HiActs-Conference-Poll/seed_demo.py

"""
Demo seeder for boss presentation.
Populates 20 simulated participant responses, then prints instructions.

Usage:
    python3 seed_demo.py
"""

import urllib.request
import urllib.error
import json

BASE_URL = "http://localhost:5001"


def post(path, data):
    body = json.dumps(data).encode()
    req = urllib.request.Request(
        BASE_URL + path,
        data=body,
        headers={"Content-Type": "application/json"},
        method="POST"
    )
    with urllib.request.urlopen(req) as resp:
        return json.loads(resp.read())


# 20 simulated participants, one response set each
# Designed to produce an interesting, realistic bubble chart

SIMULATED_RESPONSES = [
    # Q1: Role, Q2: Logbook word, Q3: Obstacle, Q4: Standardisation, Q5: Documentation, Q6: Vision
    (
        "Beamline Scientist",
        "chaotic",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Automated cross-facility comparison without manual conversion"
    ),
    (
        "Beamline Scientist",
        "fragmented",
        "Inconsistent logbook formats",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Shared analysis pipelines that work on any beamline data"
    ),
    (
        "Industry User",
        "opaque",
        "Missing or incomplete metadata",
        "Necessary and overdue",
        "Barely documented",
        "Getting our own data back without asking the beamline scientist to translate it"
    ),
    (
        "Beamline Scientist",
        "manual",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Desirable but very difficult",
        "Mostly documented, some gaps",
        "Reproducible experiments across DESY, ESRF and Diamond"
    ),
    (
        "Instrument Scientist",
        "inconsistent",
        "Manual re-scripting for every beamline",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "One analysis script that runs on data from any facility worldwide"
    ),
    (
        "Industry User",
        "black box",
        "Missing or incomplete metadata",
        "Necessary and overdue",
        "Barely documented",
        "Direct data delivery to our in-house systems without post-processing"
    ),
    (
        "Control System / IT",
        "custom",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Desirable but very difficult",
        "Mostly documented, some gaps",
        "A single integration layer instead of ten bespoke adapters"
    ),
    (
        "Beamline Scientist",
        "paper",
        "Inconsistent logbook formats",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Pooled datasets across facilities for machine learning models"
    ),
    (
        "Industry User",
        "incomplete",
        "No common file format standard",
        "Desirable but very difficult",
        "Barely documented",
        "Automated quality reports in a format our QA team actually understands"
    ),
    (
        "Beamline Scientist",
        "scattered",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Publishing FAIR data with zero extra effort after the experiment"
    ),
    (
        "Instrument Scientist",
        "fragmented",
        "Inconsistent logbook formats",
        "Desirable but very difficult",
        "Mostly documented, some gaps",
        "Collaborative experiments where two facilities contribute data to one analysis"
    ),
    (
        "Industry User",
        "slow",
        "Missing or incomplete metadata",
        "Too rigid for our experimental needs",
        "Mostly documented, some gaps",
        "Faster time from experiment to usable result for our product development"
    ),
    (
        "Beamline Scientist",
        "mixed",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Finding and reusing data from experiments done five years ago"
    ),
    (
        "Control System / IT",
        "outdated",
        "Manual re-scripting for every beamline",
        "Desirable but very difficult",
        "Partial, depends on the scientist",
        "Plug-and-play data connectors between control systems and analysis tools"
    ),
    (
        "Instrument Scientist",
        "informal",
        "Inconsistent logbook formats",
        "Necessary and overdue",
        "Barely documented",
        "Automatic FAIR compliance without a separate data curation step"
    ),
    (
        "Beamline Scientist",
        "unreliable",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Necessary and overdue",
        "Mostly documented, some gaps",
        "Comparing strain measurements across different beamlines in one click"
    ),
    (
        "Industry User",
        "confusing",
        "No common file format standard",
        "Not our problem to solve",
        "Barely documented",
        "Receiving data already formatted for our simulation software"
    ),
    (
        "Beamline Scientist",
        "analog",
        "Inconsistent logbook formats",
        "Desirable but very difficult",
        "Fully documented, reproducible",
        "A European data commons for synchrotron experiments"
    ),
    (
        "Instrument Scientist",
        "messy",
        "Different control systems (TANGO, EPICS, SPOCK)",
        "Necessary and overdue",
        "Partial, depends on the scientist",
        "Less time cleaning data, more time doing science"
    ),
    (
        "Beamline Scientist",
        "evolving",
        "Missing or incomplete metadata",
        "Too rigid for our experimental needs",
        "Mostly documented, some gaps",
        "Real-time data quality feedback during the experiment itself"
    ),
]


def main():
    print("Resetting existing data...")
    post("/admin/reset", {})

    print("Seeding 20 simulated participants...")
    entries = []
    for row in SIMULATED_RESPONSES:
        q1, q2, q3, q4, q5, q6 = row
        entries += [
            {"question_id": 1, "answer": q1},
            {"question_id": 2, "answer": q2},
            {"question_id": 3, "answer": q3},
            {"question_id": 4, "answer": q4},
            {"question_id": 5, "answer": q5},
            {"question_id": 6, "answer": q6},
        ]

    result = post("/admin/seed", entries)
    print(f"Done. {result['total']} total answers loaded.\n")

    print("=" * 56)
    print("  PRESENTATION READY")
    print("=" * 56)
    print()
    print("  Display (bubble chart, fullscreen):")
    print("  http://localhost:5001/display?demo=1")
    print()
    print("  Poll (for boss to try on phone):")
    print("  http://localhost:5001")
    print()
    print("  Admin (reset / view raw data):")
    print("  http://localhost:5001/admin")
    print("=" * 56)


if __name__ == "__main__":
    main()