projectStructure

The query executor is designed as a purely RESTful web service. It does not provide any html rendered pages, but rather exposes a number of endpoints delivering and accepting data. The FLASK framework (https://flask.palletsprojects.com/en/1.1.x/) is used to create the endpoints.

The query executor makes use of the application factory pattern (https://flask.palletsprojects.com/en/1.1.x/patterns/appfactories/) with the creation of an flask instance triggered by the create_app() function in the app/ini.py.

the overall collection of endpoints is grouped into several blueprints, each responsible for a certain functional area. The most important ones are:

• projects (creating and manipulating projects)
• query (handling of the queries in the internal XML format, transforming into Scopus queries)
• EIDs (collecting eids for queries, downloads of complete and random lists)
• collector (collecting the full records for the list of eids)
• status (endpoints to monitor the status of individual collections/queries)
• scival (upload of scival data)
• relevance_measures (calculation of relevance measures with externally provided data)
• facettes (uploading facettes in order to generate survey data)
• query_viewer (applying Stylesheet to display queries from XML syntax)
• survey_analyzer (downloading and analyzing data from SurveyGizmo v4 to v5 update)

In addition several blueprints exist, which are created to achieve thought-of functionalaties and future developments:

• wheel (started creation of donuts for SDG assignment display)
• keywords (extraction of keywords from indexed data, not maintained)
• analysis (automated overlap analysis, very slow, not yet reasonably implemented)
• identifier (planned abstraction from EIDs)
• crossref (using crossref fuzzy search tzo obtain EIDs)
• main (abandoned start of implementing html pages)

each of the blueprints resides in its own module within the app folder. Each contains a XXX_routes.py, defining the individual endpoints.