You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I want to get the biotoolsSchema paper out of the door and the main outstanding thing is elegant handling of the serialisation formats:
- biotoolsSchema-compliant XML, JSON and YAML formats
- shims for interconverting these
- validation against biotoolsSchema / biotoolsSchemaJ
... all of this also useful for ETP "ecosystem".
Until now (for purposes of XML support in bio.tools) I'd created shims as XSLT transforms, which worked fine. But this gets unmanageable for any-to-any interconversion of XML, JSON and YAML - especially as we might need JSON-LD, and perhaps other formats in future (codemeta, whatever).
So I figure the elegant solution is to write a single utility in python (biotoolsshim.py) for all our tool metadata shimming/transformation needs. Now, from my newbie-python perspective this isn't looking too hard BUT a spanner in the works is the <Summary> and <Labels> elements in the XSD - these complicate what (I think) would otherwise be a fairly straightforward exercise. I could program around it, but it seems daft to do so!
<Summary> and <Labels> are purely "organisational" (do not appear in biotoolsSchemaJ as documented here), so I'm thinking to release biotoolsSchema 3.3.0 which drops these elements, paving the way for easier coding.
PS. The XSD I think has mostly served its purpose, and expect in future to focus on the JSON schema variant (biotoolsSchemaJ).
I'm writing to get your input on this - any thoughts or suggestions much appreciated!
Thanks a million
The text was updated successfully, but these errors were encountered:
Chaps (@hansioan @hmenager @matuskalas @bgruening @albangaignard) - please help.
I want to get the biotoolsSchema paper out of the door and the main outstanding thing is elegant handling of the serialisation formats:
- biotoolsSchema-compliant XML, JSON and YAML formats
- shims for interconverting these
- validation against biotoolsSchema / biotoolsSchemaJ
... all of this also useful for ETP "ecosystem".
Until now (for purposes of XML support in bio.tools) I'd created shims as XSLT transforms, which worked fine. But this gets unmanageable for any-to-any interconversion of XML, JSON and YAML - especially as we might need JSON-LD, and perhaps other formats in future (codemeta, whatever).
So I figure the elegant solution is to write a single utility in python (biotoolsshim.py) for all our tool metadata shimming/transformation needs. Now, from my newbie-python perspective this isn't looking too hard BUT a spanner in the works is the
<Summary>
and<Labels>
elements in the XSD - these complicate what (I think) would otherwise be a fairly straightforward exercise. I could program around it, but it seems daft to do so!<Summary>
and<Labels>
are purely "organisational" (do not appear in biotoolsSchemaJ as documented here), so I'm thinking to release biotoolsSchema 3.3.0 which drops these elements, paving the way for easier coding.PS. The XSD I think has mostly served its purpose, and expect in future to focus on the JSON schema variant (biotoolsSchemaJ).
I'm writing to get your input on this - any thoughts or suggestions much appreciated!
Thanks a million
The text was updated successfully, but these errors were encountered: