import sys, os, re, json
sys.stdout.reconfigure(encoding='utf-8')

base_dir = "resources/knowledge/original/원가계산/건설공사_표준품셈"
merged_file = os.path.join(base_dir, "2026년_건설공사_표준품셈.md")

with open(merged_file, 'r', encoding='utf-8') as f:
    merged_text = f.read()

# Helper to count tables in text
def count_tables_in_text(text):
    lines = text.split('\n')
    tables = 0
    curr = []
    for l in lines:
        ls = l.strip()
        if ls.startswith('|') and ls.endswith('|'):
            curr.append(ls)
        else:
            if curr:
                if any(re.match(r'^\|(?:\s*:?-+:?\s*\|)+$', r) for r in curr):
                    tables += 1
                curr = []
    if curr and any(re.match(r'^\|(?:\s*:?-+:?\s*\|)+$', r) for r in curr):
        tables += 1
    return tables

# Count tables in each split file
all_split_files = []
for root, dirs, files in os.walk(base_dir):
    for f in files:
        if f.endswith('.md') and not f.startswith('_') and '개정사항' not in f and '2026년_건설공사_표준품셈.md' not in f:
            all_split_files.append(os.path.join(root, f))

total_split_tables = 0
split_counts = {}
for p in sorted(all_split_files):
    rel = os.path.relpath(p, base_dir)
    with open(p, 'r', encoding='utf-8') as f:
        t = f.read()
    c = count_tables_in_text(t)
    split_counts[rel] = c
    total_split_tables += c

print(f"Total split tables: {total_split_tables}")
merged_total = count_tables_in_text(merged_text)
print(f"Total merged tables: {merged_total}")