from nahiarhdNLP.preprocessing import Pipeline

# Create a pipeline with configuration
config = {
    "clean_html": True,
    "clean_mentions": True,
    "remove_urls": True,
    "stopword": True
}

pipeline = Pipeline(config)

# Process text
text = "Haii @user!! Cek website kita di https://example.com ya 😊"
result = pipeline.process(text)

print(result)
# Output: "Haii Cek website kita ya 😊"

from nahiarhdNLP.preprocessing import Pipeline

# Configure pipeline
config = {
    "clean_html": True,
    "clean_mentions": True,
    "remove_urls": True
}

pipeline = Pipeline(config)

# Input
text = "Hello <b>World</b>! Mention @user123 and visit https://example.com"

# Process
result = pipeline.process(text)

print(f"Input : {text}")
print(f"Output: {result}")

from nahiarhdNLP.preprocessing import Pipeline

config = {
    "clean_html": True,
    "clean_mentions": True,
    "clean_hashtags": True,
    "remove_urls": True,
    "remove_emoji": True,
    "remove_extra_spaces": True
}

pipeline = Pipeline(config)

# Input - Typical social media post
text = """
Haiii gengs!! 😍😍
Jangan lupa follow @nahiarhdNLP ya!
Cek website kita di https://github.com/nahiarhd
#NLP #IndonesianNLP #TextProcessing 🚀
"""

result = pipeline.process(text)

print("=" * 60)
print("INPUT:")
print(text)
print("=" * 60)
print("OUTPUT:")
print(result)
print("=" * 60)

from nahiarhdNLP.preprocessing import Pipeline

# Initial configuration
config = {"clean_html": True, "remove_urls": True}
pipeline = Pipeline(config)

text = "<p>Visit https://example.com for more info</p>"

print(f"Initial Output: {pipeline.process(text)}")
# Output: Visit for more info

# Update configuration
pipeline.update_config({"remove_punctuation": True})

print(f"Updated Output: {pipeline.process(text)}")
# Output: Visit for more info

# Check enabled steps
print(f"Enabled steps: {pipeline.get_enabled_steps()}")
# Output: ['clean_html', 'remove_urls', 'remove_punctuation']

from nahiarhdNLP.preprocessing import Pipeline

# Get all available features
all_features = Pipeline.get_available_steps()

print("All Available Features:")
for feature_name, description in sorted(all_features.items()):
    print(f"  {feature_name:25} - {description}")

print(f"\nTotal Features: {len(all_features)}")

# Get features organized by category
features_by_category = Pipeline.get_available_steps_by_category()

print("\nFeatures by Category:")
for category, feature_names in features_by_category.items():
    print(f"\n{category}:")
    for feature_name in feature_names:
        description = all_features.get(feature_name, "No description")
        print(f"  {feature_name:25} - {description}")

from nahiarhdNLP.preprocessing import Pipeline

config = {"clean_html": True}
pipeline = Pipeline(config)

# Test various HTML tags
examples = [
    "<p>This is a paragraph</p>",
    "<div class='container'>Content here</div>",
    "Normal text <b>bold text</b> <i>italic</i>",
    "<script>alert('test')</script>Clean text"
]

for text in examples:
    result = pipeline.process(text)
    print(f"Input : {text}")
    print(f"Output: {result}")
    print("-" * 60)

from nahiarhdNLP.preprocessing import Pipeline

# Remove URLs completely
config_remove = {"remove_urls": True}
pipeline_remove = Pipeline(config_remove)

# Clean URLs (remove protocol only)
config_clean = {"clean_urls": True}
pipeline_clean = Pipeline(config_clean)

text = "Visit https://github.com and http://example.com for more info"

print(f"Original     : {text}")
print(f"Remove URLs  : {pipeline_remove.process(text)}")
print(f"Clean URLs   : {pipeline_clean.process(text)}")

from nahiarhdNLP.preprocessing import Pipeline

text = "Hey @john_doe and @jane! Check out #Python #MachineLearning #AI"

# Remove mentions and hashtags
config_remove = {"remove_mentions": True, "remove_hashtags": True}
pipeline_remove = Pipeline(config_remove)

# Clean mentions and hashtags (keep text)
config_clean = {"clean_mentions": True, "clean_hashtags": True}
pipeline_clean = Pipeline(config_clean)

print(f"Original        : {text}")
print(f"Remove @#       : {pipeline_remove.process(text)}")
print(f"Clean @# (keep) : {pipeline_clean.process(text)}")

from nahiarhdNLP.preprocessing import Pipeline

config = {"remove_emoji": True}
pipeline = Pipeline(config)

examples = [
    "I love Python 🐍❤️",
    "Great work! 👍😊🎉",
    "Weather today ☀️🌧️⛈️",
]

for text in examples:
    result = pipeline.process(text)
    print(f"Input : {text}")
    print(f"Output: {result}")
    print()

from nahiarhdNLP.preprocessing import Pipeline

config = {"remove_repeated_chars": True}
pipeline = Pipeline(config)

examples = [
    "Haiiiii guys!!!",
    "Kangennnnn bangetttt",
    "Wowwwww kerennn",
    "Makasiiih yaaaa"
]

for text in examples:
    result = pipeline.process(text)
    print(f"Input : {text}")
    print(f"Output: {result}")

from nahiarhdNLP.preprocessing.normalization.emoji import EmojiConverter

emoji = EmojiConverter()
emoji._load_data()

# Emoji to Text
text_with_emoji = "Hari ini cuaca cerah ☀️ dan saya senang 😊"
result = emoji.emoji_to_text_convert(text_with_emoji)
print(f"Emoji to Text:")
print(f"Input : {text_with_emoji}")
print(f"Output: {result}")
print()

# Text to Emoji (example - depends on your emoji dataset)
text = "saya senang wajah tersenyum"
result = emoji.text_to_emoji_convert(text)
print(f"Text to Emoji:")
print(f"Input : {text}")
print(f"Output: {result}")

from nahiarhdNLP.preprocessing.normalization.spell_corrector import SpellCorrector

spell = SpellCorrector()

# Single word correction
words = ["sya", "tdk", "gk", "org", "yg", "dgn"]
print("Word Correction:")
for word in words:
    corrected = spell.correct_word(word)
    print(f"  {word:10s} → {corrected}")

print("\n" + "="*60 + "\n")

# Sentence correction
sentences = [
    "gw lg di rmh",
    "gmn kabar lo?",
    "knp gk dtg?",
    "jgn lupa ya"
]

print("Sentence Correction:")
for sent in sentences:
    corrected = spell.correct_sentence(sent)
    print(f"Input : {sent}")
    print(f"Output: {corrected}")
    print()

from nahiarhdNLP.preprocessing import Pipeline

# Comprehensive normalization pipeline
config = {
    "clean_html": True,
    "clean_mentions": True,
    "clean_hashtags": True,
    "remove_urls": True,
    "remove_emoji": True,
    "remove_extra_spaces": True,
    "remove_repeated_chars": True,
    "spell_corrector_sentence": True,
    "remove_lowercase": True
}

pipeline = Pipeline(config)

# Messy Indonesian text
text = """
Haiii @temans!! 😍 Kmrn gw udh coba apps baruu loh di https://example.com
#KerenBanget #Recommended Gkkkk nyesel dehhhh!!! 🚀🚀
"""

result = pipeline.process(text)

print("=" * 70)
print("ORIGINAL TEXT:")
print(text)
print("=" * 70)
print("NORMALIZED TEXT:")
print(result)
print("=" * 70)

from nahiarhdNLP.preprocessing.linguistic.stemmer import Stemmer

stemmer = Stemmer()

# Test various Indonesian words
words = [
    "bermain",      # playing
    "berlari",      # running
    "kebahagiaan",  # happiness
    "pembelajaran", # learning
    "menyenangkan", # enjoyable
    "berkomunikasi" # communicate
]

print("Indonesian Stemming:")
print(f"{'Word':<20} → {'Stem'}")
print("-" * 40)
for word in words:
    stem = stemmer.stem(word)
    print(f"{word:<20} → {stem}")

print("\n" + "="*60 + "\n")

# Sentence stemming
sentences = [
    "Saya sedang belajar pemrograman Python",
    "Mereka bermain bola di lapangan",
    "Kebahagiaan adalah kunci kesuksesan"
]

print("Sentence Stemming:")
for sent in sentences:
    stemmed = stemmer.stem(sent)
    print(f"Input : {sent}")
    print(f"Output: {stemmed}")
    print()

from nahiarhdNLP.preprocessing.linguistic.stopword import StopwordRemover

stopword = StopwordRemover()
stopword._load_data()

# Test sentences
sentences = [
    "Saya sedang belajar bahasa pemrograman Python untuk data science",
    "Mereka akan pergi ke pasar besok pagi",
    "Ini adalah contoh kalimat dengan banyak stopwords yang harus dihapus"
]

print("Stopword Removal:")
print("=" * 70)
for sent in sentences:
    cleaned = stopword.remove_stopwords(sent)
    print(f"Original: {sent}")
    print(f"Cleaned : {cleaned}")
    print("-" * 70)

from nahiarhdNLP.preprocessing import Pipeline

# Linguistic processing pipeline
config = {
    "remove_lowercase": True,
    "stopword": True,
    "stem": True,
    "remove_extra_spaces": True
}

pipeline = Pipeline(config)

texts = [
    "Saya sedang mengembangkan aplikasi pembelajaran online",
    "Mereka bermain musik dengan sangat menyenangkan",
    "Kebahagiaan adalah perjalanan bukan tujuan"
]

print("Complete Linguistic Processing:")
print("=" * 70)
for text in texts:
    result = pipeline.process(text)
    print(f"Original : {text}")
    print(f"Processed: {result}")
    print("-" * 70)

from nahiarhdNLP.preprocessing.tokenization.tokenizer import Tokenizer

tokenizer = Tokenizer()

texts = [
    "Ini adalah contoh kalimat sederhana",
    "Python, Java, dan JavaScript adalah bahasa pemrograman",
    "Email: [email protected], Website: https://example.com"
]

print("Tokenization Examples:")
print("=" * 70)
for text in texts:
    tokens = tokenizer.tokenize(text)
    print(f"Text  : {text}")
    print(f"Tokens: {tokens}")
    print("-" * 70)

from nahiarhdNLP.preprocessing import Pipeline

# Configure replacement pipeline
config = {
    "replace_email": True,
    "replace_link": True,
    "replace_user": True
}

pipeline = Pipeline(config)

examples = [
    "Contact me at [email protected] for more info",
    "Visit https://github.com/nahiarhd for the code",
    "Thanks @john and @jane for your help!",
    "Email: [email protected] | Web: https://company.com | Twitter: @company"
]

print("Text Replacement:")
print("=" * 70)
for text in examples:
    result = pipeline.process(text)
    print(f"Input : {text}")
    print(f"Output: {result}")
    print("-" * 70)

from nahiarhdNLP.preprocessing import Pipeline

# Complete anonymization pipeline
config = {
    "replace_email": True,
    "replace_link": True,
    "replace_user": True,
    "remove_phones": True,
    "clean_html": True
}

pipeline = Pipeline(config)

# Sensitive data example
text = """
<div class="contact">
Customer: @johndoe
Email: [email protected]
Phone: +62-812-3456-7890
Website: https://customer-site.com
</div>
"""

result = pipeline.process(text)

print("DATA ANONYMIZATION")
print("=" * 70)
print("ORIGINAL:")
print(text)
print("=" * 70)
print("ANONYMIZED:")
print(result)
print("=" * 70)

from nahiarhdNLP.datasets import DatasetLoader

loader = DatasetLoader()

# Load stopwords
stopwords = loader.load_stopwords_dataset()
print(f"📚 Stopwords Dataset:")
print(f"   Total words: {len(stopwords)}")
print(f"   Sample: {stopwords[:10]}")
print()

# Load slang dictionary
slang_dict = loader.load_slang_dataset()
print(f"💬 Slang Dictionary:")
print(f"   Total entries: {len(slang_dict)}")
print(f"   Sample mappings:")
for slang, formal in list(slang_dict.items())[:5]:
    print(f"      {slang:10s} → {formal}")
print()

# Load emoji dictionary
emoji_dict = loader.load_emoji_dataset()
print(f"😊 Emoji Dictionary:")
print(f"   Total emojis: {len(emoji_dict)}")
print(f"   Sample mappings:")
for emoji, text in list(emoji_dict.items())[:5]:
    print(f"      {emoji:5s} → {text}")
print()

# Load wordlist
wordlist = loader.load_wordlist_dataset()
print(f"📖 Wordlist Dataset:")
print(f"   Total words: {len(wordlist)}")
print(f"   Sample: {wordlist[:10]}")

config = {
    # ===== TEXT CLEANING =====
    # HTML & Tags
    "clean_html": True,              # Remove HTML tags

    # URLs
    "remove_urls": True,             # Remove complete URLs
    "clean_urls": True,              # Remove URL protocols (http://, https://)

    # Social Media
    "remove_mentions": True,         # Remove @mentions completely
    "clean_mentions": True,          # Remove @ but keep username
    "remove_hashtags": True,         # Remove #hashtags completely
    "clean_hashtags": True,          # Remove # but keep tag text

    # Content Removal
    "remove_emoji": True,            # Remove emoji characters
    "remove_punctuation": True,      # Remove punctuation marks
    "remove_numbers": True,          # Remove numbers
    "remove_emails": True,           # Remove email addresses
    "remove_phones": True,           # Remove phone numbers
    "remove_currency": True,         # Remove currency symbols

    # Text Cleaning
    "remove_special_chars": True,    # Remove special characters
    "remove_extra_spaces": True,     # Normalize whitespace
    "remove_repeated_chars": True,   # Normalize repeated characters (e.g., "haiiii" → "haii")
    "remove_whitespace": True,       # Clean tabs, newlines, etc.
    "remove_lowercase": True,        # Convert to lowercase

    # ===== TEXT NORMALIZATION =====
    "emoji_to_text": True,           # Convert emojis to text description
    "text_to_emoji": True,           # Convert text to emojis
    "spell_corrector_word": True,    # Correct spelling for single words
    "spell_corrector_sentence": True, # Correct spelling for sentences

    # ===== LINGUISTIC PROCESSING =====
    "stem": True,                    # Apply stemming (reduce to root form)
    "stopword": True,                # Remove stopwords
    "tokenizer": True,               # Tokenize text

    # ===== TEXT REPLACEMENT =====
    "replace_email": True,           # Replace emails with <email>
    "replace_link": True,            # Replace URLs with <link>
    "replace_user": True,            # Replace mentions with <user>
}

class Pipeline:
    """
    Configurable text preprocessing pipeline for Indonesian text.

    Args:
        config (dict): Dictionary of preprocessing steps {step_name: True/False}

    Methods:
        process(text: str) -> str: Process text through the pipeline
        update_config(new_config: dict) -> None: Update pipeline configuration
        get_enabled_steps() -> list: Get list of enabled processing steps
        __call__(text: str) -> str: Allow pipeline to be called as a function

    Example:
        >>> config = {"clean_html": True, "stopword": True}
        >>> pipeline = Pipeline(config)
        >>> result = pipeline.process("<p>Saya sedang belajar NLP</p>")
        >>> # or use as callable
        >>> result = pipeline("<p>Saya sedang belajar NLP</p>")
    """