GBIF EML Profile

[nicokant]

Add support for importing and transforming from GBIF EML profile of metadata.

fixes #259

[nicokant]

I'm not sure of how to deal with additional metadata that this format requires, @tomkralidis do you have any advice on how to deal with such values and to provide some validation?

[tomkralidis]

I'm not sure of how to deal with additional metadata that this format requires, @tomkralidis do you have any advice on how to deal with such values and to provide some validation?

(sorry for the delay/late response)

Simply add the values in the MCF as you see fit. You will want to document them and add a schema definition. Using WCMP2 as an example:

• https://github.com/geopython/pygeometa/blob/master/docs/content/reference/extensions/wmo-wcmp2.md
• https://github.com/geopython/pygeometa/blob/master/pygeometa/schemas/mcf/wmo-wcmp2.yaml

As for validation, would this mean validating the MCF with the EML extensions, or the EML output, or both? We currently only have core MCF validation, but could add validation as needed. Please feel free to chime in on #264.

[nicokant]

Thank you :)

I meant validating the MCF with EML extension.

My main focus now is to allow converting a EML document to other metadata formats (so mostly the import_ method), rather than producing a fully validated EML. But I'd still like to keep somewhere the additional metadata.
Would it be ok if this PR only provides an import feature rather than a bidirectional conversion?

[tomkralidis]

Would it be ok if this PR only provides an import feature rather than a bidirectional conversion?

Yes, this is fine.

[tomkralidis]

@nicokant let us know when this PR is ready for review.

[pvgenuchten]

Hi @nicokant, nice work here, set pr as ready for review, when you want us to have a look, thanx

[nicokant]

@pvgenuchten sorry it's been a while, but I've rebased now and the PR is ready for a review :)

[pvgenuchten]

Hi @nicokant, thanx for the work. One comment though, if you use mcf[‘identification’][‘doi’] python assumes this value will always be available, consider to use mcf[‘identification’].get(‘doi’) instead.

[nicokant]

Hi @nicokant, thanx for the work. One comment though, if you use mcf[‘identification’][‘doi’] python assumes this value will always be available, consider to use mcf[‘identification’].get(‘doi’) instead.

Thank you for the feedback, I noticed with a search that this pattern is also used in wmo_cmp so you might want to fix it also there

[tomkralidis]

CI failures unrelated.