Files
cosec_automation/scripts/common.py
T

749 lines
28 KiB
Python

"""
AUTHOR:
Khushal P Soonderji
DATE:
CREATED: Thu, 5th Feb, 2026
UPDATED: Thu, 5th Feb, 2026
OBJECTIVE:
There will be some common actions across various scripts. This script holds those common actions.
REFERENCES:
N/A
DOWNLOADS:
N/A
"""
# *****************************************************************************************************************
# ***** ****
# *** IMPORT ***
# ***** ****
# *****************************************************************************************************************
# To make sibling directories accessible for imports:
import sys
sys.path.append(".")
sys.path.append("..")
# For system-level activities:
import os
import copy
# To work with date and time:
import time
import datetime
# To work with tabulate data:
import pandas as pd
# Cosec-related:
from cosec_web.cosec_web import CosecWeb
# TCAOFF-related:
from helpers.async_tcaoff import AsyncTheCAOffice
# My utils:
from utils_v2.system import files
from utils_v2.string import json
from utils_v2.string import regex
from utils_v2.date_time import date_time
# To work with datatypes:
from typing import List, Dict, Any, Union
from collections import defaultdict
# To run a cron-like scheduler:
from apscheduler.schedulers.asyncio import AsyncIOScheduler
# For async activities:
import asyncio
# For debugging:
from icecream import IceCreamDebugger
# *****************************************************************************************************************
# ***** ****
# *** MACROS / ONE-TIME INIT ***
# ***** ****
# *****************************************************************************************************************
# File paths:
FILE_DIR = files.get_file_directory(include_filename = False)
PROJ_DIR = files.get_parent_directory(FILE_DIR, depth = 1)
CACHE_DIR = os.path.join(PROJ_DIR, "local", "cache")
CREDS_DIR = os.path.join(PROJ_DIR, "creds")
# ---
COSEC_CREDS_FILE = os.path.join(CREDS_DIR, "cosec.json")
TCAOFF_CREDS_FILE = os.path.join(CREDS_DIR, "tcaoff.json")
MUSTER_ROLL_CACHE_FILE = os.path.join(CACHE_DIR, "muster_roll_cache.json")
IN_OUT_SUMMARY_CACHE_FILE = os.path.join(CACHE_DIR, "in_out_summary_cache.json")
PREV_DAY_IN_OUT_SUMMARY_CACHE_FILE = os.path.join(CACHE_DIR, "prev_day_in_out_summary_cache.json")
CHROME_DRIVER_DIR = os.path.join(PROJ_DIR, "drivers", "chrome")
USER_DATA_DIR = os.path.join(PROJ_DIR, "browser", "user_data")
DOWNLOADS_DIR = os.path.join(PROJ_DIR, "downloads")
# ---
TEST_MODE_MUSTER_ROLL_FILE_PATH = os.path.join(PROJ_DIR, "cosec_web", "sample_files", "muster_roll.xls")
TEST_MODE_IN_OUT_SUMMARY_FILE_PATH = os.path.join(PROJ_DIR, "cosec_web", "sample_files", "in_out_summary.xls")
# Debugging:
printer = IceCreamDebugger(prefix = "Common | ", includeContext = True)
err_printer = IceCreamDebugger(prefix = "[ERR] Common | ", includeContext = True)
# *****************************************************************************************************************
# ***** ****
# *** VARIABLES ***
# ***** ****
# *****************************************************************************************************************
# --- Nothing Yet
# *****************************************************************************************************************
# ***** ****
# *** CLASSES ***
# ***** ****
# *****************************************************************************************************************
# --- Nothing Yet
# *****************************************************************************************************************
# ***** ****
# *** FUNCTIONS ***
# ***** ****
# *****************************************************************************************************************
def kill_chrome() -> None:
"""
To kill running Chrome processes so that they don't interfere with the one that will be spun up by automation code.
:return: None
"""
# First kill the previous processes,
# then wait if old processes were killed:
kill_count = CosecWeb.kill_chrome_processes()
if kill_count > 0: time.sleep(2.5)
# ---------------------------------------------------------------------------------------------------------------------
def get_muster_roll(
cosec_creds: dict,
on_date: datetime.datetime,
cache_file: str = MUSTER_ROLL_CACHE_FILE,
test_mode: bool = False
) -> bool:
"""
Get the latest Muster-Roll from Matrix Cosec. It saves the data into a local cache file.
:param cosec_creds: The credentials (and config) to operate Cosec Matrix.
:param on_date: The date on which you want .
:param cache_file: The cache file to store the results in.
:param test_mode: If set to True, a past file will be used instead of getting new reports from Cosec.
:return: True if the automated fetch was successful, else False.
"""
# Start by assuming failure:
success = False
report_path = None
# If test mode:
if test_mode:
report_path = TEST_MODE_MUSTER_ROLL_FILE_PATH
# When not working in test mode:
else:
# Force close other running Chrome processes:
kill_chrome()
# Create an instance of the automation object:
cosec = CosecWeb(
cosec_url = cosec_creds["creds"]["url"],
username = cosec_creds["creds"]["username"],
password = cosec_creds["creds"]["password"],
driver_dir = CHROME_DRIVER_DIR,
user_data_dir = USER_DATA_DIR,
downloads_dir = DOWNLOADS_DIR,
)
# Perform the login:
cosec.login(initial_sleep = 2.5)
# Get the in/out report:
report_path = cosec.get_muster_roll(
initial_sleep = 1.0,
on_date = on_date,
group_ids = cosec_creds["musterRollConfig"]["groupIds"],
download_timeout = 60.0
)
# Log out to end the cycle:
cosec.logout()
# Close the browser window:
cosec.quit()
# Now process the report,
# and save it to the JSON file:
if report_path is not None:
# Read the data:
report_data = CosecWeb.read_muster_roll_xls(report_path)
# Do the remaining cleanup and formatting:
report_data = report_data[[
"User ID", "User Name", "Category Name",
"Grade Name", "Branch Name", "Department Name",
"Direct Reporting", "Level-1"
]]
report_data = report_data.where(report_data.notna(), None)
report_data = report_data.to_dict(orient = "records")
# Loop through the report for inferred data.
# Add the ids of the "Direct Reporting" and "Level-1" values:
for person in report_data:
person["Direct Reporting ID"] = None
person["Level-1 ID"] = None
for check_against in report_data:
if person["Direct Reporting"] == check_against["User Name"]:
person["Direct Reporting ID"] = check_against["User ID"]
if person["Level-1"] == check_against["User Name"]:
person["Level-1 ID"] = check_against["User ID"]
# Sort by hierarchy (BFS):
pass
# print(f"MUSTER ROLL ({len(report_data)}):", json.to_string(report_data))
# Save the data to a JSON file:
json.to_file(
file = cache_file,
python_data = {
"ts": date_time.get_current_utc_date_time(as_string = False).timestamp(),
"report": report_data
},
no_space = True
)
# Note down success:
success = True
# Done here:
return success
# ---------------------------------------------------------------------------------------------------------------------
async def sync_branches_to_tcaoff(
tcaoff_client: AsyncTheCAOffice,
cosec_muster_roll: dict,
) -> Dict[str, int]:
# Start with a basic response structure:
response = defaultdict(int)
# Get the list of existing branches from TCAOFF:
tcaoff_branches = await tcaoff_client.branch_list()
tcaoff_branches = [_["branch_name"].lower() for _ in tcaoff_branches]
tcaoff_branches = list(set(tcaoff_branches))
print("TCAOFF BRANCHES: ", tcaoff_branches)
# Get only the branch names from Cosec:
cosec_branches = [tcaoff_client.remove_special_chars(_["Branch Name"]) for _ in cosec_muster_roll["report"]]
cosec_branches = list(set(cosec_branches))
print("COSEC BRANCHES: ", cosec_branches)
# Loop through the data from Cosec and add missing branches to TCAOFF:
for cosec_branch in cosec_branches:
if cosec_branch.lower() not in tcaoff_branches:
success = await tcaoff_client.branch_add(branch_name = cosec_branch)
response["total"] += 1
if success: response["success"] += 1
else: response["fail"] += 1
# Done here:
return response
# ---------------------------------------------------------------------------------------------------------------------
async def sync_departments_to_tcaoff(
tcaoff_client: AsyncTheCAOffice,
cosec_muster_roll: dict,
) -> Dict[str, int]:
# Start with a basic response structure:
response = defaultdict(int)
# Get the list of existing departments from TCAOFF:
tcaoff_depts = await tcaoff_client.department_list()
tcaoff_depts = [_["department_name"].lower() for _ in tcaoff_depts]
tcaoff_depts = list(set(tcaoff_depts))
# Get only the department names from Cosec:
cosec_depts = [tcaoff_client.remove_special_chars(_["Department Name"]) for _ in cosec_muster_roll["report"]]
cosec_depts = list(set(cosec_depts))
# Loop through the data from Cosec and add missing departments to TCAOFF:
for cosec_dept in cosec_depts:
if cosec_dept.lower() not in tcaoff_depts:
success = tcaoff_client.department_add(department_name = cosec_dept)
response["total"] += 1
if success: response["success"] += 1
else: response["fail"] += 1
# Done here:
return response
# ---------------------------------------------------------------------------------------------------------------------
async def sync_teams_to_tcaoff(
tcaoff_client: AsyncTheCAOffice,
cosec_muster_roll: dict,
) -> Dict[str, int]:
# Start with a basic response structure:
response = defaultdict(int)
# Map out branch ids:
tcaoff_branches = await tcaoff_client.branch_list()
tcaoff_branches_lookup = {d["branch_name"].lower(): d["branch_id"] for d in tcaoff_branches}
# Map out dept. ids:
tcaoff_depts = await tcaoff_client.department_list()
tcaoff_depts_lookup = {d["department_name"].lower(): d["department_id"] for d in tcaoff_depts}
# Get the list of existing team members from TCAOFF:
# NOTE: `pseudonym` is the unique username of the user.
tcaoff_teams = await tcaoff_client.team_list()
# # OLD WAY - Based on human-readable user names:
# # Identify which users have not already been sync'd:
# synced_team_ids = []
# for t in tcaoff_teams:
# app_notes = json.from_string(t["json_notes"]).get("applicantNotes", {})
# if isinstance(app_notes, str): app_notes = json.from_string(app_notes)
# if app_notes is None:
# app_notes = {}
# cosec_notes = app_notes.get("cosec", {})
# # print("COSEC NOTES:", cosec_notes)
# if cosec_notes:
# user_id = cosec_notes.get("User ID")
# if not user_id: user_id = cosec_notes["UserID"]
# synced_team_ids.append(user_id)
# tcaoff_teams = [_["pseudonym"] for _ in tcaoff_teams]
# tcaoff_teams = list(set(tcaoff_teams))
# NEW WAY - Based on unique Employee Ids:
# Identify which users have not already been sync'd:
synced_team_ids = [_["pseudonym"] for _ in tcaoff_teams]
tcaoff_teams = [_["pseudonym"] for _ in tcaoff_teams]
tcaoff_teams = list(set(tcaoff_teams))
# Loop through the data from Cosec and add missing team-members to TCAOFF:
for cosec_team in cosec_muster_roll["report"]:
# Extract Cosec Details:
cosec_user_id = cosec_team["User ID"].strip()
cosec_user_name = cosec_team["User Name"].strip()
cosec_branch_name = tcaoff_client.remove_special_chars(cosec_team["Branch Name"]).lower()
cosec_dept_name = tcaoff_client.remove_special_chars(cosec_team["Department Name"]).lower()
cosec_grade_name = (cosec_team.get("Grade Name") or "Unknown").strip()
if cosec_user_id not in synced_team_ids:
# Count the user:
response["total"] += 1
# Check the branch id and department id:
branch_id = tcaoff_branches_lookup.get(cosec_branch_name)
dept_id = tcaoff_depts_lookup.get(cosec_dept_name)
if branch_id is None:
err_printer("TEAM SYNC ERR", "Branch Not Found in TCAOFF", cosec_branch_name)
response["fail"] += 1
continue
if dept_id is None:
err_printer("TEAM SYNC ERR", "Dept. Not Found in TCAOFF", cosec_dept_name)
response["fail"] += 1
continue
# Add the team:
tcaoff_username = regex.replace(
text = cosec_user_name,
pattern = r"[^\w\d\._]",
substitute_text = ""
).strip().lower()
tcaoff_email_id = tcaoff_username + "@velankanigroup.com"
# print("Need to Add:", json.to_string(team_json))
success = await tcaoff_client.team_add(
branch_id = branch_id,
dept_id = dept_id,
reporting_to = None,
team_name = cosec_user_name,
email = tcaoff_email_id,
phone_no = "9876543210",
role = cosec_grade_name,
username = cosec_user_id,
password = "Vispl@123",
applicant_notes = {"cosec": cosec_team}
)
if success: response["success"] += 1
else:
err_printer("COULDN'T ADD", cosec_team)
response["fail"] += 1
# Done here:
printer("TCAOFF-Cosec Team Sync.:", json.to_string(response))
return response
# ---------------------------------------------------------------------------------------------------------------------
def get_in_out_summary(
cosec_creds: dict,
from_dt: date_time.datetime = None,
to_dt: date_time.datetime = None,
cache_file: str = IN_OUT_SUMMARY_CACHE_FILE,
test_mode: bool = False
) -> bool:
"""
Get the latest In-Out-Summary from Matrix Cosec. It saves the data into a local cache file.
:param cosec_creds: The credentials (and config) to operate Cosec Matrix.
:param from_dt: The date from which to fetch In-Out-Summary.
:param to_dt: The date till which to fetch In-Out-Summary.
:param cache_file: The cache file to use to store the results.
:param test_mode: If set to True, a past file will be used instead of getting new reports from Cosec.
:return: True if the automated fetch was successful, else False.
"""
# Start by assuming failure:
success = False
report_path = None
# If test mode:
if test_mode:
report_path = TEST_MODE_IN_OUT_SUMMARY_FILE_PATH
# When not working in test mode:
else:
# Force close other running Chrome processes:
kill_chrome()
# Create an instance of the automation object:
cosec = CosecWeb(
cosec_url = cosec_creds["creds"]["url"],
username = cosec_creds["creds"]["username"],
password = cosec_creds["creds"]["password"],
driver_dir = CHROME_DRIVER_DIR,
user_data_dir = USER_DATA_DIR,
downloads_dir = DOWNLOADS_DIR,
)
# Perform the login:
cosec.login(initial_sleep = 2.5)
# Get the in/out report:
if to_dt is None: to_dt = date_time.get_current_ist_date_time()
if from_dt is None: from_dt = to_dt - datetime.timedelta(days = 1)
report_path = cosec.get_in_out_summary(
initial_sleep = 1.0,
from_date = from_dt,
to_date = to_dt,
group_ids = cosec_creds["inOutConfig"]["groupIds"],
download_timeout = 60.0,
timezone = cosec_creds["generalConfig"]["timezone"],
)
# Log out to end the cycle:
cosec.logout()
# Close the browser window:
cosec.quit()
# Now process the report,
# and save it to the JSON file:
if report_path is not None:
# Read the data:
report_data = CosecWeb.read_in_out_summary_xls(report_path)
# Assume that the punch time in the data is IST data,
# then normalize it to UTC:
def parse_dt(x):
if pd.isnull(x): return None
else: return date_time.to_timezone(
datetime_object = date_time.as_if_timezone(
datetime_object = date_time.parse_date_time(x),
timezone = cosec_creds["generalConfig"]["timezone"]
),
timezone = date_time.TIMEZONE_UTC
).timestamp()
report_data["Punch Time"] = report_data["Punch Time"].apply(lambda x: parse_dt(x))
# Do the remaining cleanup and formatting:
report_data = report_data.where(report_data.notna(), None)
report_data = report_data.to_dict(orient = "records")
report_data = {
"ts": date_time.get_current_utc_date_time(as_string = False).timestamp(),
"report": report_data
}
# Save the data to a JSON file:
json.to_file(
file = cache_file,
python_data = report_data,
no_space = True
)
# Note down success:
success = True
# Done here:
return success
# ---------------------------------------------------------------------------------------------------------------------
def compute_work_done(
in_out_df: pd.DataFrame
) -> List[Dict[str, Union[str, int, float, None]]]:
"""
To calculate the full work done by all the employees on all the provided dates.
:param in_out_df: The table that contains all the work done by all the employees on the date-range that was
selected.
:return: A list of dicts that contain the information of all the work done.
"""
# Create the structure that will be given as the output:
flattened_work_reports = []
# Create an internal dict with the structure:
# work_reports["user"]["date"] = {...}
work_reports = defaultdict(lambda: defaultdict(dict))
# Loop through the full DataFrame once,
# and figure out the first in and last out times:
for index, row in in_out_df.iterrows():
# Ignore if the event wasn't a success:
if row["Event Status"] != "Allowed":
continue
# Extract user and date:
user_id = row["User ID"]
punch_dt = date_time.parse_date_time(row["Punch Time"], timezone = date_time.TIMEZONE_IST)
punch_date = punch_dt.strftime("%Y-%m-%d")
punch_ts = punch_dt.timestamp()
punch_loc = row[" Device/Source Detail"]
# Handle the first in time:
if row["I/O Type"] == "In":
if work_reports[user_id][punch_date].get("first_in") is None:
work_reports[user_id][punch_date]["first_in"] = punch_ts
work_reports[user_id][punch_date]["first_in_loc"] = punch_loc
# Handle the last out time:
if row["I/O Type"] == "Out":
if work_reports[user_id][punch_date].get("first_in") is not None:
work_reports[user_id][punch_date]["last_out"] = punch_ts
work_reports[user_id][punch_date]["last_out_loc"] = punch_loc
# Now use the first_in and last out times of each record to figure out the amount of work done:
for user_id, user_reports in work_reports.items():
for punch_date, punch_info in user_reports.items():
# Some defaults:
min_work_seconds = 10.0 * 60.0 * 60.0
work_seconds = 0.0
work_ot_seconds = 0.0
work_status = "A"
# Extract, clean and compute punch timing:
first_in = punch_info.get("first_in")
first_in_loc = punch_info.get("first_in_loc")
last_out = punch_info.get("last_out")
last_out_loc = punch_info.get("last_out_loc")
# When the user has a valid in-time, but no known out time,
# we assume that he worked a full day:
if first_in is not None and not last_out:
last_out = first_in + min_work_seconds
work_seconds = last_out - first_in
work_ot_seconds = 0.0
# When the user has neither an in-time, nor an out-time,
# we assume that he was absent the whole day:
elif not first_in and not last_out:
work_seconds = 0.0
work_ot_seconds = 0.0
# When we have both - an in-time and an out-time - we compute the work done.
# Any work over 10 hours will be counted as overtime:
elif first_in and last_out:
work_seconds = last_out - first_in
work_ot_seconds = max(0.0, work_seconds - min_work_seconds)
# Finally, compute the work status.
# 'A' --> Absent
# 'H' --> Half Day
# 'P' --> Present (Full Day)
# 'OT' -> Overtime
work_hours = work_seconds / (60 * 60)
if work_hours > 10.0: work_status = "OT"
elif 7.5 < work_hours <= 10.0: work_status = "P"
elif 2.5 < work_hours <= 5.0: work_status = "H"
else: work_status = "A"
# Save the data:
flattened_work_reports.append({
"user_id": user_id,
"work_date": punch_date,
"first_in": first_in,
"first_in_loc": first_in_loc,
"last_out": last_out,
"last_out_loc": last_out_loc,
"work_seconds": work_seconds,
"work_hours": work_hours,
"work_ot_seconds": work_ot_seconds,
"work_ot_hours": work_ot_seconds / (60.0 * 60.0),
"work_status": work_status,
})
# Done here:
return flattened_work_reports
# ---------------------------------------------------------------------------------------------------------------------
async def sync_attendance_to_tcaoff(
tcaoff_client: AsyncTheCAOffice,
cosec_in_out_summary: dict,
verbose: bool = False
) -> Dict[str, int]:
"""
:param tcaoff_client: The asynchronous client object that interfaces with TCAOFF.
:param cosec_in_out_summary: he table that contains all the work done by all the employees on the date-range that
was selected.
:param verbose: If True, internal debugging print will be more verbose.
:return: The dict that gives you the count of the successful and failed attendance marking API calls.
"""
# Results:
results = defaultdict(int)
# Convert the In/Out data to a DataFrame:
in_out_df = pd.DataFrame(cosec_in_out_summary["report"])
# Compute the work done:
work_reports = compute_work_done(in_out_df)
# Get the list of existing team members from TCAOFF:
# NOTE: `pseudonym` is the unique username of the user.
tcaoff_teams = await tcaoff_client.team_list()
# Get a mapping from Cosec id to TCAOFF record:
cosec_id_to_tcaoff_team = {}
for t in tcaoff_teams:
app_notes = json.from_string(t["json_notes"]).get("applicantNotes", {})
if isinstance(app_notes, str): app_notes = json.from_string(app_notes)
if not app_notes: continue
cosec_notes = app_notes.get("cosec", {})
if not cosec_notes: continue
# print("Cosec Notes:", cosec_notes)
cosec_id_to_tcaoff_team[cosec_notes.get("User ID") or cosec_notes.get("UserID")] = t
cosec_id_to_tcaoff_team = {k:v for k, v in cosec_id_to_tcaoff_team.items() if k == v["pseudonym"]}
# Create the tasks to fire:
tasks = []
for wr in work_reports:
# Extract basic details:
cosec_user_id = wr["user_id"]
# Match it to the TCAOFF team-member:
tcaoff_team = cosec_id_to_tcaoff_team.get(cosec_user_id)
if tcaoff_team is None:
printer(f"TCAOFF SYNC ERR: Cosesc User Id not found in TCAOFF", cosec_user_id)
continue
# Create the task for this attendance:
tasks.append(
tcaoff_client.attendance_mark(
user_id = tcaoff_team["user_id"],
status = wr["work_status"],
over_time = wr["work_ot_hours"],
attendance_date = wr["work_date"],
json_notes = {
"totHours": wr["work_hours"],
"firstIn": wr["first_in"],
"lastOut": wr["last_out"],
}
)
)
printer("TASKS COUNT", len(tasks))
def chunks(lst, size = 10):
for i in range(0, len(lst), size):
yield lst[i:i + size]
count = 0
for chunk in chunks(tasks[:], size = 25):
count += 1
printer("Task Chunk:", count)
_res = await asyncio.gather(*chunk)
success = sum(_res)
failure = len(_res) - success
results["success"] += success
results["failure"] += failure
results["total"] += len(_res)
# Done here:
printer(results)
return results
# *****************************************************************************************************************
# ***** ****
# *** MAIN PROGRAM ***
# ***** ****
# *****************************************************************************************************************
if __name__ == "__main__":
pass