segyio_header_buf

segyio header Field.buf always big endian?

Investigation into the segyio.field.Field.buf byteorder. It was expected that the raw bytes of the header would match the byteorder of the segy file, i.e. endian parameter in the segyio.open() context manager. However, it doesn't appear to do so. Two segy files were created: (1) simply reads in the first five traces and writes out the bytes with the original byteorder, and (2) changes the byteorder to little endian (system native byteorder). When converting the buf raw bytes class variable to python integer objects, the little endian file required big endian decoding to get the correct value.

• create_example_segy_big_ibm.py: creates first 5 traces with original byteorder, no formatting
• create_example_segy_little_ieee.py: creates first 5 traces with little endian byteorder

The examples use segytools to parse the raw segy binary data for reading and reformatting in the case of the little endian segy.

open source data

dataunderground penobscot - Penobscot_3D_gathers_part1/3D_gathers_pstm_nmo_X1001.sgy

version

import segyio
print(segyio.__version__)
# 1.9.3