search

LANL-Seismoacoustics / e1

Python interface to the e1 compression format.
Other
1 stars 2 forks source link

Wrap compression functions in the C library #4

Open jkmacc-LANL opened 3 years ago

jkmacc-LANL commented 3 years ago

Currently, only e_decomp in the C library is wrapped, but there are a number of other useful functions. For example, I've not exposed any compression routines. This was partly due to time constraints (and I'm not convinced the world needs more e1 data 😉 ), but certainly this could be done.

samuelchodur commented 9 months ago

Could you please confirm that a big-endian machine is required to compress using e1? If so, any suggestions on how to develop/test the wrapped code? I went down the road of using QEMU to create a big-endian emulation but TBH it was not as straight-forward as I had hoped. I am having trouble finding an OS (and the patience) to do the emulation. I tried FreeBSD and AlmaLinux. I didn't have the gumption to learn how to install an updated version of Python in FreeBSD nor even get the AlmaLinux installation completed.

jkmacc-LANL commented 9 months ago

@samuelchodur We use this library on little endian machines (not via Python, but the same library). Big endian should not be necessary.

jkmacc-LANL commented 9 months ago

@samuelchodur, have you tried this on little-endian?

samuelchodur commented 9 months ago

@samuelchodur We use this library on little endian machines (not via Python, but the same library). Big endian should not be necessary. @samuelchodur, have you tried this on little-endian?

Thanks for the information about using the C library. I was not getting the expected results when compressing, so I was hoping lack of big endian was the issue :smiley:. Admittedly, I haven't used the ctypes library very much (nor written much C) so when I have some time I will try some more to get compression from Python working.

jkmacc-LANL commented 9 months ago

That'd be awesome! ctypes wasn't too hard to work with👍