srt-cleaner-project.py

import difflib
import re
from pathlib import Path

# --- CONFIG ---
SRT_FILE = "input.srt"
SCRIPT_FILE = "ceremony_script.txt"
NAMES_FILE = "graduate_names.txt"
OUTPUT_FILE = "cleaned_output.srt"
LOG_FILE = "review_log.txt"
FILLER_WORDS = {"uh", "um", "you know", "okay", "alright", "all right", "let's do it again"}

# --- UTILITIES ---
def load_script_lines(path):
    with open(path, 'r', encoding='utf-8') as f:
        return [line.strip() for line in f if line.strip()]

def load_names(path):
    names = set()
    with open(path, 'r', encoding='utf-8') as f:
        for line in f:
            names.add(line.strip().upper())
    return names

def srt_blocks(path):
    with open(path, 'r', encoding='utf-8') as f:
        content = f.read()
    blocks = content.strip().split("\n\n")
    parsed = []
    for block in blocks:
        lines = block.splitlines()
        if len(lines) >= 3:
            idx = lines[0]
            timestamp = lines[1]
            text = " ".join(lines[2:])
            parsed.append((idx, timestamp, text))
    return parsed

def clean_text(text):
    lowered = text.lower()
    for word in FILLER_WORDS:
        lowered = lowered.replace(word, "")
    return " ".join(lowered.split()).strip()

def match_script_block(caption_text, script_lines):
    for line in script_lines:
        ratio = difflib.SequenceMatcher(None, caption_text.lower(), line.lower()).ratio()
        if ratio > 0.85:
            return line, ratio
    return None, 0

def correct_names(text, grad_names):
    corrected = text
    flagged_names = []
    for name in grad_names:
        parts = name.split()
        if len(parts) == 2:
            first, last = parts
            pattern = re.compile(rf"{first[:3]}\w*\s+{last[:3]}\w*", re.IGNORECASE)
            match = pattern.search(text)
            if match and name.title() not in corrected:
                flagged_names.append(match.group())
                corrected = corrected.replace(match.group(), name.title())
    return corrected, flagged_names

# --- MAIN ---
script_lines = load_script_lines(SCRIPT_FILE)
grad_names = load_names(NAMES_FILE)
blocks = srt_blocks(SRT_FILE)

output_lines = []
log_lines = []

for idx, timestamp, text in blocks:
    cleaned = clean_text(text)
    replacement, ratio = match_script_block(cleaned, script_lines)
    corrected_text, name_flags = correct_names(text, grad_names)
    final_text = text
    
    flags = []

    if ratio < 0.85:
        flags.append("SCRIPT_MISMATCH")
        final_text = corrected_text

    if any(word.lower() in cleaned.lower() for word in FILLER_WORDS):
        flags.append("FILLER_DETECTED")

    if name_flags:
        flags.append(f"NAME_FIX({', '.join(name_flags)})")
        final_text = corrected_text

    if flags:
        final_text = f"[?] {final_text}"
        log_lines.append(f"Block {idx} | Flags: {', '.join(flags)}\nOriginal: {text}\nFinal: {final_text}\n")

    output_lines.append(f"{idx}\n{timestamp}\n{final_text}\n")

with open(OUTPUT_FILE, 'w', encoding='utf-8') as out:
    out.write("\n".join(output_lines))

with open(LOG_FILE, 'w', encoding='utf-8') as log:
    log.write("\n\n".join(log_lines))

print(f"✅ Done! Cleaned captions saved to {OUTPUT_FILE}")
print(f"📋 Review log saved to {LOG_FILE}")