nanocomp / harminv Goto Github PK

===>  Building for harminv-1.4.2
gmake[1]: Entering directory '/usr/ports/science/harminv/work/harminv-1.4.2'
gmake[1]: *** No rule to make target 'copyright.h', needed by 'all'.  Stop.
gmake[1]: Leaving directory '/usr/ports/science/harminv/work/harminv-1.4.2'

Version: 1.4.2
clang-14
FreeBSD 13.1

I made my best guess as to autogen, did a subsequent configure, and now make runs into a missing copyright.h file.

Dear Steven,

coming from the Fortran/Matlab world I decided to make Julia to my language of choice. Recently, I have discovered harminv which I would really like to use. However, I am clueless about anything related to c (compiling, running etc). Do you have an idea on how to make your library run under Julia ?? I am on windows 7 btw.

Thank you

Hi,

It seems that sines-test is failing on my computer

Example output from harminv

$ ./sines 0.1+0.01i 0.08+0.001i | ./harminv 0.05-0.15
frequency, decay constant, Q, amplitude, phase, error

(yes no frequencies were found)

It seems to work if we change the range in which we look for frequencies

./sines 0.1+0.01i 0.08+0.001i | ./harminv 0.05-0.16
frequency, decay constant, Q, amplitude, phase, error
0.08, 1.000000e-03, 251.327, 2, 3.14159, 8.834857e-16
0.1, 1.000000e-02, 31.4159, 1, 4.78299e-14, 1.692517e-15

both frequencies correctly found!!!

Any clue as to why this might happen?

Trying to install harminv on macOS Sierra (10.12.6). Cloned harminv from GitHub.com...

git clone https://github.com/stevengj/harminv
cd ./harminv

sh autogens.sh runs to completion.
./configure runs to completion.
make stops with the following message,
make: *** No rule to make target copyright.h', needed by all'. Stop.

A few days ago I installed harminv on another iMac, running macOS Sierra, from the tarball without incident.

I use this software to calculate the frequency of a signal that is superimposed from three sinusoidal signals
y=sin(2*pi*t*12)+sin(2*pi*t*12.1)+sin(2*pi*t*12)
The length of the signal is 5.23s, the sampling interval is 0.01s, and the number of sampling points is 524；
I use the harminv :
harminv -t -0.01 10-15 < input.data
the result is:
frequency:
-12.1882 -12.0181 12.018 12.182
The frequency of 12.1 is missing，I wonder how harminv‘s parameter should be set to calculate these three frequencies

9f94898 has changed the api. (Now released with version 1.4)

The arguments of harminv_get_omega and harminv_get_amplitude have changed to require a harminv_complex as first argument.

Was this change necessary? Do you expect it to be permanent?

I can update my python interface (aaren/pharminv) to match the change without much problem, but I'm not sure whether you intended this change or not.

(harminv 1.3.1 is distributed with Debian Jessie)

The documentation states "Harminv comes with a GNU-style configure script, " which is not true.

I tried running autogen.sh but I can not figure out how to supply my environment to it. There doesn't seem to be a --help options or so. Could you document the possible flags?