#!/usr/bin/env python3
"""Hypothetical elapsed-calendar-day schedule, not a trial planning standard.

Save beside clinical-trial-critical-path-example.csv. No packages required:
  python3 clinical-trial-critical-path.py
  python3 clinical-trial-critical-path.py --duration EDC=10
  python3 clinical-trial-critical-path.py --duration S=35

Finish dates are exclusive boundaries: a task starting Jan 4 with duration 10
finishes Jan 14; its successor can start Jan 14. No holidays/business-day logic,
resource limits, probabilistic estimates, clinical holds or rework are modeled.
"""

import argparse
import csv
import json
from datetime import date, timedelta
from pathlib import Path


def schedule(path, start, overrides):
    with path.open(newline="") as handle:
        rows = list(csv.DictReader(handle))
    tasks = {}
    for row in rows:
        key = row["id"]
        if not key or key in tasks:
            raise ValueError("Task ids must be nonempty and unique")
        duration = int(row["duration_calendar_days"])
        if duration < 0:
            raise ValueError("Durations must be nonnegative integer calendar days")
        tasks[key] = {"id": key, "task": row["task"], "duration": duration,
                      "predecessors": row["predecessors"].split(";") if row["predecessors"] else []}
    if not tasks:
        raise ValueError("Schedule must contain tasks")
    for key, value in overrides.items():
        if key not in tasks or value < 0:
            raise ValueError("Overrides need a known task and nonnegative integer duration")
        tasks[key]["duration"] = value
    for task in tasks.values():
        if any(key not in tasks for key in task["predecessors"]):
            raise ValueError("Unknown predecessor")
    ordered = []
    remaining = dict(tasks)
    while remaining:
        ready = [key for key, task in remaining.items()
                 if all(p not in remaining for p in task["predecessors"])]
        if not ready:
            raise ValueError("Dependency cycle")
        for key in ready:
            task = remaining.pop(key)
            task["early_start"] = max((tasks[p]["early_finish"] for p in task["predecessors"]), default=0)
            task["early_finish"] = task["early_start"] + task["duration"]
            ordered.append(key)
    finish = max(task["early_finish"] for task in tasks.values())
    for key in reversed(ordered):
        task = tasks[key]
        successors = [t for t in tasks.values() if key in t["predecessors"]]
        task["late_finish"] = min((t["late_start"] for t in successors), default=finish)
        task["late_start"] = task["late_finish"] - task["duration"]
        task["total_float"] = task["late_start"] - task["early_start"]
        task["start_date"] = (start + timedelta(days=task["early_start"])).isoformat()
        task["finish_date"] = (start + timedelta(days=task["early_finish"])).isoformat()
    return {"assumption": "Hypothetical; all durations are elapsed calendar days; no resource constraints",
            "start_date": start.isoformat(), "finish_date": (start + timedelta(days=finish)).isoformat(),
            "elapsed_days": finish, "critical_tasks": [k for k in ordered if tasks[k]["total_float"] == 0],
            "tasks": [tasks[k] for k in ordered]}


def main():
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--csv", type=Path, default=Path(__file__).with_name("clinical-trial-critical-path-example.csv"))
    parser.add_argument("--start", type=date.fromisoformat, default=date(2027, 1, 4))
    parser.add_argument("--duration", action="append", default=[], metavar="ID=DAYS")
    args = parser.parse_args()
    try:
        overrides = {}
        for item in args.duration:
            key, value = item.split("=", 1)
            overrides[key] = int(value)
        print(json.dumps(schedule(args.csv, args.start, overrides), indent=2))
    except (ValueError, KeyError) as error:
        parser.error(str(error))


if __name__ == "__main__":
    main()
