This is a prototype for the paper on FLOWER (FLOW + ER): A technique for extending Entity-Relationship diagrams. This paper is published in the proceedings of DaWaK 2023.
Authors: Elijah Mitchell, Nabila Berkani, Carlos Ordonez, Ladjel Bellatreche
Paper citation: E. Mitchell, N. Berkani, L. Bellatreche, C. Ordonez 'FLOWER: Viewing Data Flow in ER Diagrams', DAWAK Conference 2023
@InProceedings{Mitchell2023,
author = {Elijah Mitchell, Nabila Berkani, Ladjel Bellatreche, Carlos Ordonez},
booktitle = {DaWaK Conference},
title = {FLOWER: Viewing Data Flow in ER Diagrams},
year = {2023},
}
Paper hosted at https://www2.cs.uh.edu/~ordonez/co_research_proceedings.html :: pdf :: powerpoint
ER is useful but limited to static relationships: We cannot see how data moves or their relationship with other data. By extending our view of ER to include the concept of "Flows" between entities, we can better understand systems at a glance.
This prototype is meant for demonstrating a working implementation of the algorithm for analyzing code for use with FLOWER. Provided with a set of files, the program will parse the top level semantic trees (non-functional, non class, current module only) and construct an analysis of the inputs, outputs and relationships between these through the pipeline.
The analysis portion of this project uses python 3.10 -- no external libraries are required.
usage: __init__.py [-h] [-v] [-o OUTPUT] [-e ERDIAG] [-d] filename
positional arguments:
filename Relative or absolute path to file(s) to parse. Supports glob.
options:
-h, --help show this help message and exit
-v, --verbose
-o OUTPUT, --output OUTPUT
Write the analysis (a json file). If unset, prints output directly.
-e ERDIAG, --erdiag ERDIAG
Output FLOWER diagram json files. Will be appended with _ents.json and _rels.json.
-d, --derived Also include derived entities (internal state nodes like dataframes) in ER output.
The diagram generator uses node.js. View the README in diagram/ for instructions on how to run it.