The Support Library for Developing Scientific Data Collections Access Portals

The article discusses an approach to the automated construction of web-portals for access to scientific data collections based on the data domain model and a support library used to generate a web interface “online”. Automation of construction is based on an architectural generation scheme, which contains: a modeling apparatus represented by a template graph of a metalanguage and prototypes of description elements in a metalanguage; a domain model, consisting of a graphical ER-model and its objects representation; and a life cycle, consisting of a model in terms of Petri nets and an object representation for events in the model. Source code generation is performed in two stages. At the first stage, the objects of the original model form the view in each functional layer. In the second stage, generator produces output files for each functional layers corresponding to life cycle events. The support library presented in this article simplifies the generation of an interconnected set of files necessary to build the interface using the Django framework, which consists of: a page template, a CSS-style file for the page design, Javascript functions for the browser-side user interface, a Python file with a class corresponding to an interface page, as well as AJAX functions on the web server side required to display complex data. To ensure the consistency of the resulting set of files, URLs for AJAX functions are generated automatically and proxy functions in Javascript are produced that make a remote call to the server. The support library described in the article and the approach to automating the construction of information systems allows to formalize the users' understanding of the subject area, describe the interface and specify the data processing methods.