import glob
import os
import shutil
import pandas as pd
from openpyxl import load_workbook
from openpyxl.styles import Font, PatternFill, Border, Side, Alignment
from openpyxl.utils import get_column_letter
from datetime import date, datetime

# Paths
src_dir = os.path.dirname(os.path.abspath(__file__)) + "/"
out_dir = "U:/Dropbox/!!!Days/Downloads Z230/"

# Find source files
src_files = glob.glob(src_dir + "Protocol 77242113UCO3001 - All Samples*.xlsx")
assert src_files, "Source file not found!"
src_file = src_files[0]
print(f"Source xlsx: {src_file}")

csv_files = glob.glob(src_dir + "_EDCStdRpt-DataListing.csv")
assert csv_files, "CSV file not found!"
csv_file = csv_files[0]
print(f"Source csv:  {csv_file}")

# Delete old output report if exists
today = date.today().strftime("%Y-%m-%d")
out_filename = f"{today} 77242113UCO3001 Speciment Inventory report.xlsx"
out_path = out_dir + out_filename

for old in glob.glob(out_dir + "*77242113UCO3001 Speciment Inventory report*.xlsx"):
    os.remove(old)
    print(f"Deleted old: {old}")

# Copy source file to output — preserves all formatting perfectly
shutil.copy2(src_file, out_path)

# Load data with pandas for analysis
df = pd.read_excel(src_file, sheet_name=0, header=0)

# All unique patients
all_patients = sorted(df['Patient No.'].dropna().unique())

# BXSCR and DNA rows
bxscr = df[df['Protocol Visit Code'] == 'BXSCR']
dna   = df[df['Protocol Visit Code'] == 'DNA']

# Parse date value to datetime object
def fmt_date(val):
    if pd.isna(val):
        return None
    if isinstance(val, str):
        return datetime.strptime(val, '%d-%b-%Y')
    return pd.to_datetime(val).to_pydatetime()

# Get Container Receipt Date + Excel row for patient+specimen from given visit dataframe
OK_STATUSES = {'Received', 'In Inventory', 'Shipped'}

# Excel row = pandas df index + 2 (row 1 header, data from row 2)
def get_specimen_info(visit_df, patient, specimen_type=None):
    rows = visit_df[visit_df['Patient No.'] == patient]
    if specimen_type:
        rows = rows[rows['Specimen Type'] == specimen_type]
    rows = rows[rows['Sample Status'].isin(OK_STATUSES)]
    if rows.empty:
        return '', None
    row = rows.iloc[0]
    return fmt_date(row['Container Receipt Date']), rows.index[0] + 2

# Get Container Receipt Date + Excel row by Container Label Line 1 code and visit code
def get_label_info(patient, label_code, visit_code):
    rows = df[(df['Patient No.'] == patient) &
              (df['Protocol Visit Code'] == visit_code) &
              (df['Container Label Line 1'] == label_code)]
    rows = rows[rows['Sample Status'].isin(OK_STATUSES)]
    if rows.empty:
        return '', None
    row = rows.iloc[0]
    return fmt_date(row['Container Receipt Date']), rows.index[0] + 2

# Open copied workbook and add analysis sheet
out_wb = load_workbook(out_path)

# Rename and autofit first sheet
src_ws = out_wb.active
src_ws.title = "Zdroj"
for col in src_ws.columns:
    max_len = max((len(str(cell.value)) if cell.value is not None else 0) for cell in col)
    src_ws.column_dimensions[get_column_letter(col[0].column)].width = min(max_len + 2, 50)

# Create analysis sheet
analysis_ws = out_wb.create_sheet("Přehled vzorků")

thin = Side(style='thin')
border = Border(left=thin, right=thin, top=thin, bottom=thin)

# Column definitions: (header, width)
# No "Visit code" column
columns = [
    ("Investigator Name",   24),
    ("Číslo pacienta",      20),
    ("Máme biopsii SM11",   20),  # col 3  → SCREENING C:E
    ("Máme RNA",            16),
    ("Máme Cryostor",       16),
    ("DNA",                 14),  # col 6  → no group
    ("PLASMPK I-0 TROUGH",  18),  # col 7  → RANDOMIZACE I-0 G:L
    ("PLASMA PK I-0 PEAK",  18),
    ("SERUM ADA I-0 PRE",   18),
    ("SM06/SERUM BIOM",     16),
    ("SM07/WB RNA",         14),
    ("SM10/FECAL",          14),
    ("PLASMPK I-2 TROUGH",  18),  # col 13 → I-2 M:P
    ("PLASMA PK I-2 PEAK",  18),
    ("SERUM ADA I-2 PRE",   18),
    ("STOOL I-2",           12),
    ("PLASMPK I-4 TROUGH",  18),  # col 17 → I-4 Q:V
    ("PLASMA PK I-4 PEAK",  18),
    ("SERUM ADA I-4 PRE",   18),
    ("SM06/SERUM BIOM",     16),
    ("SM07/WB RNA",         14),
    ("STOOL I-4",           12),
]

# Row 1 — group headers
group_font  = Font(name='Calibri', bold=True, size=11)
group_fill  = PatternFill("solid", fgColor="FFFFFF")  # white, same as user's theme=0
group_border = Border(left=thin, right=thin, top=thin, bottom=thin)

groups = [
    (3,  5,  "SCREENING"),
    (7,  12, "RANDOMIZACE I-0"),
    (13, 16, "I-2"),
    (17, 22, "I-4"),
]
for start_col, end_col, label in groups:
    analysis_ws.merge_cells(start_row=1, start_column=start_col, end_row=1, end_column=end_col)
    cell = analysis_ws.cell(row=1, column=start_col, value=label)
    cell.font = group_font
    cell.fill = group_fill
    cell.alignment = Alignment(horizontal='center', vertical='center')
    cell.border = group_border
    # apply border to all merged cells
    for c in range(start_col, end_col + 1):
        analysis_ws.cell(row=1, column=c).border = group_border

analysis_ws.row_dimensions[1].height = 20

# Row 2 — column headers
header_fill = PatternFill("solid", fgColor="4472C4")
header_font = Font(name='Calibri', bold=True, size=11, color="FFFFFF")

for col_idx, (header, width) in enumerate(columns, 1):
    cell = analysis_ws.cell(row=2, column=col_idx, value=header)
    cell.font = header_font
    cell.fill = header_fill
    cell.border = border
    cell.alignment = Alignment(horizontal='center', vertical='center', wrap_text=True)
    analysis_ws.column_dimensions[get_column_letter(col_idx)].width = width

analysis_ws.row_dimensions[2].height = 30

# Freeze first 2 columns and first 2 rows
analysis_ws.freeze_panes = "C3"

# Data rows start at row 3
date_font_link = Font(name='Calibri', size=11, color="000000", underline='single')
yes_fill = PatternFill("solid", fgColor="E2EFDA")
no_fill  = PatternFill("solid", fgColor="FFE7E7")
data_font = Font(name='Calibri', size=11)

src_sheet_name = out_wb.sheetnames[0]
pat_sheet_name = "Seznam pacientů"

# Build patient → first Excel row in "Seznam pacientů" (header=row1, data from row2)
# pat_df is built later, but we need sorted order — pre-sort here too
_csv_df_pre = pd.read_csv(csv_file, encoding='utf-8')
_pat_pre = _csv_df_pre[['SiteNumber', 'Subject', 'Field4Value']].copy()
_pat_pre['Field4Value'] = _pat_pre['Field4Value'].apply(lambda v: datetime.strptime(str(v).strip(), '%d %b %Y') if pd.notna(v) else None)
_pat_pre = _pat_pre.sort_values(['SiteNumber', 'Subject', 'Field4Value']).reset_index(drop=True)
patient_row_map = {}
for i, row in _pat_pre.iterrows():
    pat = row['Subject']
    if pat not in patient_row_map:
        patient_row_map[pat] = i + 2  # +1 for 1-based, +1 for header row

# Only patients with any BXSCR record
bxscr_patients = sorted(bxscr['Patient No.'].dropna().unique())

for row_idx, patient in enumerate(bxscr_patients, 3):
    investigator = bxscr[bxscr['Patient No.'] == patient].iloc[0]['Investigator Name']
    sm11, sm11_row    = get_specimen_info(bxscr, patient, 'Tissue , Paraffin Block')
    rna,  rna_row     = get_specimen_info(bxscr, patient, 'Biopsy RNA Later')
    cryo, cryo_row    = get_specimen_info(bxscr, patient, 'Biopsy, Frozen Tissue')
    dna_date, dna_row = get_specimen_info(dna, patient)
    trough,  trough_row  = get_label_info(patient, 'PLASMPK I-0 TROUGH', 'I-0')
    peak,    peak_row    = get_label_info(patient, 'PLASMA PK I-0 PEAK',  'I-0')
    ada,     ada_row     = get_label_info(patient, 'SERUM ADA I-0 PRE',   'I-0')
    sm06,    sm06_row    = get_label_info(patient, 'SM06/SERUM BIOM',     'I-0')
    sm07,    sm07_row    = get_label_info(patient, 'SM07/WB RNA',          'I-0')
    sm10,    sm10_row    = get_label_info(patient, 'SM10/FECAL',           'I-0')
    trough2, trough2_row = get_label_info(patient, 'PLASMPK I-2 TROUGH',  'I-2')
    peak2,   peak2_row   = get_label_info(patient, 'PLASMA PK I-2 PEAK',  'I-2')
    ada2,    ada2_row    = get_label_info(patient, 'SERUM ADA I-2 PRE',   'I-2')
    stool2,  stool2_row  = get_label_info(patient, 'STOOL I-2',            'I-2')
    trough4, trough4_row = get_label_info(patient, 'PLASMPK I-4 TROUGH',  'I-4')
    peak4,   peak4_row   = get_label_info(patient, 'PLASMA PK I-4 PEAK',  'I-4')
    ada4,    ada4_row    = get_label_info(patient, 'SERUM ADA I-4 PRE',   'I-4')
    sm064,   sm064_row   = get_label_info(patient, 'SM06/SERUM BIOM',     'I-4')
    sm074,   sm074_row   = get_label_info(patient, 'SM07/WB RNA',          'I-4')
    stool4,  stool4_row  = get_label_info(patient, 'STOOL I-4',            'I-4')

    # col 1-2: plain text, col 3+: (date, excel_row) tuples
    row_data = [investigator, patient,
                (sm11, sm11_row), (rna, rna_row), (cryo, cryo_row), (dna_date, dna_row),
                (trough, trough_row), (peak, peak_row), (ada, ada_row),
                (sm06, sm06_row), (sm07, sm07_row), (sm10, sm10_row),
                (trough2, trough2_row), (peak2, peak2_row), (ada2, ada2_row), (stool2, stool2_row),
                (trough4, trough4_row), (peak4, peak4_row), (ada4, ada4_row),
                (sm064, sm064_row), (sm074, sm074_row), (stool4, stool4_row)]

    for col_idx, value in enumerate(row_data, 1):
        if col_idx <= 2:
            cell = analysis_ws.cell(row=row_idx, column=col_idx, value=value)
            if col_idx == 2 and patient in patient_row_map:
                cell.hyperlink = f"#'{pat_sheet_name}'!B{patient_row_map[patient]}"
                cell.font = Font(name='Calibri', size=11, underline='single')
            else:
                cell.font = data_font
        else:
            dt, excel_row = value
            cell = analysis_ws.cell(row=row_idx, column=col_idx, value=dt)
            if dt and excel_row is not None:
                cell.hyperlink = f"#'{src_sheet_name}'!A{excel_row}"
                cell.font = date_font_link
                cell.fill = yes_fill
                cell.number_format = 'DD-MMM-YYYY'
            else:
                cell.font = Font(name='Calibri', size=11, color="C00000")
                cell.fill = no_fill
        cell.border = border
        cell.alignment = Alignment(horizontal='center', vertical='center')

# ── Seznam pacientů sheet ────────────────────────────────────────────────────
csv_df = pd.read_csv(csv_file, encoding='utf-8')

patients_ws = out_wb.create_sheet("Seznam pacientů")

pat_columns = [
    ("Číslo centra",   20),
    ("Číslo pacienta", 20),
    ("Kód návštěvy",   20),
    ("Datum návštěvy", 16),
    ("Typ návštěvy",   16),
]

# Header row
for col_idx, (col_name, width) in enumerate(pat_columns, 1):
    cell = patients_ws.cell(row=1, column=col_idx, value=col_name)
    cell.font = header_font
    cell.fill = header_fill
    cell.border = border
    cell.alignment = Alignment(horizontal='center', vertical='center', wrap_text=True)
    patients_ws.column_dimensions[get_column_letter(col_idx)].width = width

patients_ws.row_dimensions[1].height = 30
patients_ws.freeze_panes = "A2"

# Prepare and sort data
def parse_date(val):
    if pd.isna(val) or str(val).strip() == '':
        return None
    try:
        return datetime.strptime(str(val).strip(), '%d %b %Y')
    except:
        return None

pat_df = csv_df[['SiteNumber', 'Subject', 'InstanceName', 'Field4Value', 'Field5Value']].copy()
pat_df['Field4Value'] = pat_df['Field4Value'].apply(parse_date)
pat_df = pat_df.sort_values(['SiteNumber', 'Subject', 'Field4Value']).reset_index(drop=True)

# Data rows
for row_idx, row in enumerate(pat_df.itertuples(index=False), 2):
    for col_idx, value in enumerate(row, 1):
        cell = patients_ws.cell(row=row_idx, column=col_idx, value=value)
        cell.font = data_font
        cell.border = border
        cell.alignment = Alignment(horizontal='center', vertical='center')
        if col_idx == 4 and value is not None:
            cell.number_format = 'DD-MMM-YYYY'

out_wb.save(out_path)
print(f"Saved: {out_path}")
print(f"Patients with BXSCR: {len(bxscr_patients)}")
print(f"All unique patients: {len(all_patients)}")