222 lines
No EOL
7.2 KiB
Python
222 lines
No EOL
7.2 KiB
Python
# --- install (if needed) ---
|
|
# !pip install pdfplumber pandas openpyxl
|
|
|
|
import re
|
|
import os
|
|
import pandas as pd
|
|
import pdfplumber
|
|
import openpyxl
|
|
from openpyxl.styles import Font, Alignment
|
|
from openpyxl.worksheet.table import Table, TableStyleInfo
|
|
|
|
# -----------------------------
|
|
# Paths
|
|
# -----------------------------
|
|
pdf_path = r"C:\Users\jmulhall\Downloads\DTB_janfeb2026.pdf"
|
|
out_xlsx = "meditech_trial_balance_parsed_0630.xlsx"
|
|
|
|
# -----------------------------
|
|
# Regex
|
|
# -----------------------------
|
|
money_re = re.compile(r"-?\d{1,3}(?:,\d{3})*(?:\.\d{2})")
|
|
acct_re = re.compile(r"^(?P<account>\d{2}\.\d{5}\.\d{5})\s+-\s+(?P<fund>\d{2})\s+(?P<name>.+)$")
|
|
journal_re = re.compile(
|
|
r"^(?P<journal>[A-Z][A-Z0-9/]*)\s+"
|
|
r"(?P<date>\d{2}/\d{2}/\d{2})(?P<bch>\d{1,2})?\s+"
|
|
r"(?P<entry>\d+(?:-\d+)?)\b"
|
|
)
|
|
|
|
# -----------------------------
|
|
# Helpers
|
|
# -----------------------------
|
|
def parse_money(s):
|
|
if not s:
|
|
return None
|
|
m = money_re.search(s)
|
|
return float(m.group(0).replace(",", "")) if m else None
|
|
|
|
def is_page_header_noise(s):
|
|
s = " ".join(s.split()).upper()
|
|
return (
|
|
("FISCAL CALENDAR" in s or "FIRST CALENDAR" in s)
|
|
and ("ROUND MONEY" in s or "TRIAL" in s)
|
|
)
|
|
|
|
def clean_description(desc):
|
|
if not desc or is_page_header_noise(desc):
|
|
return ""
|
|
patterns = [
|
|
r"FISCAL\s+CALENDAR\s+[A-Z]{3,6}\s*,?\s*[A-Z]{3}\s+20\d{2}",
|
|
r"FIRST\s+CALENDAR\s+[A-Z]{3,6}\s*,?\s*[A-Z]{3}\s+20\d{2}",
|
|
r"ROUND\s+MONEY\s*:\s*\d+\.\d+",
|
|
r"\bTRIAL\b",
|
|
]
|
|
for p in patterns:
|
|
desc = re.sub(p, "", desc, flags=re.IGNORECASE)
|
|
return " ".join(desc.split())
|
|
|
|
# -----------------------------
|
|
# Parse PDF
|
|
# -----------------------------
|
|
transactions = []
|
|
acct_summaries = []
|
|
report_title_parts = {"fiscal": None, "period": None, "trial": None}
|
|
|
|
with pdfplumber.open(pdf_path) as pdf:
|
|
for page in pdf.pages:
|
|
text = page.extract_text(layout=True) or ""
|
|
lines = text.splitlines()
|
|
|
|
# Capture report title once
|
|
for ln in lines:
|
|
s = ln.strip()
|
|
if report_title_parts["fiscal"] is None and "Fiscal Calendar" in s:
|
|
report_title_parts["fiscal"] = " ".join(s.split())
|
|
if report_title_parts["period"] is None and re.search(r"\b[A-Z]{3}\s+20\d{2}\b", s):
|
|
report_title_parts["period"] = " ".join(s.split())
|
|
if report_title_parts["trial"] is None and s.upper() == "TRIAL":
|
|
report_title_parts["trial"] = "TRIAL"
|
|
|
|
# Locate headers
|
|
txn_header = acct_header = None
|
|
for i, ln in enumerate(lines):
|
|
if "JOURNAL" in ln and "DESCRIPTION" in ln and "DEBITS" in ln:
|
|
txn_header = i
|
|
if "ACCOUNT" in ln and "OPEN" in ln and "CLOSE" in ln:
|
|
acct_header = i
|
|
|
|
deb_pos = cred_pos = desc_pos = None
|
|
open_pos = adeb_pos = acredit_pos = change_pos = close_pos = None
|
|
|
|
if txn_header is not None:
|
|
h = lines[txn_header]
|
|
deb_pos = h.find("DEBITS")
|
|
cred_pos = h.find("CREDITS", deb_pos)
|
|
desc_pos = h.find("DESCRIPTION", cred_pos)
|
|
|
|
if acct_header is not None:
|
|
h = lines[acct_header]
|
|
open_pos = h.find("OPEN")
|
|
adeb_pos = h.find("DEBITS", open_pos)
|
|
acredit_pos = h.find("CREDITS", adeb_pos)
|
|
change_pos = h.find("CHANGE", acredit_pos)
|
|
close_pos = h.find("CLOSE", change_pos)
|
|
|
|
current_account = None
|
|
current_account_name = None
|
|
pending_totals = False
|
|
last_txn = None
|
|
i = 0
|
|
|
|
while i < len(lines):
|
|
ln = lines[i]
|
|
s = ln.strip()
|
|
|
|
if not s or is_page_header_noise(s):
|
|
i += 1
|
|
continue
|
|
|
|
# ACCOUNT header
|
|
m = acct_re.match(s)
|
|
if m:
|
|
current_account = m.group("account")
|
|
current_account_name = m.group("name").strip()
|
|
pending_totals = True
|
|
last_txn = None
|
|
i += 1
|
|
continue
|
|
|
|
# ACCOUNT totals
|
|
if pending_totals and money_re.search(s):
|
|
nums = [float(x.replace(",", "")) for x in money_re.findall(s)]
|
|
nums += [None] * 5
|
|
acct_summaries.append({
|
|
"ACCOUNT_NAME": current_account_name,
|
|
"ACCOUNT": current_account,
|
|
"OPEN": nums[0],
|
|
"TOTAL_DEBITS": nums[1],
|
|
"TOTAL_CREDITS": nums[2],
|
|
"NET_CHANGE": nums[3],
|
|
"CLOSE": nums[4],
|
|
})
|
|
pending_totals = False
|
|
i += 1
|
|
continue
|
|
|
|
# TRANSACTION
|
|
m = journal_re.match(s)
|
|
if m and current_account:
|
|
journal = m.group("journal")
|
|
date = m.group("date")
|
|
bch = int(m.group("bch")) if m.group("bch") else None
|
|
entry = m.group("entry")
|
|
|
|
debit = credit = None
|
|
desc = ""
|
|
|
|
if deb_pos is not None and cred_pos is not None:
|
|
debit = parse_money(ln[deb_pos:cred_pos])
|
|
credit = parse_money(ln[cred_pos:desc_pos])
|
|
desc = ln[desc_pos:].strip()
|
|
else:
|
|
after = s[m.end():].strip()
|
|
nums = money_re.findall(after)
|
|
if nums:
|
|
debit = float(nums[0].replace(",", ""))
|
|
desc = money_re.sub("", after).strip()
|
|
|
|
row = {
|
|
"ACCOUNT_NAME": current_account_name,
|
|
"ACCOUNT": current_account,
|
|
"JOURNAL": journal,
|
|
"DATE": date,
|
|
"BCH": bch,
|
|
"ENTRY": entry,
|
|
"DEBITS": debit,
|
|
"CREDITS": credit,
|
|
"DESCRIPTION": clean_description(desc),
|
|
}
|
|
transactions.append(row)
|
|
last_txn = row
|
|
i += 1
|
|
continue
|
|
|
|
# Continuation DESCRIPTION
|
|
if last_txn:
|
|
add = clean_description(s)
|
|
if add:
|
|
last_txn["DESCRIPTION"] += " " + add
|
|
i += 1
|
|
continue
|
|
|
|
i += 1
|
|
|
|
# -----------------------------
|
|
# Finalize
|
|
# -----------------------------
|
|
title = ", ".join(v for v in report_title_parts.values() if v)
|
|
|
|
df_txn = pd.DataFrame(transactions)
|
|
df_acct = pd.DataFrame(acct_summaries)
|
|
|
|
# -----------------------------
|
|
# Export XLSX
|
|
# -----------------------------
|
|
with pd.ExcelWriter(out_xlsx, engine="openpyxl") as writer:
|
|
df_txn.to_excel(writer, "Transactions", index=False, startrow=2)
|
|
df_acct.to_excel(writer, "Account Summary", index=False, startrow=2)
|
|
|
|
wb = openpyxl.load_workbook(out_xlsx)
|
|
for sheet in ["Transactions", "Account Summary"]:
|
|
ws = wb[sheet]
|
|
ws["A1"] = title
|
|
ws.merge_cells(start_row=1, start_column=1, end_row=1, end_column=ws.max_column)
|
|
ws["A1"].font = Font(bold=True)
|
|
ws.freeze_panes = "A4"
|
|
|
|
wb.save(out_xlsx)
|
|
|
|
print("✔ Parsed successfully")
|
|
print("Transactions:", df_txn.shape)
|
|
print("Account Summary:", df_acct.shape)
|
|
print("Output:", out_xlsx) |