiler will work just fine with a file that has the h5 filenames anyway. The haloes file created by .write_particle_lists will have the same prefix name as the h5 files, making it easy to relate the files together.
>
> I really think this is the more elegant solution. I can change it back and add a separate file if both of you still feel that that's the better solution. Of course I haven't comitted any of this, and I'm in no hurry to do so.
>
>  _______________________________________________________
> sskory at physics.ucsd.edu           o__  Stephen Skory
> http://physics.ucsd.edu/~sskory/ _.>/ _Graduate Student
> ________________________________(_)_\(_)_______________
>
> _______________________________________________
> Yt-dev mailing list
> Yt-dev at lists.spacepope.org
> http://lists.spacepope.org/listinfo.cgi/yt-dev-spacepope.org
>