noaa-ocs-s100 / s100py Goto Github PK

The S-102 specification is node based (rather than pixel based).

The geotransform provided by GDAL is pixel (cell) based.

This has the following impact:

1. The corner point provided by the GDAL geotransform needs to be offset by + 1/2 X resolution and - 1/2 Y resolution.
2. The offset to other corners should be calculated as reference_corner + resolution * (N -1)

Data coding format 7 supports TINs and we just got a clarification that the indices are zero based, so we could add proper support. Do we use MDAL to make a general reader utility?

I note in your docs:

"""
This package requires the GDAL Python bindings be present, so it usually can't just be installed using pip install gdal. We recommend installing GDAL either through a package manager (e.g. conda, apt, yum, pacman) or by compiling from scratch. Miniconda is probably the easiest method.
"""
It seems it would be helpful to your users to make a conda package available for s100py itself, so folks could simply do

conda install -c conda-forge s100py

It would be pretty straightforward to add a new conda-forge "feedstock", and once in, it's pretty self maintaining.

Let me know if you want some help -- we've done a fair bit of this to support NOAA / ORR projects, notably PyGNOME.

HMM -- it looks like there hasn't been an update it a while -- maybe this could wait 'till the next release.

(https://github.com/NOAA-ORR-ERD/PyGnome)

Use pyinstaller to make .exe so hydrographic offices can use/try the capabilities of s100py. Remember to use the smaller numpy (no MKL) to make the executable much smaller on Windows at least.

Add support and samples for data coding formats 1, 8 - fixed stations.

Add command line usage for the utility functions, like bag/geotiff to S102 or similar for S104, S111. Do we use "fire" or built-in argparse to avoid adding a dependency?