administrator/janssen
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

z230

Browse Source
This commit is contained in:
vladimir.buzalka
2026-05-05 12:19:51 +02:00
parent 5103cac2c9
commit 10eba225e7
5 changed files with 772 additions and 341 deletions
Download Patch File Download Diff File Expand all files Collapse all files
IWRS/Drugs/Working/create_accountability_report.py
+226 -230
View File
@@ -1,3 +1,6 @@
import sys
import os
import mysql.connector
import pandas as pd
from datetime import date
from pathlib import Path
@@ -5,51 +8,15 @@ from openpyxl import load_workbook
from openpyxl.styles import Font, PatternFill, Alignment, Border, Side
from openpyxl.utils import get_column_letter
STUDY = "42847922MDD3003"
#STUDY = "77242113UCO3001"
sys.path.insert(0, os.path.join(os.path.dirname(os.path.abspath(__file__)), ".."))
import db_config
INVENTORY_DIR = Path(f"xls_reports_{STUDY}")
DESTRUCTION_DIR = Path(f"xls_ip_destruction_{STUDY}")
SHIPMENTS_FILE = Path(f"xls_shipments_{STUDY}/shipments_report_{STUDY}.xlsx")
DETAILS_DIR = Path(f"xls_shipment_details_{STUDY}")
OUTPUT_DIR = Path("output")
OUTPUT_FILE = OUTPUT_DIR / f"{date.today().strftime('%Y-%m-%d')} {STUDY} CZ IWRS overview.xlsx"
STUDY = "42847922MDD3003"
# STUDY = "77242113UCO3001"
SHIPMENT_DROP_COLS = {
"Location", "Shipped Date", "Delivered Date [UTC]",
"Delivery Recipient", "Delivery Details", "Cancelled Date",
"Tracking #", "Total Medication IDs",
"Shipping Category", "Study", "Destination Location", "Destination Site",
"Medication type", "Container ID", "Quantity of Medication IDs",
"Packaged Lot description",
}
# ── Shared constants ──────────────────────────────────────────────────────────
COLUMN_RENAMES = {
"Site": "Site",
"Medication ID": "Med ID",
"Packaged Lot number": "Lot No.",
"Original Expiration Date when Packaged Lot was Added": "Orig Exp Date",
"Expiration date": "Exp Date",
"Received Date": "Rcv Date",
"Shipment Receipt User": "Rcpt User",
"Subject Identifier": "Subject ID",
"Quantity Assigned": "Qty Asgn",
"IRT Transaction": "IRT Tx",
"Date Assigned": "Date Asgn",
"Assignment User": "Asgn User",
"Dispensation Status": "Disp Status",
"Dispensing Date": "Disp Date",
"Dispensing date": "Disp Date",
"Quantity Dispensed": "Qty Disp",
"Dispensing User": "Disp User",
"Quantity Returned": "Qty Ret",
"Date Returned": "Date Ret",
"Return User": "Ret User",
"DestroyedOn": "Destroyed",
"Basket number": "Basket No.",
}
BASE_DIR = Path(os.path.dirname(os.path.abspath(__file__)))
OUTPUT_DIR = BASE_DIR / "output"
OUTPUT_FILE = OUTPUT_DIR / f"{date.today().strftime('%Y-%m-%d')} {STUDY} CZ IWRS overview.xlsx"
DATE_COLUMNS = {
"Orig Exp Date", "Exp Date", "Rcv Date",
@@ -81,44 +48,197 @@ COLUMN_WIDTHS = {
"Max Visit Date": 16,
}
# ── Helpers ───────────────────────────────────────────────────────────────────
def read_inventory(path):
df = pd.read_excel(path, header=None)
# Support both "Medication ID" (MDD3003) and "Medication" (UCO3001)
mask = df[0].isin(["Medication ID", "Medication"])
meta = {}
for i in range(len(df)):
val = str(df.iloc[i, 0]) if pd.notna(df.iloc[i, 0]) else ""
if val.startswith("Site:"):
meta["site"] = val.replace("Site:", "").strip()
if not mask.any():
print(f" {path.name}: no data (skipping)")
return None, meta
header_row = df[mask].index[0]
data = pd.read_excel(path, header=header_row)
data = data.rename(columns={"Medication": "Medication ID"})
return data, meta
# shipments sheet: kolík kde začínají detail sloupce (1-based, pro format_shipment_sheet)
N_SHIP_COLS = 9
def read_destruction_lookup():
lookup = {}
for path in DESTRUCTION_DIR.glob("*.xlsx"):
df = pd.read_excel(path, header=None)
basket_id = None
destroyed_on = None
for i in range(15):
val = str(df.iloc[i, 0]) if pd.notna(df.iloc[i, 0]) else ""
if val.startswith("Basket ID:"):
basket_id = val.replace("Basket ID:", "").strip()
if val.startswith("Drug Destruction Created Date:"):
destroyed_on = val.replace("Drug Destruction Created Date:", "").strip()
header_row = df[df[0] == "Medication ID Description"].index[0]
data = pd.read_excel(path, header=header_row)
for med_id in data["Medication ID"].dropna():
lookup[int(med_id)] = (basket_id, destroyed_on)
return lookup
# ── DB ────────────────────────────────────────────────────────────────────────
def get_conn():
return mysql.connector.connect(
host=db_config.DB_HOST, port=db_config.DB_PORT,
user=db_config.DB_USER, password=db_config.DB_PASSWORD,
database=db_config.DB_NAME,
)
def get_latest_import_id(cursor, study):
cursor.execute(
"SELECT MAX(import_id) AS mid FROM iwrs_import WHERE study=%s AND report_type='drugs'",
(study,),
)
row = cursor.fetchone()
mid = row["mid"]
if mid is None:
raise RuntimeError(f"Žádná data v MySQL pro studii {study}")
return mid
# ── Načítání dat z MySQL ──────────────────────────────────────────────────────
def load_inventory(cursor, study, import_id):
"""
Vrátí DataFrame s inventory + destruction join.
Sloupce jsou rovnou přejmenované pro downstream funkce.
"""
sql = """
SELECT
i.site AS Site,
i.medication_id AS `Med ID`,
i.packaged_lot_no AS `Lot No.`,
i.original_expiration_date AS `Orig Exp Date`,
i.expiration_date AS `Exp Date`,
i.received_date AS `Rcv Date`,
i.receipt_user AS `Rcpt User`,
i.subject_identifier AS `Subject ID`,
i.quantity_assigned AS `Qty Asgn`,
i.irt_transaction AS `IRT Tx`,
i.date_assigned AS `Date Asgn`,
i.assignment_user AS `Asgn User`,
i.dispensation_status AS `Disp Status`,
i.dispensing_date AS `Disp Date`,
i.quantity_dispensed AS `Qty Disp`,
i.dispensing_user AS `Disp User`,
i.quantity_returned AS `Qty Ret`,
i.date_returned AS `Date Ret`,
i.return_user AS `Ret User`,
d.destruction_date AS Destroyed,
d.basket_id AS `Basket No.`
FROM iwrs_inventory i
LEFT JOIN (
SELECT medication_id,
ANY_VALUE(basket_id) AS basket_id,
ANY_VALUE(destruction_date) AS destruction_date
FROM iwrs_destruction
WHERE study = %s
GROUP BY medication_id
) d ON d.medication_id = i.medication_id
WHERE i.import_id = %s
AND i.study = %s
ORDER BY i.site, i.received_date, i.medication_id
"""
cursor.execute(sql, (study, import_id, study))
rows = cursor.fetchall()
df = pd.DataFrame(rows)
for col in DATE_COLUMNS:
if col in df.columns:
df[col] = pd.to_datetime(df[col], errors="coerce")
print(f" Inventory: {len(df)} kitu")
return df
def load_shipments(cursor, study, import_id):
"""
Vrátí DataFrame se spojenými shipments + items.
"""
sql = """
SELECT
s.shipment_id AS `Shipment ID`,
s.status AS `IRT Shipment Status`,
s.type AS Type,
s.ship_from AS `Shipment From`,
s.ship_to_site AS `Ship To:`,
s.request_date AS `Request Date`,
s.received_date AS `Received Date`,
s.received_by AS `Received by`,
s.expected_arrival AS `Expected Arrival`,
i.investigator AS Investigator,
i.medication_description AS `Medication Description`,
i.medication_id AS `Medication ID`,
i.packaged_lot_no AS `Packaged Lot number`,
i.expiration_date AS `Expiration Date`,
i.item_status AS Status
FROM iwrs_shipments s
JOIN iwrs_shipment_items i
ON i.study = s.study
AND i.shipment_id = s.shipment_id
AND i.import_id = %s
WHERE s.import_id = %s
AND s.study = %s
ORDER BY s.ship_to_site, s.shipment_id, i.medication_id
"""
cursor.execute(sql, (import_id, import_id, study))
rows = cursor.fetchall()
df = pd.DataFrame(rows)
for col in ("Request Date", "Received Date", "Expiration Date", "Expected Arrival"):
if col in df.columns:
df[col] = pd.to_datetime(df[col], errors="coerce")
print(f" Shipments: {df['Shipment ID'].nunique() if len(df) else 0} zásilek, {len(df)} kitu")
return df
# ── Odvozené sheety ───────────────────────────────────────────────────────────
def build_site_summary(shipments_df):
STATUS_COLS = ["Available", "Assigned", "Dispensed", "Returned by Subject"]
pivot = shipments_df.groupby("Ship To:")["Status"].value_counts().unstack(fill_value=0)
for s in STATUS_COLS:
if s not in pivot.columns:
pivot[s] = 0
pivot = (
pivot[STATUS_COLS]
.reset_index()
.rename(columns={"Ship To:": "Site", "Returned by Subject": "Returned"})
.sort_values("Site")
.reset_index(drop=True)
)
pivot["Total"] = pivot[["Available", "Assigned", "Dispensed", "Returned"]].sum(axis=1)
print(f" Site Summary: {len(pivot)} center")
return pivot
def build_expired(df):
today = date.today()
mask = (
df["Basket No."].isna() &
df["Subject ID"].isna() &
(df["Exp Date"] < pd.Timestamp(today))
)
filtered = df[mask].copy().reset_index(drop=True)
sheet_name = f"Expired as of {today.strftime('%d-%b-%Y')}"
print(f" Expired: {len(filtered)}")
return filtered, sheet_name
def build_assigned_not_dispensed(df):
mask = df["Subject ID"].notna() & df["Disp Date"].isna()
filtered = df[mask].copy().reset_index(drop=True)
print(f" Assigned not dispensed: {len(filtered)}")
return filtered
def build_not_returned(df):
no_ret = df[
df["Date Ret"].isna() &
df["Subject ID"].notna() &
(df["Disp Status"].fillna("").str.upper() != "NOT DISPENSED")
].copy()
max_asgn = df.groupby("Subject ID")["Date Asgn"].max().rename("Max Visit Date")
no_ret = no_ret.join(max_asgn, on="Subject ID")
filtered = no_ret[no_ret["Date Asgn"] < no_ret["Max Visit Date"]].copy()
filtered = filtered.drop(columns=["Qty Ret", "Date Ret", "Ret User", "Destroyed", "Basket No."])
filtered = filtered.reset_index(drop=True)
print(f" Not returned: {len(filtered)}")
return filtered
def build_kits_for_destruction(df):
mask = (
df["Basket No."].isna() &
(df["Date Ret"].notna() | (df["Disp Status"].fillna("").str.upper() == "NOT DISPENSED"))
)
filtered = (
df[mask]
.copy()
.sort_values(["Site", "Date Ret"], ascending=[True, True])
.drop(columns=["Destroyed", "Basket No."])
.reset_index(drop=True)
)
print(f" Kits for destruction: {len(filtered)}")
return filtered
# ── Formátování ───────────────────────────────────────────────────────────────
def format_sheet(ws, header_color, highlight_col=None, highlight_color=None):
thin = Side(style="thin", color="000000")
@@ -155,62 +275,11 @@ def format_sheet(ws, header_color, highlight_col=None, highlight_color=None):
ws.freeze_panes = "A2"
# ── Shipment helpers ─────────────────────────────────────────────────────────
def build_shipments():
sh = pd.read_excel(SHIPMENTS_FILE, sheet_name=0, header=5)
sh.columns = sh.columns.str.strip()
sh = sh.dropna(how="all")
sh["Shipment ID"] = sh["Shipment ID"].astype(str).str.strip()
sh = sh.drop(columns=[c for c in SHIPMENT_DROP_COLS if c in sh.columns])
shipment_cols = list(sh.columns)
all_rows = []
for _, s_row in sh.iterrows():
sid = s_row["Shipment ID"]
path = DETAILS_DIR / f"shipment_details_{sid}.xlsx"
if not path.exists():
continue
det = pd.read_excel(path, sheet_name=0, header=5)
det.columns = det.columns.str.strip()
det = det.dropna(how="all")
det["Shipment"] = det["Shipment"].astype(str).str.strip()
extra_cols = [c for c in det.columns if c not in shipment_cols and c != "Shipment" and c not in SHIPMENT_DROP_COLS]
for _, d_row in det.iterrows():
all_rows.append({**s_row.to_dict(), **{c: d_row[c] for c in extra_cols}})
result = pd.DataFrame(all_rows)
all_cols = shipment_cols + [c for c in extra_cols if c in result.columns]
result = result[all_cols]
for col in ["Request Date", "Received Date", "Expiration Date"]:
if col in result.columns:
result[col] = pd.to_datetime(result[col], errors="coerce")
print(f" Shipments: {result['Shipment ID'].nunique()} shipments, {len(result)} kitu")
return result
def build_site_summary(result):
STATUS_COLS = ["Available", "Assigned", "Dispensed", "Returned by Subject"]
pivot = result.groupby("Ship To:")["Status"].value_counts().unstack(fill_value=0)
for s in STATUS_COLS:
if s not in pivot.columns:
pivot[s] = 0
pivot = pivot[STATUS_COLS].reset_index().rename(columns={
"Ship To:": "Site", "Returned by Subject": "Returned"
})
pivot = pivot.sort_values("Site").reset_index(drop=True)
pivot["Total"] = pivot[["Available", "Assigned", "Dispensed", "Returned"]].sum(axis=1)
print(f" Site Summary: {len(pivot)} center")
return pivot
def format_shipment_sheet(ws, header_color_ship, header_color_detail, n_ship_cols):
thin = Side(style="thin", color="000000")
border = Border(left=thin, right=thin, top=thin, bottom=thin)
hfont = Font(bold=True, color="FFFFFF", name="Arial", size=10)
dfont = Font(name="Arial", size=10)
thin = Side(style="thin", color="000000")
border = Border(left=thin, right=thin, top=thin, bottom=thin)
hfont = Font(bold=True, color="FFFFFF", name="Arial", size=10)
dfont = Font(name="Arial", size=10)
fill_ship = PatternFill("solid", start_color=header_color_ship)
fill_detail = PatternFill("solid", start_color=header_color_detail)
@@ -219,7 +288,9 @@ def format_shipment_sheet(ws, header_color_ship, header_color_detail, n_ship_col
cell.font = hfont
cell.alignment = Alignment(horizontal="center", vertical="center", wrap_text=True)
cell.border = border
ws.column_dimensions[get_column_letter(cell.column)].width = min(len(str(cell.value or "")) + 4, 35)
ws.column_dimensions[get_column_letter(cell.column)].width = min(
len(str(cell.value or "")) + 4, 35
)
ws.row_dimensions[1].height = 30
for row in ws.iter_rows(min_row=2, max_row=ws.max_row):
@@ -234,101 +305,29 @@ def format_shipment_sheet(ws, header_color_ship, header_color_detail, n_ship_col
ws.freeze_panes = "A2"
# ── Build DataFrames ──────────────────────────────────────────────────────────
def build_main(lookup):
all_rows = []
for path in sorted(INVENTORY_DIR.glob("onsite_inventory_detail_*.xlsx")):
df, meta = read_inventory(path)
if df is None:
continue
df["DestroyedOn"] = df["Medication ID"].apply(
lambda x: lookup.get(int(x), (None, None))[1] if pd.notna(x) else None)
df["Basket number"] = df["Medication ID"].apply(
lambda x: lookup.get(int(x), (None, None))[0] if pd.notna(x) else None)
df.insert(0, "Site", meta.get("site", path.stem))
all_rows.append(df)
print(f" {path.name}: {len(df)} kits")
combined = pd.concat(all_rows, ignore_index=True)
combined.rename(columns=COLUMN_RENAMES, inplace=True)
for col in DATE_COLUMNS:
if col in combined.columns:
combined[col] = pd.to_datetime(combined[col], dayfirst=True, errors="coerce")
combined.sort_values(["Site", "Rcv Date", "Med ID"], inplace=True, ignore_index=True)
return combined
def build_expired(df):
today = date.today()
mask = (
df["Basket No."].isna() &
df["Subject ID"].isna() &
(df["Exp Date"] < pd.Timestamp(today))
)
filtered = df[mask].copy().reset_index(drop=True)
sheet_name = f"Expired as of {today.strftime('%d-%b-%Y')}"
print(f" Expired: {len(filtered)}")
return filtered, sheet_name
def build_assigned_not_dispensed(df):
mask = df["Subject ID"].notna() & df["Disp Date"].isna()
filtered = df[mask].copy().reset_index(drop=True)
print(f" Assigned not dispensed: {len(filtered)}")
return filtered
def build_not_returned(df):
no_ret = df[
df["Date Ret"].isna() &
df["Subject ID"].notna() &
(df["Disp Status"].str.upper() != "NOT DISPENSED")
].copy()
max_asgn = df.groupby("Subject ID")["Date Asgn"].max().rename("Max Visit Date")
no_ret = no_ret.join(max_asgn, on="Subject ID")
filtered = no_ret[no_ret["Date Asgn"] < no_ret["Max Visit Date"]].copy()
filtered = filtered.drop(columns=["Qty Ret", "Date Ret", "Ret User", "Destroyed", "Basket No."])
filtered = filtered.reset_index(drop=True)
print(f" Not returned: {len(filtered)}")
return filtered
def build_kits_for_destruction(df):
mask = (
df["Basket No."].isna() &
(df["Date Ret"].notna() | (df["Disp Status"].str.upper() == "NOT DISPENSED"))
)
filtered = df[mask].copy().sort_values(["Site", "Date Ret"], ascending=[True, True])
filtered = filtered.drop(columns=["Destroyed", "Basket No."]).reset_index(drop=True)
print(f" Kits for destruction: {len(filtered)}")
return filtered
# ── Main ──────────────────────────────────────────────────────────────────────
def main():
# Prepare output dir, remove any previous overview file
OUTPUT_DIR.mkdir(exist_ok=True)
for old in OUTPUT_DIR.glob(f"*{STUDY} CZ IWRS overview.xlsx"):
old.unlink()
print(f"Removed old file: {old.name}")
lookup = read_destruction_lookup()
print(f"Loaded {len(lookup)} kits from destruction reports")
print(f"\nNačítám data z MySQL pro {STUDY}...")
conn = get_conn()
cursor = conn.cursor(dictionary=True)
import_id = get_latest_import_id(cursor, STUDY)
print(f" import_id = {import_id}")
df = build_main(lookup)
df = load_inventory(cursor, STUDY, import_id)
shipments_df = load_shipments(cursor, STUDY, import_id)
cursor.close()
conn.close()
expired_df, expired_sheet = build_expired(df)
assigned_df = build_assigned_not_dispensed(df)
not_returned_df = build_not_returned(df)
destruction_df = build_kits_for_destruction(df)
site_summary_df = build_site_summary(shipments_df)
shipments_df = build_shipments()
site_summary_df = build_site_summary(shipments_df)
n_ship_cols = shipments_df.columns.tolist().index("Investigator") # first detail col index (0-based)
# Write all sheets
with pd.ExcelWriter(OUTPUT_FILE, engine="openpyxl") as writer:
df.to_excel( writer, index=False, sheet_name="CountryMedicationOverview")
expired_df.to_excel( writer, index=False, sheet_name=expired_sheet)
@@ -338,30 +337,27 @@ def main():
shipments_df.to_excel( writer, index=False, sheet_name="Shipments")
site_summary_df.to_excel( writer, index=False, sheet_name="Site Summary")
# Format all sheets
wb = load_workbook(OUTPUT_FILE)
# Main sheet — dark blue, green highlight for Destroyed/Basket No.
ws_main = wb["CountryMedicationOverview"]
ws_main = wb["CountryMedicationOverview"]
format_sheet(ws_main, header_color="1F4E79")
# Extra: green fill for Destroyed and Basket No. columns
new_col_fill = PatternFill("solid", start_color="E2EFDA")
headers_main = [c.value for c in ws_main[1]]
new_col_fill = PatternFill("solid", start_color="E2EFDA")
headers_main = [c.value for c in ws_main[1]]
for row in ws_main.iter_rows(min_row=2, max_row=ws_main.max_row):
for cell in row:
col_name = headers_main[cell.column - 1] if cell.column <= len(headers_main) else None
if col_name in ("Destroyed", "Basket No."):
cell.fill = new_col_fill
format_sheet(wb[expired_sheet], header_color="C00000", highlight_col="Exp Date", highlight_color="FFE0E0")
format_sheet(wb["Assigned not dispensed"], header_color="833C00", highlight_col="Subject ID", highlight_color="FFF2CC")
format_sheet(wb["Not returned"], header_color="375623", highlight_col="Max Visit Date", highlight_color="E2EFDA")
format_sheet(wb["Kits for destruction"], header_color="595959")
format_shipment_sheet(wb["Shipments"], "1F4E79", "375623", n_ship_cols)
format_sheet(wb["Site Summary"], header_color="1F4E79")
format_sheet(wb[expired_sheet], header_color="C00000", highlight_col="Exp Date", highlight_color="FFE0E0")
format_sheet(wb["Assigned not dispensed"], header_color="833C00", highlight_col="Subject ID", highlight_color="FFF2CC")
format_sheet(wb["Not returned"], header_color="375623", highlight_col="Max Visit Date", highlight_color="E2EFDA")
format_sheet(wb["Kits for destruction"], header_color="595959")
format_shipment_sheet(wb["Shipments"], "1F4E79", "375623", N_SHIP_COLS)
format_sheet(wb["Site Summary"], header_color="1F4E79")
wb.save(OUTPUT_FILE)
print(f"\nSaved: {OUTPUT_FILE} ({len(df)} rows on main sheet, {wb.sheetnames})")
print(f"\nUloženo: {OUTPUT_FILE} ({len(df)} řádků, sheety: {wb.sheetnames})")
if __name__ == "__main__":