Guide for TUdatalib administrators

Status: 2023-06-29

licensed under CC0 1.0, no copyright protection

1 About TUdatalib

TUdatalib is the institutional repository of the TU Darmstadt for research data that were created or worked with at the TU.

TUdatalib enables structured storage of research data and descriptive metadata, long-term archiving (at least 10 years) and - if desired - publication of metadata and/or files including DOI assignment. In addition, there is a fine-grained rights and role management.

TUdatalib is operated jointly by ULB and HRZ and is based on the open source software DSpace, the web interface of TUdatalib can be reached at tudatalib.ulb.tu-darmstadt.de.

The service has been available to all members of TU Darmstadt since 2019 and is in regular operation since 2022.

2 Registration

When logging in, the members of the TU must authenticate themselves with their TU-ID. As a rule, users are automatically assigned to their research group (via the T-group stored in the IDM) and can then use TUdatalib according to the rights granted to them by the research group administrator.

External users can register with their email address and must be activated by a research group administrator.

Non-registered users can only access freely available data sets.

3 Charges

The use of TUdatalib is free of charge for up to 2 TB total volume of new data per year and research group of the TU.

For larger volumes of new data, a one-time cost sharing by the research group is required for the data volume exceeding 2 TB. This is currently 250 € / TB for 10-year archiving. Invoicing is handled by the HRZ. TUdatalib sends notification e-mails to the admins of a department if the free data volume is filled to 90% and 100%.

4 System times / maintenance windows

Every Tuesday from 06:30 to 08:30 TUdatalib may be unavailable due to maintenance work.

5 Organization and role model in TUdatalib

5.1 The individual entities

In TUdatalib, the organizational units of the TU are preset as communities, sorted by departments, institutes, and research groups. If a research group or similar is missing, please contact us. Communities can be divided into further sub-communities. For each community in TUdatalib there should be at least one administrator (two are recommended). Administrators manage the roles and rights for their community and can create collections.

Any number of collections can be created within a community in order to map different authorization groups (groups). Any number of data sets (items) can be created in a collection. Each data set contains a descriptive set of metadata, any number of individual files and a persistent identifier (see section "Assigning Persistent Identifiers" below).

_{(Source: https://wiki.lyrasis.org/display/DSDOC6x/Functional+Overview)}

5.2 Rights and role management

In TUdatalib, rights and roles are assigned exclusively to groups (and not to individual users). Users can be assigned four different roles using these authorization groups:

Administrators: (see also in particular the section "Administration of the data sets"). The main rights include:
1. Creating collections
2. Assignment and administration of roles and rights
3. Editing metadata
4. Registration of DOIs
5. Mirroring data sets between bollections
Readers: Users who are allowed to view the data sets in a collection / community. This is the default right for users who have been automatically assigned to a community upon login.
Submitters: Users who are allowed to submit new data sets to a collection.
Controllers: Users who are allowed to accept or reject submitted data sets and edit their metadata.

6 Workflows in the web interface of TUdatalib

6.1 Metadata assignment for the community

If the administrator has been assigned to a community (usually the research group), he can first edit the metadata of his community.

To edit the metadata of your community, navigate from the start page to your community. To do this, click on the corresponding community for which you have administrator rights. Click on "Edit Community" on the left side of the navigation area under the section CONTEXT. In the "Edit Metadata" tab, you can enter and maintain a community description with name and an introductory text (optionally in HTML), a logo, etc.

6.2 Allocation of administrator rights

To grant administrator rights to additional users, the authorization group must be extended.

Click on the tab "Assign roles". A group in the form COMMUNITY_xxx_ADMIN is then displayed. You can edit the group by clicking on the group name. Users can be found via the search slot by entering their last and/or first name and can be added as members if they have already been logged in to TUdatalib.

You can then manage the members of this group. Members of the group can be individual users or groups of users.

If a user is not found in the user list, please make sure that the user has logged in to TUdatalib before.
If you always want to assign the same group as administrators for sub-communities, please contact us. The system administrator can predefine such a group in TUdatalib.

6.3 Creating substructures (sub-community & collections) in a community

6.3.1 Creating a sub-community within a community

It is possible (but not obligatory!) to create sub-communites in one community. Navigate to the community in which you want to create a sub-community. Then click on "Create Subcommunity" on the left in the navigation area under the CONTEXT section (see picture).

A sub-community can be edited in the same way as a community and has the same metadata fields.

6.3.2 Creating a collection within a community

The creation of at least one collection is mandatory to be able to create data sets (see 5.1). After the user group has been defined with administrator rights, you can go on to create a collection in your community. Navigate from the start page to the community in which you want to create a collection. Then click on "Create Collection" on the left side of the navigation area under the CONTEXT section (see picture).

6.4 Assigning metadata for the collection

When creating a new collection, various metadata can be specified. This includes the name of the collection, a short description, an introductory text (optionally in HTML), a logo and the indication of a license and provenance of the collection. You can maintain the metadata in the tab "Edit Metadata" afterwards.

Under the tab "Content Source" you can specify whether the collection is a standard TUdatalib collection or whether the collection obtains its content via harvesting from an external source.

6.5 Assignment of rights and roles within a collection

In the tab "Assign roles" you can set, for example, who is allowed to view the files and metadata of the data sets. Four different roles can be assigned to users and groups, see section 5.2 "Rights and Role Management". The group "Readers" is automatically generated when a new collection is created and by default includes a group to which all members of the corresponding organizational unit are automatically added based on their TU ID (after the first login to TUdatalib).

All community administrators are always also collection administrators for all collections in your community.

Collection administrators do not automatically have the controller role, since this role does not necessarily have to be filled.

To further edit rights and roles within a collection, navigate from the home page to the collection you want to edit. Then click "Edit Collection" in the CONTEXT navigation pane.

6.6 Assigning roles for agents via groups

When assigning the above roles, a group must be created for each role assigned. This group automatically receives a name in the form COMMUNITY_xxx_ or COLLECTION_xxx_ (see table below), where the middle part is the internal ID of the community specified by the system. The name of these groups cannot be changed. For some processing steps it is therefore useful to know a group's internal ID.

Role	Name of the respective group	Example	Rights of group members
Administrator	COLLECTION__ADMIN	COLLECTION_ae910c2b-d734-4777-9571-0f116d7ff41a_ADMIN	• Creating of collections • Granting and administration of roles and rights • Acception or rejection of data sets • Editing metadata • Registration of DOIs • Mirroring data sets from other collections (if those carry the same rights as the required data sets)
Reader	COLLECTION__DEFAULT_READ	COLLECTION_9c4f4d8e-6c6b-43ea-b4c1-88b46cbc6af3_DEFAULT_READ	• Read access without ability to change content in TUdatalib • Download of data sets
Submitter	COLLECTION__SUBMIT	COLLECTION_9c4f4d8e-6c6b-43ea-b4c1-88b46cbc6af3_SUBMIT	• Submission of data sets to TUdatalib
Controller	COLLECTION__WORKFLOW_ROLE_editor	COLLECTION_9c4f4d8e-6c6b-43ea-b4c1-88b46cbc6af3_WORKFLOW_ROLE_editor	• Acception or rejection of submitted data sets • Editing metadata of those

If one of the four possible right groups does not yet exist (only the group for readers is created automatically), it can be created in the "Assign roles" tab by clicking on "Create".

Attention

Authorizations are not checked recursively in the DEFAULT_READ groups, therefore only users should be members of the DEFAULT_READ group, no other groups. In particular, the following applies: If a collection is to be publicly visible, the relevant DEFAULT_READ group should be completely deleted.

The administrator of a community or a collection can edit the members of the groups, e.g. add users or other groups.

To add a user to a group, the user must be found via the search and then added with the "Add" button. Don't forget to save your changes.

Attention

To maintain scientific traceability, only users who can be unambiguously identified as a person may be given roles with write permissions. This means that users who were created as external users via a functional mailbox may not be included in the administrator or controller group.

6.7 Creation and description of data sets

To add data sets to your collection, you can do the following:

Click the name of the collection in your pane. Then click "Submit a new item to this collection".
Or navigate via the start page to the section "MY ACCOUNT" and select "Submissions". Click on "Start another submission" and then select the collection to which you want to add data.

6.7.1 Naming of data sets

It is advisable to give data sets names that are meaningful even for outsiders and to avoid special characters.

6.7.2 Assignment of metadata

When collecting your research data as a data set, various aspects are queried. The mandatory fields are marked with an * (see table). The main purpose of descriptive metadata is to make records more discoverable and to provide an understanding of their content.

All information about a data set - except for the authorizations - should be represented via the metadata to ensure optimal visibility and reusability of the data. This also applies, for example, to a data set's relations to (third-party) funded projects and other resources like, e.g., publications.

Metadata field	Description	Input type	Field repeatable
Titel (Title)*	Name for the entire data set	Free text	✔
Beteiligte Personen (Involved persons)*	Persons responsible for a data set. We recommend the use of ORCID to identify persons, see: http://www.ulb.tu-darmstadt.de/orcid. By clicking on "Lookup" the name can be searched in the ORCID registry and then inserted into TUdatalib.	Free text Lookup in ORCID Registry	✔
Datum (Creation date)*	Date at which the data set was created. Only the year has to be specified, day and month can be left open.	Year and day: Free text Month: drop-down menu	✘
Weitere verknüpfte Ressourcen (Other linked resources)	Relationship to and unique identifier of any linked resources (e.g. a journal article based on the stored data)	Relation type: Drop-down Menu Identifer type: Drop-down menu Identifier: Free text	✔
Typen (Types)*	Specification of the type of data set	Checkboxes	✔
Spache(Language)	Primary language of the data set, if applicable	Drop-down menu	✘
Schlagworte (Keywords)	Free keywords to describe the content of the data set. Due to international reusability, it is recommended to use only English keywords.	Free text	✔
DFG-Fächer (DFG subjects)*	Assignment of the data set to the DFG subject classification	Drop-down menu	✔
Beschreibung (Description)*	Free short description of the data set ("ReadMe")	Free text	✘
Beschreibungsversion (Description version)	Version name of the data set	Free text	✘
Drittmittelprojekt (Third-party funded project)	Attribution to third-party funded projects at the TU Darmstadt. You can select from a maintained list of projects using the lookup button.	Lookup	✔

The metadata field "Drittmittelprojekt" enables a standardized selection of third-party funded projects to be attributed to the data set via a selection list (lookup). This is called up by clicking on "Lookup".

In the lookup list, projects can be searched for using the search line. It contains about 6000 third-party funded projects of the last years. Due to the still improvable quality of the entries in the list, it is best to search for the grant number, only alternatively for the funder or acronym/title of a project.

Attention

You can search here only with one term, not with several!

These metadata fields are minimal. TUdatalib therefore allows the addition of further metadata fields for your discipline-specific context (i.e. your community or collection) - please contact us! Only this way an efficient search for data based on subject-specific parameters becomes possible.

6.7.3 Research data documentation

In addition to the descriptive metadata, we recommend uploading a research documentation in the form of a *.txt file as a file included with the dataset. Such documentation should help third parties to interpret your data as clearly as possible.

6.7.4 Assignment of additional read permissions

A data set first inherits the authorization groups from the collection in which it resides. Before you upload files, you can specify if any of the additional authorization groups you have previously created should also receive read permission for the entire data set, i.e. files and metadata. To do this, select and add the desired groups one by one. The group "Anonymous" includes all users, even those who are not logged on. You should therefore select "Anonymous" if the data set is to be freely available on the web (Open Access). You can give names to the individual access authorizations in order to distinguish them. Finally, click on "Confirm Policy & add another" to make your selection effective.

6.7.5 Uploading files

After you have entered all required information, you can upload your files to the data set via the upload mask.

6.7.6 File formats

TUdatalib is not limited in the choice of file formats. However, there is generally no guarantee that certain formats will still be readable after many years. It is therefore advisable to convert your files into one of the recommended formats before uploading, if possible. Recommended formats for long-term archiving and information on possible conversions can be found here: https://www.forschungsdaten.info/themen/bewahren-und-nachnutzen/formate-erhalten/. The TUdata team will also be happy to advise you on this.

In general you should not upload your files in a container format (zip. or similar) as this makes long-term archiving and indexing very difficult. Container formats are only recommended if you need to maintain a folder structure for your data.

6.7.7 File size limit

We do not recommend uploading single files larger than 4 GB via the web interface. Large files or many single files are better uploaded via the REST interface of TUdatalib. Sample code in the form of a Python script is available for their use. Please contact us if you want to use this. Additionally, we offer to add large data files from internet locations via a form after dataset submission.

6.7.8 Licensing

A license regulates the rights of third parties to use your research data if you publish it. Please refer to the recommendations and further references given in the input mask. Then decide which rights of use you wish to grant and select the appropriate license from the list provided. If your desired license is not found in the list, you can enter it in the free text field at the end. Please also enter a URL where the license can be viewed.

6.7.9 Final review and archiving

In the last tab "Review" you can check all entered files and metadata again and correct them section by section. Please check your entries carefully. Do not forget to accept the user agreement. Select "Complete submission" to add your data set to the collection. If no group for the controller role is available in the collection, the data sets will now be immediately visible to all readers in the collection without further verification. TUdatalib allows curation through a controller before final archiving of data, if the appropriate role is assigned. The procedure is then like this:

A submitter adds a data set to a collection.
The archived data set initially remains invisible even to those with read access and cannot be changed until it is released by a controller.
A controller is informed about the new data set by e-mail and can decide to accept (or reject) the data set and edit the metadata if necessary.
All submitted data sets will only become visible to the read-authorized persons after the controller has approved them.

To enable this workflow for a collection, the administrator must assign a controller group to the collection in the "Edit Collection" section of the "Assign Roles" tab. All members of this group can then perform the curation step before final archiving.

After the submission has been approved, the data set will be archived in TUdatalib and a unique and persistent identifier will be assigned which can be found in the URL (example: https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/1969). The identifier can be used to uniquely and permanently reference the data set with its URL.

6.7.10 Adding large data files

Currently, the DSpace user interface only allows for uploading files that are no larger than 4 GB. Larger files are possible by connecting to the REST-API, but even here, there are technical limitations to uploading large files. For this reason, we offer to attach larger files to TUdatalib datasets from internet locations. You can request this service for existing datasets via a form linked from the dataset landing page.

A typical workflow would look like this:

Submit the datset to TUdatalib using the web interface with a small (dummy) file. Submission of the dataset has to be completed to attach large files as described here, an unfinished submission is not sufficient.
Upload the large file(s) to a location on the internet that can be accessed by us. We recommend using the GigaMove transfer service for files up to 100 GB (more information).
Access the large file upload form from the landing page of the submitted dataset and send us the locations from the previous step. The form is pre-filled with all necessary values except for the file locations (=URLs)
We add the files to the dataset and get in touch with you once done or if there are issues
Remove the dummy file if this should not be included in the final dataset

If you use packaged files such as .zip files, please only use digits, letters, and underscores in the package names.

If using the form with URLs is not a suitable approach in your case, please send us an email and we will discuss alternative solutions.

6.8 Administration of the data sets

The rights of an administrator include additional powers once the submission has been approved. He or she can register DOIs, edit access rights to data sets (metadata and files) after archiving, identify older versions of data sets, and mirror data sets in other collections.

Clicking on "Edit this item" in the navigation pane displays further administration options, including discarding, moving to another collection or making the data set invisible.

6.8.1 Publishing datasets

If all records in a collection are to be publicly visible, it is recommended to follow the procedure described in Section 6.6, where the DEFAULT_READ group of the collection is deleted. If datasets are in a collection that is not publicly visible, they can still be published. For this purpose, the complete dataset or only the metadata of the dataset can be made publicly visible via the two buttons "Remove all restrictions (metadata+bitstream)" or "Remove restriction of metadata". In the latter case, the dataset can be found on the web, but the actual files cannot be accessed. Interested parties are thus given the opportunity to find datasets and, if necessary, to ask the respective administrator by e-mail whether access to the files can be granted (see Section 6.8.4).

Translated with www.DeepL.com/Translator (free version)

6.8.2 Allocation of DOIs

All data sets in TUdatalib receive a unique addressing with a so-called handle (see "Final testing and archiving"). In addition, TUdatalib allows you to assign DOIs (Digital Object Identifiers), which make your data sets permanently publicly findable and citable (example: https://doi.org/10.25534/tudatalib-36.9). They can be compared to a fingerprint, which identifies each data set unambiguously.

6.8.2.1 Register DOI

If the Anonymous group has read permission for at least the metadata of a data set, the collection administrator can register a DOI for that data set on the "Edit this item" tab.

If access restrictions exist for the metadata, DOI assignment is not possible, but you can remove the access restrictions by clicking "Remove all restrictions" or "Remove restrictions of metadata", which will allow DOI assignment.

Clicking on "Register DOI" in the "Item Status" tab will trigger this.

After TUdatalib has received a positive response to the request, the assigned DOI is displayed immediately.

If the data set with a DOI is public, this data set cannot be set to private or deleted in TUdatalib. If a new version of the data set is created by clicking on "Create version of this item", the DOI will not apply to the new version. A separate DOI must be assigned for each new version of a data set in the same way as above. If you do not want to register the DOI until later, but already need the DOI's URL in advance, you can use the URL generated and reserved (but not registered) by TUdatalib (see picture).

6.8.2.2 Update data set with DOI

If the metadata of a DOI data set is changed in TUdatalib, the changed metadata must also be updated at the DOI agency. This is done automatically every night. Alternatively, you can click on "Update DOI" in the "Item Status" tab to manually update the metadata immediately.

6.8.3 Subsequent modification and deletion of access rights

Rights for read-access of a data set are automatically inherited from a collection on its data sets (cf. sction 6.5). The access rights of a data set (metadata and files) can nevertheless be changed by the collection administrator, even at a later date. These rights include: READ, WRITE, ADD, REMOVE, ADMIN, WITHDRAWN_READ.

	Community	Collection	Item (full)	Item (bundle)	Bitstream
READ	x	x	x	x	x	read-access to all content
WRITE	x	x	x	x	x	write-access to all content
ADD	x	x	x	x		addition of objects
REMOVE	x	x	x	x		removal of objects
ADMIN	x	x	x	x	x	full access
WITHDRAWN_READ			x	x		read-access for withdrawn objects

Existing permissions for a data set can be deleted individually for each data set. Furthermore, permissions for a data set's metadata can be set/deleted separately from the permissions for its files.

Removing restrictions allows read-access to the data set, or the files or metadata, for all users, even those not logged in (Anonymous group). This is confirmed by a message, as shown in the picture:

We recommend using these features sparingly, except for the applications given in the following subsections.

6.8.3.1 Removing all restrictions on a data set

This deletes all restrictions on reading a given data set (files and metadata) and sets the permissions to "Anonymous", the data set is then freely available on the web (Open Access).

Click on "Edit this item" in the navigation community. Selecting "Authorizations..." opens a dialog for listing by files with rights.
Click on "Remove all restrictions"

The result corresponds to the procedure described in section 6.8.1.

6.8.3.2 Publishing metadata of restricted data sets

This removes all restrictions for reading the metadata of a data set and the permissions are set to "Anonymous". The data set can then be found in the network, but the actual files cannot be accessed. Interested parties are thus given the opportunity to find data sets and, if necessary, to ask the respective administrator by e-mail whether access to the files can be granted (cf. section 6.8.4).

Click on "Edit this item" in the navigation community. Selecting "Authorizations..." opens a dialog for listing by files with rights.
Click on "Remove restriction of metadata"

The result corresponds to the procedure described in section 6.8.1.

6.8.3.3 Different restrictions for different files of a data set (partial publication)

It is also possible to give different files of a data set different restrictions, allowing publication of only selected files of a data set.

Click on "Edit this item" in the navigation community. Selecting "Authorizations..." opens a dialog for listing by files with rights.
Click on "Remove restriction" for each file you want to publish.

6.8.4 Processing access requests to restricted data sets

Data sets whose metadata can be freely viewed can be found (but not viewed) by unauthorized users. In TUdatalib, these users have the possibility to request access to the files. This access request is sent to the collection administrator responsible by email. We ask you to answer the user. If desired you can

then add her or him to the authorization group after he has logged on to TUdatalib once.
send him a temporary link (valid for 48 hours) where she or he can download a single file or all files of a data set. This access is logged by the system so that it can be traced later.

6.8.5 Versioning

A versioning of data sets is possible, i.e. a new, slightly modified data set (e.g. small changes in the actual files) replaces a previous data set. The metadata does not have to be re-entered for the new data set. For this step you can select the button "Create version of this item" in the navigation bar in the EDIT section.

After versioning, the old data set can no longer be found by searching in TUdatalib, but remains directly accessible via URL or DOI (Digital Object Identifier). A note about the newer version is displayed.

Versioning instead of modification is the method of choice for all data sets with DOI to ensure the traceability of the data genesis.

6.8.6 Subsequent modification and addition of metadata

As collection admin, you can subsequently add or modify the metadata of a data set. To do so, click the "Edit this item" button and then the "Item Metadata" tab. Please note: Manually changing and adding metadata requires knowledge of the metadata level used (Dublin Core), the entries are not validated. You are responsible for entering the correct data in the correct format. Therefore, we recommend that you create a new version of the data set (see above) when making additions or major changes to the metadata, in order to be guided through the summission process again. You can then delete the old version (unless you have registered a DOI for the data set).

6.8.7 Adding a thumbnail

Thumbnails are usually generated automatically, e.g. when you upload a pdf. However, this is not possible for all file types.

In this case, you can add a thumbnail manually afterwards. To do this, go to "Edit this item", select the tab "Item bitstreams", and click on "Upload a new bitstream" at the bottom. Here, select the bundle "Thumbnails", then you can simply upload a jpg file from your computer and optionally add a description. This file will then be displayed as a thumbnail.

To add the thumbnail, you need the rights to modify the item. If you do not have these, please contact the TUdatalib admin of your working group.

6.8.8 Mirror data sets in multiple collections

With the tool "Item mapper", a data set can be displayed in several collections at the same time, as it can be useful for cooperation projects.

To do this, click on "Communitys & Collections" in the right-hand menu and select the target collection or navigate from the home page to the target collection into which you want to mirror the data sets.

Click on "Item mapper" on the left in the CONTEXT navigation area. You can search for data sets by typing in the search line shown and then select the data sets to be mirrored. Then click on the button "Map selected items". You can list all mirrored data sets and delete them from the list if required.

If a data set is mirrored, it does not receive new access rights. All permissions assigned to the original collection by the administrator are retained. If you do not have read access to the data set, you cannot mirror the data set to the target collection.

6.8.9 Moving data sets between collections

Collection administrators can move a data set to another collection if they have admin privileges in both collections involved. To do this, click the "Move..." button. button in the "Item Status" tab.

6.8.10 Deleting data sets

Only the system administrator can finally delete an entire data set, please contact us for this purpose. Community and collection administrators can withdraw data sets, however, in the "Item Status" tab. These are then no longer searchable in the interface.

6.8.11 Links for temporary access tu datasets

For example, for peer review, you may want to give external people temporary access to an unpublished dataset in a collection you administer. For this purpose, the system offers the possibility to generate access links that are valid for 30 days. These links can be created for individual files as well as for an entire dataset. The links are created directly on the landing page of the dataset by clicking on the corresponding icon. You can then copy the links and pass them on.

Screenshot to demonstrate the admin token functionality in TUdatalib

7 Using TUdatalib via REST interface

With the REST interface there is a way to create data sets and add files and metadata to them without using the web interface of TUdatalib. This is presented here using the programming language Python 3 together with the Python package 'requests'. You can also use any other programming language.

7.1 Required authorizations

You usually need administrator privileges for the collection in which you want to create a data set.

You also need administrator privileges if you want to add files and metadata to an existing data set after its submission is completed, otherwise the publication authorization is sufficient to add files and metadata to a data set whose submission was started but not completed.

7.2 Installing the Python Requests Library

Debian: To install the requests library on your system pip3 is required. Open the command line (Terminal) and execute the following commands:

sudo apt-get install python3-pip
sudo pip3 install requests

Alternatively, the requests library can also be installed via apt-get. To do this, execute the following commands:

sudo apt-get upgrade
sudo apt-get update
sudo apt-get install python3-requests

Further information on requests and installation can be found here.

7.3 Download the sample scripts for Python

We provide sample scripts for creating data sets and uploading files and metadata to existing data sets, which, if Python and requests are properly installed, can be executed independently. You can find these for free use here: https://git.rwth-aachen.de/gerald.langhanke/tudatalib/.

7.4 Using the example scripts for Python

Download the directory and unzip it to any location. The exported directory contains the following files and folders:

The files 'create_item.py', 'upload_files.py' and 'upload_metadata.py' are the scripts you have to execute when you want to create a data set, upload files to a data set or add metadata to a data set, respectively. These scripts require information in corresponding configuration files located in the 'configs' folder.

For the scripts to work properly, you should fill out the configuration files before running the scripts. Each script has its own configuration file:

For example, 'create_item.py' accesses 'config_create_item.py' in order to obtain the information necessary for creating the data set. Please also refer to the notes in the 'README.md' file.

The 'sf' folder contains files necessary for the function of the scripts and nothing in it should be edited by the user! Especially do not delete the 'init.py' in the folders (if present). These are also necessary for the function of the scripts, even if they are empty.

7.5 Filling out the configuration files

Before executing the scripts in the main folder, you should fill in the corresponding configuration file in the subfolder 'configs'.

7.5.1 Content of config_create_item.py for creating data sets

The following information is required to create data sets in TUdatalib:

Internal ID of the collection in which you want to create a data set
Metadata such as data set name, author or date of submission

Specify the ID of the collection as a Python string, i.e. in quotation marks. The metadata of a data set consists of a list (Python list) of metadata entries. Each metadata entry is a Python dictionary that has at least the entries "key", "value" and "language". The entries under "key" determine which metadata field (Dublin Core Schema) it is. The entry under "value" defines the value of the field. Under "language" you can enter "None" if specification of a language is not applicable.

An example is given in the picture. For further information, please refer to the file itself or further down.

7.5.2 Content of config_upload_files.py for uploading files

If you want to add files to existing data sets, the following information is required:

Internal ID of the data set
Path to the folder of the files to be uploaded

Please note that the entire contents of the specified folder will always be added to the data set (the ID specified in the same sublist). If necessary, create a new folder with files to be added to a data set. Also, always specify the entire path to the folder, not just a part of it. An example is shown in the picture. For more information, see the file itself or below.

7.5.3 Content of config_update_metadata.py for adding metadata.

Internal ID of the data set
List of metadata entries (one per data set ID)

An example is shown in the picture. For more information, see the file itself or below.

7.5.4 How do I get the ID of a collection?

As an administrator of a collection, you can find out the ID of the target collection by navigating to the collection and then clicking "Edit Collection" on the left. In the tab "Assign Roles" you will then see the ID in the group name between "COLLECTION" and the role name (e.g. "ADMIN"). Underscores do NOT belong to the ID. Alternatively, you can also deduce the ID when creating or editing a data set in the collection. To do this, you must select the target collection in the "Authorizations" item when creating a data set and then read the ID in the group name as described above. Alternatively, you can click on "Edit this item" (in the navigation bar) and then on "Authoriazsations...". In this way, the ID of the collection can also be concluded from the group name of, e.g., the read permissions as described above.

7.5.5 How do I get the ID of a data set?

If you have permission to edit a published data set, you can click "Edit this item" in the navigation bar in the data set display. In the "Item Status" tab the ID of the item is then listed under "Item internal ID".

8 Visibility of published data sets

Data sets published in TUdatalib, especially those with DOI, are made visible in many places on the web, e.g:

Google und Google Scholar
OpenAIRE: https://explore.openaire.eu/search/dataprovider?datasourceId=re3data_____::dd850814f13a8499f57349bab338ad17
DataCite Search: https://search.datacite.org/repositories/tib.ulbd
BASE: https://www.base-search.net/Search/Results?type=all&lookfor=*&filter%5B%5D=f_dccollection%3A%22ftulbdarmstadtda%22

9 Personal data

Your research data may not contain any personal data. If necessary, the data must be made anonymous. If you have any questions, please also contact the ethics commission of the TU: https://www.intern.tu-darmstadt.de/gremien/ethikkommisson/index.en.jsp

10 External Review

TUdatalib can always create temporary (valid for 48 hours) one-time links (e.g. for external reviewers). See the point: "Processing access requests to restricted data sets". Currently there is no possibility for administrators to actively create these links themselves, requiring reviewers to submit an access request first.

11 Contact persons

For questions and suggestions please contact the TUdata team:

tudata@tu-darmstadt.de
06151/16-76413, -76420, -76373
You can also create a ticket directly in our issue tracker at tudatalib@ulb.tu-darmstadt.de.