Files
Aislo/resources/tester/scratch/find_unclassified_grouped_tables.py
T

106 lines
4.9 KiB
Python

import os, sys, re
sys.stdout.reconfigure(encoding='utf-8')
target_files = [
# 02_토목부문
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제1장_도로포장공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제2장_하천공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제3장_터널공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제4장_궤도공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제5장_강구조공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제6장_관부설및접합공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제7장_항만공사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제8장_지반조사.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/02_토목부문/제9장_측량.md',
# 05_유지관리부문
'resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문/제1장_공통.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문/제2장_토목.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문/제3장_건축.md',
'resources/knowledge/original/원가계산/건설공사_표준품셈/05_유지관리부문/제4장_기계설비.md',
# 04_기계설비부문 제13장
'resources/knowledge/original/원가계산/건설공사_표준품셈/04_기계설비부문/제13장_플랜트설비공사.md',
]
print("=== Scanning for Condition/Grouped Tables without Item Classification Numbers ===")
grouped_tables = []
for fpath in target_files:
fname = os.path.basename(fpath)
with open(fpath, 'r', encoding='utf-8') as f:
lines = f.readlines()
tables = []
cur_t = []
cur_start = 0
cur_title = ""
for idx, l in enumerate(lines):
line_str = l.strip()
if not line_str.startswith('|'):
if line_str and not line_str.startswith('>'):
cur_title = line_str
if cur_t:
tables.append((cur_start, cur_title, cur_t))
cur_t = []
else:
if not cur_t:
cur_start = idx + 1
cur_t.append((idx + 1, line_str))
if cur_t:
tables.append((cur_start, cur_title, cur_t))
for t_start, t_title, t_rows in tables:
data_rows = []
for lno, r_str in t_rows:
cells = [c.strip() for c in r_str.split('|')[1:-1]]
if all(re.match(r'^:?-+:?$', c) for c in cells if c):
continue
if cells and cells[0] in ["비고", "비 고", "주", "[주]"]:
continue
data_rows.append((lno, r_str, cells))
if len(data_rows) < 3:
continue
# Check if rows have item numbers like 1-1, 1-2, 1-1-1 or 4-3-1 etc.
has_item_numbers = False
for lno, r_str, cells in data_rows[1:]:
c0 = cells[0] if cells else ""
if re.match(r'^\d+-\d+(?:-\d+)?', c0) or re.match(r'^[A-Z]\d{4}', c0):
has_item_numbers = True
break
if has_item_numbers:
continue # Safe, has explicit classification numbers
# Check if this table has condition/branch groups
# (e.g. multi-level condition headers or blank cells inheriting above groups)
has_grouping = False
group_col_values = []
for lno, r_str, cells in data_rows[1:]:
c0 = cells[0] if cells else ""
c1 = cells[1] if len(cells) > 1 else ""
# Inherited blank cell or condition keyword
if c0 == "" or c0 == "〃" or any(k in c0 for k in ["기층", "보조기층", "포장", "Type", "식", "식재", "토사", "암반", "외업", "내업", "직선", "곡선"]):
has_grouping = True
if c0:
group_col_values.append(c0)
if has_grouping and len(set(group_col_values)) >= 2:
grouped_tables.append({
'file': fname,
'line': t_start,
'title': t_title[:60],
'rows_count': len(data_rows),
'groups': list(set(group_col_values))[:6],
'sample_header': data_rows[0][1][:70]
})
print(f"Total Condition/Grouped Tables found in our scope: {len(grouped_tables)}")
for gt in grouped_tables:
print(f"\n[{gt['file']}:L{gt['line']}] {gt['title']}")
print(f" Header: {gt['sample_header']}")
print(f" Groups: {gt['groups']} (Rows: {gt['rows_count']})")