[Meep-discuss] FT fields output

[Vincent Wheeler]

Hi Meep users,

I would like to calculate the Fourier transformed fields at every point in
space.  According to this old email:
https://www.mail-archive.com/meep-discuss@ab-initio.mit.edu/msg00108.html
and the current documentation, there does not seem to be a built-in way to
do this conveniently without digging into the C++ source code, a prospect I
briefly tried but would rather like to avoid.

In short, I think it is possible to get all of the FT fields from the h5
file produced by the libctl save-flux routine, and I am looking for
confirmation that it is possible and for advice on doing so.  A more
detailed discussion and question follows for those who care to read it.

Let's pick a particular direction to simplify the discussion with concrete
names of components, say X.  Also, suppose I have a 40x40x40 grid of
solution nodes and I have chosen nfreq = 6.  A file saved using save-flux
should contain four field components, each with a real and complex part, in
order to calculate the flux in a direction normal to the chosen direction.
If the h5 file produced by save-flux indeed has all the information I need,
it should have (40*40*40)*(4*2)*6 numbers in it: the total number of nodes
times the number of (complex) field components per node times the number of
frequencies.  When I check out the contents of one such h5 file I can see
that data sets with the names "ey_dft" and "hz_dft" are inside.  This is
rather unexpected since I would think there is also "ez_dft" and "hy_dft."
However, if I look at the size of these data sets I get size("ey_dft") =
(1536000,1), a one-dimensional array with 1536000 elements.  If it is truly
just the fourier transform of the y component of the electric field, then
it would have a size equal to (40*40*40)*2*6 = 768000, yet it is exactly
twice that.  Since I have two data sets of this size, it would seem all of
the data I need is contained in the h5 file output by save-flux.

This is encouraging, but leads me to my questions: how is the FT field data
organized in the h5 files generated by save-flux?  Is it done in a
meaningful way, so that I can write a little script to organize it into
something more convenient than two large one-dimensional arrays?  Also, is
it organized differently when using meep with MPI (as I will be)?

Thanks for any help.

Vince

===
Vincent M. Wheeler, PhD
Research School of Engineering
Australian National University
Mobile: +61 4 8459 5779 | Office: +61 2 6125 9204 <+61%202%206125%209204>
vincent.whee...@anu.edu.au
___
meep-discuss mailing list
meep-discuss@ab-initio.mit.edu
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss

[Meep-discuss] _rl_get_keymap_name with Sierra OS?

[Priscilla Kelly]

Hello,

I am installing meep according to 

https://www.mail-archive.com/meep-discuss@ab-initio.mit.edu/msg05812.html 


and I have made the configure file without issue. 

However, when I type make, I get the following error:

Making install in guile-readline
/Applications/Xcode.app/Contents/Developer/usr/bin/make  install-am
  CCLD libguilereadline-v-18.la 
Undefined symbols for architecture x86_64:
  "_rl_get_keymap_name", referenced from:
  _scm_init_readline in readline.o
ld: symbol(s) not found for architecture x86_64
clang: error: linker command failed with exit code 1 (use -v to see invocation)
make[3]: *** [libguilereadline-v-18.la ] 
Error 1
make[2]: *** [install] Error 2
make[1]: *** [install-recursive] Error 1
make: *** [install] Error 2

Based on older threads, I know that the built in apple code for readline isn’t 
going to work but I can’t change the name or move usr/lib/libreadline.dylib 
with the new update.

How can I work around this?

Thank you,
Priscilla___
meep-discuss mailing list
meep-discuss@ab-initio.mit.edu
http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/meep-discuss