alih-kata-tanpa-penyahtaksa.py

# The code mostly generated by ChatGPT (I can't code and just stitch it together part by part)
# Even I don't understand much but I will explain what the code will do inside README.md

import pysubs2
import os
import re

folder_masuk = "Masuk"
folder_keluar = "Keluar"
folder_kamus = "Kamus"

def muat_katan(file_path):
    data = {}
    with open(file_path) as file:
        lines = file.readlines()
        for line in lines:
            rumi, jawi = line.strip().split("\t")
            rumi = rumi.lower()
            if rumi in data:
                data[rumi].append(jawi)
            else:
                data[rumi] = [jawi]
    return data

def alih_kata(padanan, translations, context):
    katan = padanan.group(0)
    if katan.lower() in translations:
        translations_list = translations[katan.lower()]
        if len(translations_list) > 0:
            return translations_list[0]
    return katan

def alih_ayat(ayat, translations, padanan_tanda):
    def gantikan_tanda(padanan):
        tanda = padanan.group(0)
        if tanda in padanan_tanda:
            return padanan_tanda[tanda]
        return tanda

    ayat_diterjemah = re.sub(r'(?<!\\)(\\[Nnh])', r' {\1} ', ayat)

    ayat_diterjemah = re.sub(r'(?<!{){[^}]+}|[A-Za-z]+', lambda padanan: alih_kata(padanan, translations, ayat), ayat_diterjemah)
    ayat_diterjemah = re.sub(r'[?,;,.]', gantikan_tanda, ayat_diterjemah)

    ayat_diterjemah = ayat_diterjemah.replace('{\\N}', '\\N').replace('{\\n}', '\\n').replace('{\\h}', '\\h')

    return ayat_diterjemah

def baiki_ejaan(teks):
    # Cari dan ganti 'د ' dan 'ک ' yang berjarak untuk membetulkan perkataan cthnya 'د سکوله'
    teks = re.sub(r'\bد\s', 'د', teks)
    teks = re.sub(r'\bک\s', 'ک', teks)

    return teks

def alih_kata_sarikata(file_path, translations, padanan_tanda, padanan_tanggaman_akhiran, padanan_tanggaman_awalan):
    sarikata = pysubs2.load(file_path)
    tidak_teralih = {}
    extracted_content = {}

    for i, dialog in enumerate(sarikata.events):
        if dialog.is_comment:
            continue

        teks_dialog = dialog.text

        extracted = re.findall(r'{(.*?)}', teks_dialog)
        for extract in extracted:
            placeholder = f'__{len(extracted_content)}__'
            extracted_content[placeholder] = '{' + extract + '}'
            teks_dialog = teks_dialog.replace('{' + extract + '}', placeholder)

        dialog_terjemah = baiki_ejaan(alih_ayat(teks_dialog, translations, padanan_tanda))
        dialog_terjemah = "{\\fe-1}" + dialog_terjemah

        for placeholder, content in extracted_content.items():
            dialog_terjemah = dialog_terjemah.replace(placeholder, content)

        dialog.text = dialog_terjemah

        katan_rumi_jawi = re.findall(r'(?<!{)(?<!\\)(?:\\\\)*(?<!\\[Nnh])\b[A-Za-z]+(?<!\\)(?!})(?<!\\[Nnh])', teks_dialog)
        for katan in katan_rumi_jawi:
            if katan.lower() not in translations and not re.search(r'(?<!\\){[^}]+}', teks_dialog):
                katan_terjemah = alih_kata(re.search(katan, teks_dialog), translations, teks_dialog)
                if katan_terjemah == katan:
                    for tanggaman_akhiran in padanan_tanggaman_akhiran:
                        if katan.lower().endswith(tanggaman_akhiran):
                            katan_terjemah = alih_kata(re.search(katan[:-len(tanggaman_akhiran)], teks_dialog), translations, teks_dialog) + padanan_tanggaman_akhiran[tanggaman_akhiran]
                            break
                if katan_terjemah == katan:
                    for tanggaman_awalan in padanan_tanggaman_awalan:
                        if katan.lower().startswith(tanggaman_awalan):
                            katan_terjemah = padanan_tanggaman_awalan[tanggaman_awalan] + alih_kata(re.search(katan[len(tanggaman_awalan):], teks_dialog), translations, teks_dialog)
                            break
                if katan_terjemah != katan:
                    dialog_terjemah = dialog_terjemah.replace(katan, katan_terjemah)
                else:
                    tidak_teralih.setdefault(i, []).append(katan)

        sarikata.events[i].text = dialog_terjemah

    laluan_sarikata = os.path.join(folder_keluar, "[Dialih Kata]" + os.path.basename(file_path))
    sarikata.save(laluan_sarikata)

    return tidak_teralih

katan_kamus = {}

for file in os.listdir(folder_kamus):
    if file.endswith(".tsv"):
        file_path = os.path.join(folder_kamus, file)
        katan_kamus = {**katan_kamus, **muat_katan(file_path)}

padanan_tanda = {"?": "؟", ";": "⁏", ",": "⹁", ".": "."}
padanan_tanggaman_akhiran = {"lah": "له", "kah": "که", "nya": "ڽ", "kan": "کن", "i": "ي", "ku": "کو", "mu": "مو"}
padanan_tanggaman_awalan = {"ber": "بر", "mem": "مم", "meng": "مڠ", "se": "س", "tak": "تق", "per": "ڤر"}

files_untranslated = {}

for fail in os.listdir(folder_masuk):
    if fail.endswith((".ass", ".ssa", ".srt")):
        print(f"Mengalih kata: {fail}\n---")
        laluan_fail = os.path.join(folder_masuk, fail)
        tidak_teralih = alih_kata_sarikata(laluan_fail, katan_kamus, padanan_tanda, padanan_tanggaman_akhiran, padanan_tanggaman_awalan)
        if len(tidak_teralih) > 0:
            files_untranslated[fail] = tidak_teralih

for fail, tidak_teralih in files_untranslated.items():
    print(f"Perkataan yang tidak teralih dalam \"{fail}\":")
    for i, dialog_index in enumerate(tidak_teralih):
        dialog = files_untranslated[fail][dialog_index]
        print(f"Baris {dialog_index + 1}: {dialog}")
    print()

print(
    "Pengalih tulisan ini tidaklah sempurna, sentuhan manusia tetap juga diperlukan."
)