import re
from nltk.tokenize import word_tokenize
from nltk.corpus import stopwords
from Sastrawi.Stemmer.StemmerFactory import StemmerFactory

# ============================================================
# STEMMER SASTRAWI
# ============================================================

stemmer_factory = StemmerFactory()
stemmer = stemmer_factory.create_stemmer()

# ============================================================
# STOPWORD
# ============================================================

partikel_tambahan = {
    "sih",
    "si",
    "kan",
    "ya",
    "bang",
    "dong",
    "lah",
    "deh",
    "kah",
    "kok"
}

kata_negasi = {
    "tidak",
    "bukan",
    "belum",
    "jangan",
    "tanpa",
    "kurang",
    "malah"
}

list_stopwords = set(stopwords.words("indonesian"))
list_stopwords = list_stopwords | partikel_tambahan

# Kata negasi harus tetap dipertahankan
list_stopwords = list_stopwords - kata_negasi

# ============================================================
# KAMUS NORMALISASI
# ============================================================
#
# Ini baru struktur awal backend.
# Seluruh kamus final dari notebook penelitian nanti
# dimasukkan ke bagian ini agar hasil preprocessing web
# identik dengan penelitian.
# ============================================================

normalization_map = {
    "download": "unduh",
    "upload": "unggah",
    "scan": "pindai",
    "matkul": "mata kuliah",
    "gabisa": "tidak bisa",
    "gakbisa": "tidak bisa",
    "nggakbisa": "tidak bisa",
    "lemot": "lambat",
    "submit": "unggah",
    "userfriendly": "ramah pengguna",
    "gak": "tidak",
    "ga": "tidak",
    "nggak": "tidak"
}

phrase_normalization = {
    "di gunakan": "digunakan",
    "didalam": "di dalam",
    "diluar": "di luar"
}

# ============================================================
# 1. CASE FOLDING
# ============================================================

def case_folding(text):
    return str(text).lower()

# ============================================================
# 2. CLEANING
# ============================================================

def cleaning(text):
    text = str(text)

    # Hapus URL
    text = re.sub(r"https?://\S+|www\.\S+", " ", text)

    # Hapus mention
    text = re.sub(r"@\w+", " ", text)

    # Hapus hashtag beserta simbol #
    text = re.sub(r"#\w+", " ", text)

    # Hapus angka
    text = re.sub(r"\d+", " ", text)

    # Hanya pertahankan huruf dan spasi
    text = re.sub(r"[^a-zA-Z\s]", " ", text)

    # Rapikan spasi
    text = re.sub(r"\s+", " ", text).strip()

    return text

# ============================================================
# 3. NORMALISASI
# ============================================================

def normalization(text):
    text = str(text)

    # Normalisasi bentuk/frasa tertentu terlebih dahulu
    for source, target in phrase_normalization.items():
        text = re.sub(
            rf"\b{re.escape(source)}\b",
            target,
            text
        )

    words = text.split()
    normalized_words = []

    for word in words:
        replacement = normalization_map.get(word, word)

        # Replacement boleh menghasilkan lebih dari satu kata
        normalized_words.extend(replacement.split())

    return " ".join(normalized_words)

# ============================================================
# 4. TOKENIZING
# ============================================================

def tokenizing(text):
    return word_tokenize(str(text))

# ============================================================
# 5. STOPWORD REMOVAL
# ============================================================

def stopword_removal(tokens):
    return [
        token
        for token in tokens
        if token not in list_stopwords
    ]

# ============================================================
# 6. STEMMING
# ============================================================

def stemming(tokens):
    text = " ".join(tokens)
    return stemmer.stem(text)

# ============================================================
# PIPELINE PREPROCESSING
# ============================================================

def preprocess_text(text):
    original_text = str(text)

    casefolded_text = case_folding(original_text)

    cleaned_text = cleaning(casefolded_text)

    normalized_text = normalization(cleaned_text)

    tokenized_text = tokenizing(normalized_text)

    stopword_tokens = stopword_removal(tokenized_text)

    stopword_text = " ".join(stopword_tokens)

    stemmed_text = stemming(stopword_tokens)

    return {
        "original_text": original_text,
        "case_folding": casefolded_text,
        "cleaning": cleaned_text,
        "normalization": normalized_text,
        "tokenizing": tokenized_text,
        "stopword_removal": stopword_text,
        "stemming": stemmed_text
    }