marcoreidelbach / mardmo-export-plugin Goto Github PK

Let the user identify him/herself in the MaRDMO Questionnaire and add the creator to the Documentation Entry. Allow MaRDMO creators to update their own workflows through MaRDMO.

MaRDMO should be able to suggest/retrieve methods to/from AlgoData.

Instead of free selection, select non-math disciplines and mathematical areas.

Next to workflow doc and search, pure model documentation should be possible. Models documented like this should be exported to the TA4 Model template.

Integrate DADS (Dictionary of Algorithm and Data Structures) to have data structures selectable with a DADS ID

When Workflow is exported to Portal all non-MaRDI IDs should be replaced by the corresponding (newly created) MaRDI IDs.

Prepare MaRDMO to be compatible with RDMO 2.0.

Wikidata Query Limit

Connect Provider MaRDI Wikidata Provider

Connect MaRDMO to WikiChip for Processors, Wikichip ID is a valid Wikidata Property

Move to data as json file

MaRDMO should be able to suggest/retrieve models to/from ModelDB.

If user is present on MaRDI Portal (but not identified through orcid or zbmath id, e.g. via wikidata qid) the author is not recognized and thus workflow modifications are not possible.

If a Model is selected from MathModDB for Workflow Documentation / Model Documentation, the connected Task should be show and selectable. If no appropriate Task is present should be possible to define.

This means, collect Task related to model. if Task is selected in workflow mode, collect further Information to fill var / para table. If no suitable task (workflow /model mode) handler needs to be adjusted to allow the creation of a new task.

Ultimately, most information stored in template should end up in MaRDI KG. The template should then be rendered from the KG to prevent diverging sources of information.

Change SPARQL queries in background

Instead of having links in the individual tables, e.g. to the documentation of a software, display 'Yes' as link to user provided destination.

If user creates sets for, e.g.software, method, etc, and deletes it:

• deleted set is kept with all questions unanswered which leads to stop of MaRDMO
• in any case the answers gathered by MaRDMO are sorted wrong
• in RDMO a set deletion leads to a buggy behavior (?), i.e. in following sets conditions are ignored etc

Introduce controlled vocabulary whenever possible.

Before unanswered question had an empty string, this changed leading to missing entries in the tables in the html / wiki page export. This was solved by adding default answers to all questions. In the future this should be solved by adjusting the template creation and at the same time removing once again the standard answers.

Upon the answer of specific questions, several other question are automatically answered through a handler in the background.

Already done for publication.

To do:
Model
Method
Software
Input/Output

Add Model Only Documentation Possibility

e.g. to select software in method section, to select math area, ...

Add more checks to properly handle missing descriptions in wikidata entries.

Entries from Wikidata without a description should also be added to MaRDI KG without description.

Users can select there Processor from Wikichip.. A part of the URL is used as wikichip ID, the handler should use the ID and check if the processor is already on Wikidata / MaRDI KG.

Problems:

1. the url returned by wikichip is not the url used by wikidata as id. correct id could be extracted from requested text (only in situatiosn where there is an automatic redirect)
2. on some pages there is no direct re-direct, but a link, how to deal with this

Add conditions to prevent the display of irrelevant question. Questions with standard option sets require a standard answer etc.

Re-write certain help texts.

This will allow to have one section for Variable and Devices.

So far provider output is id <|> label <|> description as title, should be change to label - description (id). The original triple should be stored as external id, which should be used by MaRDMO.

When User selects a wikidata/mardi entry with ' in label/description, subsequent SPARQL query fails.

Merge both sections, add additional question, where users can define if the data is input or output (or both). Get Information if data is Input or Output from assignment in process step section.

Re-do Workflow search code

Move Section to end and use dynamic option sets for most questions to select methods, software, etc entered before.

Put API requests to portal / wikidata in individual sessions

Should MaRDMO be mapped to process ontology?

1. If software is selected from mardi/wikidata, relevant programming languages should be identified and added to the documentation template.

2. Identify other entries where this is possible!