Use nccsv for example datasets for simplicity/visbility #51
Conversation
Using the text-based nccsv format allows us to the self describing/embedded metadata features of NetCDF while also allowing easy visibility and browsing of the source data (text is more accessible than binary files). https://coastwatch.pfeg.noaa.gov/erddap/download/NCCSV.html
|
My concern with this is IOOS does link to these files as Gold Standard Example datasets for the IOOS Metadata Profile and I don't think we want to be recommending nccsv as the standard format. https://ioos.github.io/ioos-metadata/ioos-metadata-profile-v1-2.html#gold-standard-example-datasets IMO we should be exemplifying how to load datasets from a variety of formats. So, I propose that we only change one of these netCDF files to nccsv, but leave the rest as nc. If it would be beneficial to the community, we can provide a .cdl next to the .nc for readability on GH. |
|
@MathewBiddle That’s a valid concern. I hate violating DRY but what if we kept both nccsv and nc copies of the example datasets here, made the nc files the default, and included a script to generate the nc files from the nccsv files? That way we could use the nccsv files as the source of truth and cleanly track changes to them via git, but also demonstrate running ERDDAP with standard nc datasets. |
|
That seems like a viable approach. I'm onboard with it, and interested in what the script would look like. |
|
@srstsavage FYI The default branch has been renamed! See #52 |
As an alternate approach to using nccsv source of truth datasets in #51, instead we store text yaml representations of the gold standard datasets based on nco-json output (nco-yaml if you will, which is more readable than json due to lack of brackets). Also included is a script to easily (re-)generate the yaml representations from the source of truth NetCDF files, and to check if the yaml representations are up to date (possibly used in a future pre-commit hook, GitHub action, etc).
As an alternate approach to using nccsv source of truth datasets in #51, instead we store text yaml representations of the gold standard datasets based on nco-json output (nco-yaml if you will, which is more readable than json due to lack of brackets). Also included is a script to easily (re-)generate the yaml representations from the source of truth NetCDF files, and to check if the yaml representations are up to date (possibly used in a future pre-commit hook, GitHub action, etc).
|
Closing in favor of the CDL approach in #70 |
Using the text-based nccsv format allows us to the self describing/embedded metadata features of NetCDF while also allowing easy visibility and browsing of the source data (text is more accessible than binary files).
https://coastwatch.pfeg.noaa.gov/erddap/download/NCCSV.html