Community page for DSpace-CRIS users

  • What are you working on?
  • Where are you looking for partners for development or discussion?

Based on DSpace-CRIS 7

Who?What?NoteStatus

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

Improve the ORCID login flow solving the issue related to private email address in the ORCID profile and/or discrepancy among the ORCID's email and the one in the existing DSpace EPerson

This project has been funded by ORCID under their GPF program.

More details here Improvements to the Login Flow

IN PRODUCTION

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

Allow multiple Affiliations per person on publication level

In DSpace-CRIS it's possible to set an affiliation for authors or other persons on publication level. But in the current implementation it's only possible to set just one affiliation per author. As far as I know, this is the same as on DSpace-CRIS 7. Does anybody use setting affiliations for persons on publication level and how do you handle that?
-----

Hi Oliver, we are handeling it with further child-fields. Currently we are supporting one further affiliation - in future we want to add 3 more. If anybody has a better idea, we are interested.

Best Dirk. Fraunhofer IRB (Dirk Eisengräber-Pabst)

-----

Hi Oliver, in our settings one person has one affiliation (delivered via LDAP) and can set additional affilations by choice. E.g. https://fis.uni-bamberg.de/cris/rp/rp14224. So you enter the author's name in the submission and receive a preview of the entered affiliations, from which you can select one. That fits our needs.
Best Steffen, University of Bamberg (Steffen Illig )

IDEA

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

Importing items as private with bulk importOn DSpace-CRIS 7 2022.03.01 it's possible to set items as discoverable or not discoverable in the bulk import excel file. But it's not possible to create items, which are really private (no READ permission to Anonymous). For us, this is pretty bad as we are using the bulk import to import persons from our LDAP and need to restrict access to some of them. Does anybody else have the same problem?

PLANNED (WITH 4SCIENCE)

INTERESTED University of Bamberg (Steffen Illig )


Script Merging Entities

Our use case is: by now we regulary merge Persons (rps), which are created by mistake during the submission or by our ldap. In Dspace-Cris 5.X this could only done by our developers. In DS-CRIS 7 we would like to have a similar function operable via process to enable more colleagues to do this work.

We would like to merge two entities, so that only one entity remains. The remaining one should contain all relationships and metadatafields from the deleted entity without losing its own metadatafields.

It has been confirmed to us that this feature is part of the Data Quality Module,  which 4Sience is currently working on.  We will therefore participate in the financing of this module (to become Open Source). Help is welcome.


PLANNED (WITH 4SCIENCE)

INITIATED Hamburg University of Technology (Oliver Goldschmidt )

INTERESTED

See all languages for specific metadata fields (instead of only the values in the language, which is currently chosen on the UI)

If metadata fields with different languages are set (for instance a title in german and a title in english), the german title will only be displayed, when German is chosen as the language of the User interface. For some use cases this is a very nice feature. But for some metadata fields this could be unwanted (for instance if you have subject headings in different languages, you may want to see all of them at once, independently from the actually chosen language).

IDEA

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

Set user for command line operationsOn command line it is not possible to set a user, who should execute the operation on DSpace end. That is a problem, because it's impossible to create new entities via bulk import on command line as the user needs to be allowed to create new entities to the collection the import is targeting.

IN PRODUCTION

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

https://github.com/4Science/DSpace/issues/419 

Bitstreams listed in the advancedattachment CRIS Layout option are not sorted correctly

Bug on the itempages (at least with the advancedattachment rendering in DSpace-CRIS): the files are sorted randomly and the sort order from bundle2bitstream.bitstream_order is ignored. Anyone having trouble with that as well?

IDEA

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED

https://github.com/4Science/dspace-angular/issues/66

Version information not visible on DSpace-CRIS itempages

The version information with a reference to the latest version and a version history for versioned items is not displayed on DSpace-CRIS itempages (in fact, it is visible on the full item view page and on generic DSpace 7, but not on DSpace-CRIS)

IDEA

INITIATEDUniversity of Bamberg (Florian Gantner )


Organization Unit Tree

https://github.com/4Science/DSpace/issues/386 

https://github.com/4Science/DSpace/pull/383 

https://github.com/4Science/Rest7Contract/pull/4 

https://github.com/4Science/dspace-angular/pull/52 

Implementation of some organization unit tree which shows the hierarchical structure of Orgunits together with some metrics for each tree node.

The tree is included as some extra section component. For each tree node metrics are calculated using solr queries or they are aggregated from subordinate nodes. The tree node is delivered through some new Rest endpoint.

Some working example can be shown on https://fis.uni-bamberg.de/explore/orgunits

WORKING ONIN PRODUCTION

INITIATED Hamburg University of Technology (Oliver Goldschmidt)

INTERESTED University of Bamberg (Florian Gantner )

Orcid Push of Product Entity

https://github.com/4Science/DSpace/issues/426

https://github.com/4Science/DSpace/pull/428 

https://github.com/4Science/Rest7Contract/pull/5

https://github.com/4Science/dspace-angular/pull/73


Push of Product Entity to Orcid as Orcid Work. The Push uses some own mapping Factory and thus can be configured in the same way as the Orcid Push oif Publications.

WORKING ONIN PRODUCTION

INITIATED University of Bamberg (Florian Gantner )

INTERESTED

"Associate" Item

https://github.com/4Science/DSpace/issues/365

https://github.com/4Science/DSpace/pull/382

https://github.com/4Science/dspace-angular/pull/50

https://github.com/4Science/Rest7Contract/pull/3

Extension of the edit-item mode which allows some user to "associate" some project to some publication (e.g. Publication dc.relation.project holding the reference to the project), although the permissions for editing the publication are missing.

We use the terminus associate because connect, related, interlinking etc... and other synonyms are all already used in DSpace speech.

WORKING ONIN PRODUCTION

INITIATED University of Bamberg (Florian Gantner )

INTERESTED

MINE / MY_SELECTED orcid sync settings

https://github.com/4Science/DSpace/issues/348

https://github.com/4Science/DSpace/pull/351 

https://github.com/4Science/Rest7Contract/pull/6

https://github.com/4Science/dspace-angular/pull/38


Implementation of Orcid Sync Settings for relationship based preferences (existed in DSpace 5). This allows profile owners to push only selected publications/fundings (MY_SELECTED) or all non-hidden publications/fundings (MINE) to orcid.

WORKING ONIN PRODUCTION

INITIATED University of Bamberg (Florian Gantner )

INTERESTED

Versioning by metadata

https://github.com/uniba-ub/DSpace/tree/uniba-versioning
https://github.com/uniba-ub/dspace-angular/tree/uniba-versioning
https://github.com/uniba-ub/Rest7Contract/tree/uniba-versioning


Part of the versioning logic used by university of Bamberg, as previously presented at https://doi.org/10.20378/irb-53684 (german only) for DSpace5 by metadata. Add some context-menu button and endpoints to create some version/copy of some item as own workspaceitem. The reference between versions (isversionof/hasversions 1 to n) is hold in reciprocal metadata. Another functionality is to change the main version (which hasversions and in our case is discoverable) within some version group by another button in the context-menu

WORKING ONIN PRODUCTION

INITIATED University of Bamberg (Florian Gantner )

INTERESTED

Enhancement Tools

https://github.com/4Science/DSpace/issues/357

https://github.com/4Science/DSpace/pull/312 


The item-enhancer is usually some long-running process which runs over all items. We made some forks of the item-enhancer script: 1. only consider certain entity type or collection 2. only consider items of some solr-query, e.g. only recently changed.

Both approaches limit the amount of items being processed and allow some fine-granular (re)calculation of virtual metadata. In addition there has been some improvements to the Commitment of the Context.

The tools also contains some NameEnhancer which can be used to calculate the dc.title from several other metadatafield in their order of appearance.

WORKING ONIN PRODUCTION

INITIATED

INTERESTED



IDEAPLANNED (WITH 4SCIENCE)WORKING ONIN PRODUCTION

INITIATED

INTERESTED



IDEAPLANNED (WITH 4SCIENCE)WORKING ONIN PRODUCTION

IN PRODUCTION

Based on DSpace-CRIS 5

Interested in discussion (Searching for Co-Funders)

  • Automatic Software Deposit 

Functionality similar to Zenodo: https://guides.github.com/activities/citable-code/ It should be include local GitLab installations as well (not necessarily a CRIS feature).
Interested: Hamburg University of Technology (Beate Rajski) Fraunhofer IRB (Dirk Eisengräber-Pabst) TU Darmstadt (Gerald Jagusch)

  • Automatic detection of Identifier type

We are dealing with several Identifier types within DSpace (CRIS), mostly kinds of Persistent Identifiers like DOI, Handle, URN, as well as ISBN, ISSN, URL or similar things. In many contexts it is necessary to know about the identifier type (for instance for lookup or for submission to DataCite or other services). Handling of identifiers is different: sometimes it's qualified as a qualifier in the metadata field (for example dc.identifier.doi or dc.publisher.doi), but this is not good extensible and requires some effort. So it would be nice to have a concept how to deal with these identifiers. A useful extension for that would be the automatic detection of the identifier type. Zenodo has such a mechanism by guessing and suggesting the identifier type. We would like to have such a mechanism for DSpace as well, in any cases, where different identifiers might be used. Zenodo is using a Python toolkit for that (https://idutils.readthedocs.io), that might be useful to implement such a function.
This is not necessarily a CRIS feature only.
Interested: Hamburg University of Technology (Oliver Goldschmidt )


  • lobid integration

An integration of lobid (http://lobid.org/) would integrate the German Common Authority Record Database (GND) and other lobid services (like lobid-organisations) into DSpace. This could be useful in various contexts:

  • Lookup for subject headings (GND)

    • Lookup of person names (GND)

    • Lookup of organisation units (GND and lobid-organisations)

We would like to have that integration for the various lookup use cases as one more simplification to get content into the entities.
Interested: Hamburg University of Technology (Oliver Goldschmidt ), State and University Library Hamburg (Phu Tu )
The Library Code (Kim Shepherd and Pascal-Nicolas Becker ) are starting to implement this and will provide it back to DSpace-CRIS once it is done.

  • Better import of BibTex and RIS files and DOI imports

    The import of BibTex or RIS files and the DOI-Import in DSpace (-CRIS) using the gr.ekt code should be upgraded: the current import does not take into account differences between document types.
    For example, RIS input fields can map to different Dublin Core fields depending on document type.
    Interested in: University of Bamberg (Steffen Illig

Planned/Realized with 4Science

  • Thumbnails for CRIS-Files  (ongoing)

    E.g. the picture in the profile is too huge for displaying it directly. A thumbnail would be smarter. 
    University of Bamberg 
  • Printerfriendly version of a CRIS-Entity (done)
    University of Bamberg

  • Set priorities of nested entities (ongoing)
    University of Bamberg

  • Connection of CRIS Entities (Project, Awards) via autocomplete with Publications  (done)

    The feature is documented here.
    University of Bamberg & Hamburg University of Technology
  • Improvements for switching en/de (done)
    University of Bamberg & Hamburg University of Technology

  • Integration of the Unpaywall API in the Submission Workflow (ongoing)
    Helmut-Schmidt-University Hamburg
  • Highlight changes in the Submission Workflow for new version (ongoing)
    Helmut-Schmidt-University Hamburg
  • Vertical and horizontal tabs in CRIS Entities (done)
    It will be an alternative layout for the CRIS object page highly configurable. Instead to use tabs a left menu will be introduced, with the possibility to group different sections (i.e. tabs) under a common label to build a hierarchical menu. Tabs not listed for the vertical menu will appear on the top of the CRIS content as horizontal buttons. The feature is NOT planned to be merged in the official code base but will be available on the HSU github public repository for other interested party. The source code is available here https://github.com/ub-hsu/DSpace/tree/openhsu-work
    most come in with this commit https://github.com/ub-hsu/DSpace/commit/b4e4127361c953fa17902a60256b91e9fcc9a101
    Helmut-Schmidt-University Hamburg
  • Hierarchical tree structure (based on the DSpace-GLAM solution) for CRIS-entities in general (done)
    https://openhsu.ub.hsu-hh.de/cris/explore/orgunits

    Helmut-Schmidt-University Hamburg
  • "Add fulltext / Create new version" and "green open access publication"-feature (various entry points) (ongoing)
    "Add fulltext" enables e.g. authorized logged in users to jump directly in the submission workflow again to add the fulltext for dspaceitems, which has been published first as "only metadata"-submission. Enhancements are in progress.
    Helmut-Schmidt-University Hamburg
  • Visual clustering and distinction of researcher profiles by status (e.g. in the "people"-section of orgunit-profiles or in facets) (done)
    Helmut-Schmidt-University Hamburg
  • Visual distinction of organization units by type (done)
    Helmut-Schmidt-University Hamburg
  • Duplicate Check for imports (done)
    DSpace's duplicate check is triggered on the first page of the submission form. For manually added records this is early enough, but for records taken from a database import operation, this is a bit too late. So it would be a nice feature to see a duplicate warning already on the search result page after looking for new records in databases.
    Hamburg University of Technology (Oliver Goldschmidt ), University of Bamberg ( Steffen Illig )
  • Update ORCID integration to use ORCID API v3 (done)
    DSpace CRIS should be able to use ORCID API v3
    Hamburg University of Technology (Oliver Goldschmidt )
  • Improve ORCID Lookup (ongoing)
    The person lookup in DSpace CRIS can contain a lookup operation in ORCID. This operation is pretty slow and can delay the lookup process significantly. The ORCID lookup can be parallelized to improve performance. So it could be a good idea to split the lookup into two steps: first only the local database is searched. A button should be added to the lookup window to allow the user to lookup in ORCID as the second step. As ORCID Lookup is only allowed if the ORCID is definitely known, this would allow ORCID Lookup on demand and would improve lookup performance at once.
    Hamburg University of Technology (Oliver Goldschmidt )

Working on

Who?What?Note
University Kassel/disy (Stefan Lossow)Impact Pathway
University Kassel/disy (Stefan Lossow)API to DPMA or EPO patent database
University of Bamberg (Florian Gantner)Set the visibility of nested entitiesNot realized yet. General function for several purposes.
University of Bamberg (Florian Gantner)Showing lists of "See also" ressources based on (external) "See also" providers using the GND identifier. First draft online in : https://professorenkatalog.uni-bamberg.de/cris/person/person00832

In Production

Who?What?Note

University of Bamberg (Cornelius Matějka)

Multilingualism for discovery facets and browse indexesPR: https://github.com/4Science/DSpace/pull/89

University of Bamberg (Cornelius Matějka )

Management of private CRIS Entities for researcher

If a researcher sets a CRIS Entity as private (e.g. maybe an future project => See also RDMO integration) we wonder how he would manage these items in future (maybe to update them). A possible solution would be a special area for such cases in the MyDSpace-CRIS Site with the correct rights management (only one's private projects).

PR: https://github.com/4Science/DSpace/pull/79
University of Bamberg (@Gantner, Florian@Philipp Rumpf)

Linked Open Data für CRIS-Entities

Based on the DSpace modul Linked (Open) Data by Pascal-Nicolas Becker we created a own modul which works also for CRIS Entities.

PR: https://github.com/4Science/DSpace/pull/88

University of Bamberg (Cornelius Matějka )

Site policy modal

Can be easily applied to DSpace (without CRIS (wink)).

PR: https://github.com/4Science/DSpace/pull/80,

University of Bamberg (Cornelius Matějka

Enable ENTER key in textareas and add suitable rendering of running text (e.g. abstracts, etc.).

PR: https://github.com/4Science/DSpace/pull/51 (merged)

University of BambergPrefilled contract formulars for download or mail delivery (submission process)

The extension could not be integrated in the official master (4Science) because of specific requirements and the use of Jasper.
The jasper license is an AGPL so it is not compatible with the official DSpace-CRIS license. 

Hamburg University of TechnologyExtend journal lookup (in ZDB and in local entity objects) to lookup by ISSN instead of journal title

The ZDB integration has been pushed to github repository

University of BambergHide empty tabs on CRIS-Entity-Homepage
  • Hamburg University of Technology also has opened a ticket for that... 4Science told me, that it should be standard behaviour, that a tab without content is hidden, but for us this also does not work in any case.
  • University of Bamberg: We have asked Andrea at the AWT 2019. It should be standard but it is not at the moment. To support the development of the feature we opend a ticket right now.
  • Hamburg University of Technology: on our test system this feature is working as expected, so the problem should be solved in the latest master... Andrea Luigi explained to me, that the tab is not displayed if there is no content at all, if there are no metadata fields for the box (in jsp controlled boxes there is no need to display metadata fields either way). So, after removing the metadata from the researcher tab in an ou profile for instance, the tab will not be displayed (but, as I noted, only on our test system; on our production system the problem really is till there).
  • University of Bamberg: We have merged the latest master yesterday and can confirm that empty tabs on CRIS-Entity-Homepage are now hidden. So we moved that topic from "Planned with 4Science" to "In Production".

University of Bamberg (Jonathan Boss & Cornelius Matějka )

Browsable organisation unit tree
Browsable DDC tree

PR: https://github.com/4Science/DSpace/pull/119
4Science is also working on a more generic solution, highly scalable to visualize generic tree of objects (see ongoing project from HSU)
University of Bamberg (@Philipp Rumpf)ISBN Search (B3Kat and LoC)Merged in the official codebase commit 8ad1605388d96f75a29a058081142f140cd24bcd
University of Bamberg (@Jonathan Boss @Gantner, Florian)OData Connector (REST API)  between DSpace-CRIS and Typo3-HomepagePR: https://github.com/uniba-ub/dspace-odata
University of Bamberg
(@Jannik Lappe)
Contact form for publicationsPlease ask if you are interessted
Hamburg University of TechnologyDuplicate Check for imports

DSpace's duplicate check is triggered on the first page of the submission form. For manually added records this is early enough, but for records taken from a database import operation, this is a bit too late. So it would be a nice feature to see a duplicate warning already on the search result page after looking for new records in databases.

Feature was funded by Hamburg University of Technology and University of Bamberg