[Feature Request] Restructuring of the publication page for a better overview #5

Closed CMR248 closed 1 year ago

[x] Headline: "Welcome to the ARC Data Publication Service"
[x] Visually separate the headline from the rest of the web page with a thick separator line
[x] Dividing the remaining webpage into the publication-describing part (left, wide, white) and the administrative part (right, less wide, grey)
[x] Concerning the publication-describing part, the layout could look like common publication standards (e.g. Zenodo)
DATA PUBLICATION
Date
Name of the Publication Server (ARChive)
Investigation Title (which is not necessarily the title of the Repository)
List of Authors including Mailadresses Affiliation ** ORCiD (with Symbol)
Investigation Description
[x] The administrative part includes function buttons for publishing and hint fields about the underlying GitLab repository
No User as this might be confusing
Publish ARC Button at the top
In order that all authors are listed, it would be good to query exactly that again via a checkbox.
Instead of displaying Project, display "Source of this Data Publication is ".
The unsuccessful critical tests should be displayed underneath, with a text like "Project not ready to publish" and afterwards "Make sure that the following criteria are met:". The successful ones should be omitted, as they tend to cause confusion. An explanation of what should be adjusted to correct the pipeline would be nice ** Here one could refer to an area in the knowledge base, which goes into detail concerning individual messages, or @Brilator
[x] If Terms and Conditions/Author List checkboxes are not checked, the following should happen ** Link Terms and Conditions properly
[x] A "back" button would be desirable to return to the repo

For a better visualization and estimation of the font sizes find images below

Restructured Page

Terms and Conditions/Authors list weren´t checked

Tests failed

(Part of) Zenodo Publication Page

@omaus @HLWeil orcid and affiliation must be added to metadata.json , so that this items can be displayed on the webpage

@omaus @HLWeil orcid and affiliation must be added to metadata.json , so that this items can be displayed on the webpage

@Zerskk

We will provide email address, ORCID, and affiliation name but cannot provide affiliation ID since this has to be done by Invenio:

To me, it seems like you'd need to implement a parser that crawls a CV for a specific name and then fills in the ID if the name exists, else... I don't know, creates a random ID and updates the CV?

CC: @kMutagene

For context, here is an overview of the current (as of 2023-07-11) validation and metadata creation pipeline:

Repos:

Jobs:

Note that arc-to-invenio only runs when arc-validate succeeds without critical tests failing

flowchart LR

arc[ARC on GitLab]

arc-export[["arc-export"]]
arc-validate[["arc-validate"]]
arc-to-invenio[["arc-to-invenio"]]

arc_json("arc.json")
arc_validate_result("arc-validate-results.xml")
arc_quality("arc-quality.svg ")
invenio_metadata("metadata.json")

arc --> arc-export --> arc_json

arc --> arc-validate 
arc-validate --> arc_validate_result
arc-validate --> arc_quality

arc --> arc-to-invenio --> invenio_metadata

Files:

arc.json: contains the full arc metadata (e.g. all isa files in the arc in json format). Intended to be used for 'inspecting the arc' without downloading it, e.g. for a preview on the invenio page when the arc is published.
arc-validate-results.xml: junit-formatted file containing the results of all tests performed on the arc by the arc-validate tool.
arc-quality.svg: Indicates the quality of the arc based on passed tests, e.g.
metadata.json: Invenio metadata record containing necessary information extracted from the arc by arc-to-invenio to create a bibliographic record on invenio. See also arc-to-invenio docs for a list of supported fields.

Difference between `arc.json` and `metadata.json`

metadata.json contains only the fields required for creaing a json file adhering to the invenio rdm metadata json schema

arc.json is the serialized full arc datamodel. It is a superset of metadata.json, which in contrast only contains invenio-relevant metadata. Note that properties are also named differently in these two files, e.g.

arc.json:

"people": [
    {
      "lastName": "Weil",
      "firstName": "Lukas",
      "email": "weil@nfdi4plants.org",
      "affiliation": "RPTU",
      "comments": [{ "name": "Investigation Person ORCID", "value": "orcid1" }]
    }
]

vs. metadata.json:

"creators": [
    {
      "person_or_org": {
        "type": "personal",
        "name": "Weil, Lukas",
        "given_name": "Lukas",
        "family_name": "Weil",
        "identifiers": [
          { "scheme": "email", "identifier": "weil@nfdi4plants.org" },
          { "scheme": "orcid", "identifier": "orcid1" }
        ]
      },
      "affiliations": [{ "name": "RPTU" }]
    }
]

So in short, every information contained in metadata.json is also in arc.json, but arc.json can contain more metadata that is not relevant for the invenio record json schema.

~~I'd suggest retrieving all metadata that are not directly related to the invenio metadfata record from arc.json.~~

metadata.json should contain all information needed to fill the forms.

all tasks implemented, further review required

nfdi4plants / archigator-backend