Does threddsIso handle the 'source' CF attribute?

[mwengren]

See: http://cfconventions.org/cf-conventions/v1.6.0/cf-conventions.html#description-of-file-contents.

I did a search in the current version of UnidataDD2MI.xsl and couldn't find it called out in the variables section at the top.

Since this is mentioned as a critical file description variable in CF, shouldn't it be parsed and output in the ISO XML somewhere?

[lukecampbell]

As of ACDD 1.1 there was no mapping or use of the global source attribute.

http://wiki.esipfed.org/index.php/Attribute_Convention_for_Data_Discovery_1-1

source is defined in ACDD 1.3 but I don't believe a mapping has been done from ACDD 1.3 to ISO-19115-2 to date.

[mwengren]

Hey @lukecampbell. I read down to this comment in issue #7 noticed that is is included in the new mapping spreadsheet.

I don't know how long source has been part of CF, but since it's alongside the other high level identification variables I would guess awhile. Might have just been overlooked in earlier ncISO versions.

Hopefully, it will be included in the upcoming release! (this will help us tremendously at IOOS to identify model datasets by some automated fashion for potential inclusion in EDS via the Catalog, at least, if CKAN indexes where it ends up in the ISO XML for search/filter, a separate issue potentially). See: https://github.com/ioos/catalog/issues/40

[ocefpaf]

Just to xref. another issue here: this would really help with https://github.com/ioos/catalog/issues/40

[lesserwhirls]

Tagging @noaaroland - do the transforms in https://github.com/NOAA-PMEL/uafnciso handle the source attribute?

[noaaroland]

It does now: https://github.com/NOAA-PMEL/uafnciso/releases/tag/2.3.6

[noaaroland]

The CDN versions of the transforms I use in the code are: https://cdn.jsdelivr.net/gh/NOAA-PMEL/uafnciso@a9f3dceb0dbc092f8d7f5153ceeb58af2d30edc4/transforms/UnidataDD2MI.xsl and https://cdn.jsdelivr.net/gh/NOAA-PMEL/uafnciso@fdb7f86515c21a8b5c087978975addf9ad5d0027/transforms/UnidataDDCount-HTML.xsl

[mwengren]

@noaaroland Thanks for updating the NOAA-PMEL/uafnciso repo with fixes!

Since this is a fork of the original ncISO, I presume it won't be included in any new TDS releases, correct?

We can point our data providers to use the uafnciso release JAR files as an alternative to generate ISO files from TDS. If, that is, we are able to leverage that for filtering capability based on the values populated in 'source' - one of the original goals of ioos/catalog#40.

Not sure if this should be closed or left open until this functionality is ported here if that's desired. Feel free to close if not.

[lesserwhirls]

I think we are still in a place where there have been changes made to the transforms in this repository, and there have been changes made to the transforms in the NOAA-PMEL repository, and they have diverged enough to where I am unable to make a call on what stays and what goes (plus, I'm not exactly proficient with xslt stuff, so you probably don't want me doing it anyways). It'd be great if we could settle around a canonical set, and I'd be in favor of adopting them here (not my call, however).

[dneufeldcu]

I agree with consolidating the changes in the Unidata threddisIso repo, I think that was the intent behind merging the client code into that repository as well. Not sure who is best positioned to review and merge the XSLT at this point? @amilan17 @noaaroland

[zdefne-usgs]

We have also made some changes to XSLT to address some of the issues such as: https://github.com/Unidata/threddsIso/issues/41 The most recent copy can be found at: https://github.com/zdefne-usgs/ocean-iso-metadata/tree/master/XSL/usgs_current