#!/usr/bin/env python3
from pathlib import Path
import re

inp = Path(r"d:\TJWaterServer\epanet\szhskeleton-patternfixed-ascii.inp")
out = Path(r"d:\TJWaterServer\epanet\szhskeleton-patternfixed-ascii-fixed2.inp")
mapout = Path(r"d:\TJWaterServer\epanet\szhskeleton-patternfixed-ascii-fixed2.mapping.txt")

text = inp.read_text(encoding='utf-8')
lines = text.splitlines()

# find [VALVES] start and end
start = None
for i,l in enumerate(lines):
    if l.strip().upper() == '[VALVES]':
        start = i
        break
if start is None:
    print('No [VALVES] section found')
    raise SystemExit(1)
end = len(lines)
for j in range(start+1, len(lines)):
    if re.match(r"^\s*\[.+\]", lines[j]):
        end = j
        break

# collect valve lines with their absolute numbers
valve_entries = []  # (absolute_line_index, token, line)
for idx in range(start+1, end):
    l = lines[idx]
    if not l.strip() or l.strip().startswith(';'):
        continue
    tok = l.split()[0]
    valve_entries.append((idx, tok, l))

from collections import defaultdict
positions = defaultdict(list)
for ln, tok, l in valve_entries:
    positions[tok].append(ln)

# find duplicates
dups = {tok:lns for tok,lns in positions.items() if len(lns)>1}
print('Found', sum(1 for _ in valve_entries), 'valve entries; duplicates:', len(dups))

replacements = []  # (line_index, old, new)
counter = 1
for tok, lns in dups.items():
    # skip first occurrence, rename others
    for occ_index, ln in enumerate(lns):
        if occ_index == 0:
            continue
        # produce new name: prefix V if starts with digit
        if re.fullmatch(r"\d+", tok) or re.match(r"^\d", tok):
            base = 'V' + tok
        else:
            base = tok
        new = f'{base}_{occ_index}'
        # ensure uniqueness globally
        while any(rn == new for _,_,rn in replacements) or any(new == t for t in positions.keys()):
            counter += 1
            new = f'{base}_{occ_index}_{counter}'
        replacements.append((ln, tok, new))

# Apply replacements on the given absolute lines
for ln, old, new in replacements:
    line = lines[ln]
    # replace only first token occurrence
    parts = line.split()
    if parts:
        # find start of token in line (preserve spacing)
        m = re.search(re.escape(parts[0]), line)
        if m:
            startpos = m.start()
            endpos = m.end()
            newline = line[:startpos] + new + line[endpos:]
            lines[ln] = newline

# write new file
out.write_text('\n'.join(lines) + '\n', encoding='utf-8')
# write mapping
with mapout.open('w', encoding='utf-8') as f:
    for ln, old, new in replacements:
        f.write(f'line {ln+1}: {old} -> {new}\n')

print('Wrote', out, 'with', len(replacements), 'replacements; mapping at', mapout)