TKB_plan/web/import_excel.py

#!/usr/bin/env python3
"""Parse Excel schedule file and output JSON to stdout."""
import json
import sys
from datetime import date
import openpyxl

filepath = sys.argv[1]
wb = openpyxl.load_workbook(filepath, data_only=True)
ws = wb.active

# Also load with styles for comments
wb2 = openpyxl.load_workbook(filepath)
ws2 = wb2.active

# --- Build dayIndex from row 9 (months) and row 11 (days) ---
month_starts = {}
for col in range(7, 300):
    val = ws.cell(row=9, column=col).value
    if val is not None and hasattr(val, 'month'):
        month_starts[col] = (val.year, val.month)

if not month_starts:
    print(json.dumps({"error": "No month data found in row 9"}), file=sys.stderr)
    sys.exit(1)

day_index = []
for col in range(7, 300):
    day_val = ws.cell(row=11, column=col).value
    if day_val is None:
        continue
    day_num = int(day_val)

    current_month = None
    for mcol in sorted(month_starts.keys(), reverse=True):
        if col >= mcol:
            current_month = month_starts[mcol]
            break

    if current_month is None:
        continue

    year, month = current_month
    idx = col - 7

    try:
        d = date(year, month, day_num)
        is_weekend = d.weekday() >= 5
        week = d.isocalendar()[1]
    except ValueError:
        is_weekend = False
        week = 0

    day_index.append({
        "idx": idx,
        "day": day_num,
        "month": month,
        "year": year,
        "week": week,
        "weekend": is_weekend,
    })

# --- Extract station data ---
valid_idx = set(d["idx"] for d in day_index)
stations = []

for row in range(13, 40):
    code = ws.cell(row=row, column=1).value
    name = ws.cell(row=row, column=2).value or ""
    server = ws.cell(row=row, column=3).value or ""

    if not code:
        continue

    data = {}
    for col in range(7, 300):
        idx = col - 7
        if idx not in valid_idx:
            continue
        val = ws.cell(row=row, column=col).value
        if val is None:
            continue
        if isinstance(val, str) and val.strip() == "":
            continue

        entry = {}
        if isinstance(val, (int, float)):
            entry["v"] = int(val) if val == int(val) else val
        else:
            v = str(val).strip()
            # Normalize: lowercase z -> uppercase Z
            if v == 'z':
                v = 'Z'
            entry["v"] = v

        data[str(idx)] = entry

    stations.append({
        "code": str(code).strip(),
        "name": str(name).strip(),
        "server": str(server).strip(),
        "duration": None,
        "data": data,
    })

# --- Extract DEN row comments ---
day_comments = []
for col in range(7, 300):
    cell = ws2.cell(row=11, column=col)
    if cell.comment:
        idx = col - 7
        if idx not in valid_idx:
            continue
        text = cell.comment.text
        # Extract actual comment from threaded format (Czech or English)
        if "Komentář:\n" in text:
            note = text.split("Komentář:\n")[-1].strip()
        elif "Comment:\n" in text:
            note = text.split("Comment:\n")[-1].strip()
        else:
            note = text.strip()
        if note:
            day_comments.append({"dayIdx": idx, "text": note})

# --- Extract cell comments from data rows ---
cell_comments = []
for row in range(13, 40):
    code = ws2.cell(row=row, column=1).value
    if not code:
        continue
    code = str(code).strip()
    for col in range(7, 300):
        cell = ws2.cell(row=row, column=col)
        if cell.comment:
            idx = col - 7
            if idx not in valid_idx:
                continue
            text = cell.comment.text
            if "Komentář:\n" in text:
                note = text.split("Komentář:\n")[-1].strip()
            elif "Comment:\n" in text:
                note = text.split("Comment:\n")[-1].strip()
            else:
                note = text.strip()
            if note:
                cell_comments.append({"stationCode": code, "dayIdx": idx, "text": note})

# Deduplicate dayIndex — keep only first occurrence of each date
seen_dates = set()
deduped_day_index = []
for d in day_index:
    date_key = (d["year"], d["month"], d["day"])
    if date_key not in seen_dates:
        seen_dates.add(date_key)
        deduped_day_index.append(d)
day_index = deduped_day_index

result = {
    "dayIndex": day_index,
    "stations": stations,
    "obstacles": [],
    "dayComments": day_comments,
    "cellComments": cell_comments,
}

print(json.dumps(result, ensure_ascii=False))