satupeta-main/services/upload_file/upload.py

import json
import os
import random
import subprocess
import uuid
import asyncpg
import pandas as pd
import geopandas as gpd
import numpy as np
import re
import zipfile
import tempfile
import asyncio
from pyproj import CRS
from shapely.geometry.base import BaseGeometry
from shapely.geometry import base as shapely_base
from fastapi import Depends, File, Form, UploadFile, HTTPException
from api.routers.datasets_router import cleansing_data, publish_layer, query_cleansing_data, upload_to_main
from core.config import DB_DSN, DB_HOST, DB_NAME, DB_PASS, DB_PORT, DB_USER, UPLOAD_FOLDER, MAX_FILE_MB, GEONETWORK_URL
from services.upload_file.ai_generate import send_metadata
from services.upload_file.readers.reader_csv import read_csv
from services.upload_file.readers.reader_shp import read_shp
from services.upload_file.readers.reader_gdb import read_gdb
from services.upload_file.readers.reader_mpk import read_mpk
from services.upload_file.readers.reader_pdf import convert_df, read_pdf
from services.upload_file.utils.df_validation import process_dataframe_synchronous
from services.upload_file.utils.geometry_detector import detect_and_build_geometry, attach_polygon_geometry_auto
from database.connection import engine, sync_engine
from pydantic import BaseModel
from typing import Any, Dict, List, Optional
from shapely import MultiLineString, MultiPolygon, wkt
from sqlalchemy import text
from datetime import datetime
from response import successRes, errorRes
from utils.logger_config import log_activity
# Base.metadata.create_all(bind=engine)


def is_geom_empty(g):
    if g is None:
        return True
    if isinstance(g, float) and pd.isna(g):
        return True
    if isinstance(g, BaseGeometry):
        return g.is_empty
    return False


def safe_json(value):
    """Konversi aman untuk semua tipe numpy/pandas/shapely ke tipe JSON-serializable"""
    if isinstance(value, (np.int64, np.int32)):
        return int(value)
    if isinstance(value, (np.float64, np.float32)):
        return float(value)
    if isinstance(value, pd.Timestamp):
        return value.isoformat()
    if isinstance(value, shapely_base.BaseGeometry):
        return str(value)  # convert to WKT string
    if pd.isna(value):
        return None
    return value


def detect_zip_type(zip_path: str) -> str:
    with zipfile.ZipFile(zip_path, "r") as zip_ref:
        files = zip_ref.namelist()

    if any(f.lower().endswith(".gdb/") or ".gdb/" in f.lower() for f in files):
        return "gdb"

    if any(f.lower().endswith(ext) for ext in [".gdbtable", ".gdbtablx", ".gdbindexes", ".spx"] for f in files):
        return "gdb"

    if any(f.lower().endswith(".shp") for f in files):
        return "shp"

    return "unknown"


async def process_data(df: pd.DataFrame, ext: str, filename: str, fileDesc: str):
    result = detect_and_build_geometry(df, master_polygons=None)

    if not hasattr(result, "geometry") or result.geometry.isna().all():
        result = attach_polygon_geometry_auto(result)

    def normalize_geom_type(geom_type):
        if geom_type.startswith("Multi"):
            return geom_type.replace("Multi", "")
        return geom_type

    if isinstance(result, gpd.GeoDataFrame) and "geometry" in result.columns:
        geom_types = (
            result.geometry
            .dropna()
            .geom_type
            .apply(normalize_geom_type)
            .unique()
        )

        geom_type = geom_types[0] if len(geom_types) > 0 else "None"
        null_geom = result.geometry.isna().sum()

        print(f"[INFO] Tipe Geometry: {geom_type}")
        print(f"[INFO] Jumlah geometry kosong: {null_geom}")
    else:
        res = {
            "message": "Tidak menemukan tabel yang relevan.",
            "file_type": ext,
            "rows": 0,
            "columns": 0,
            "geometry_valid": 0,
            "geometry_empty": 0,
            "geometry_valid_percent": 0,
            "warnings": [],
            "warning_examples": [],
            "preview": []
        }

        return errorRes(message="Tidak berhasil mencocokan geometry pada tabel." ,details=res, status_code=422)

    result = result.replace([pd.NA, float('inf'), float('-inf')], None)
    if isinstance(result, gpd.GeoDataFrame) and 'geometry' in result.columns:
        result['geometry'] = result['geometry'].apply(
            lambda g: g.wkt if g is not None else None
        )

    empty_count = result['geometry'].apply(is_geom_empty).sum()
    valid_count = len(result) - empty_count
    match_percentage = (valid_count / len(result)) * 100

    warnings = []
    if empty_count > 0:
        warnings.append(
            f"{empty_count} dari {len(result)} baris tidak memiliki geometry yang valid "
            f"({100 - match_percentage:.2f}% data gagal cocok)."
        )

    if empty_count > 0:
        examples = result[result['geometry'].apply(is_geom_empty)].head(500)
        warning_examples = examples.to_dict(orient="records")
    else:
        warning_examples = []

    # preview_data = result.head(15).to_dict(orient="records")
    preview_data = result.to_dict(orient="records")

    preview_safe = [
        {k: safe_json(v) for k, v in row.items()} for row in preview_data
    ]

    warning_safe = [
        {k: safe_json(v) for k, v in row.items()} for row in warning_examples
    ]
    
    ai_context = {
        "nama_file_peta": filename,
        "nama_opd": "Badan Penanggulangan Bencana Daerah (BPBD) Provinsi Jatim",
        "tipe_data_spasial": geom_type,
        "deskripsi_singkat": fileDesc,
        "struktur_atribut_data": {},
    }
    ai_suggest = send_metadata(ai_context)
    # ai_suggest = {'judul': 'Peta Risiko Letusan Gunung Arjuna di Provinsi Jawa Timur', 'abstrak': 'Peta ini menggambarkan wilayah berisiko letusan Gunung Arjuna yang berada di Provinsi Jawa Timur. Data disajikan dalam bentuk poligon yang menunjukkan zona risiko berdasarkan analisis potensi aktivitas vulkanik.', 'tujuan': 'Data dapat digunakan untuk perencanaan mitigasi bencana dan pengambilan keputusan di wilayah Jawa Timur.', 'keyword': ['Risiko letusan', 'Gunung Arjuna', 'Bencana alam', 'Provinsi Jawa Timur', 'Geologi'], 'kategori': ['Geoscientific information', 'Environment'], 'kategori_mapset': 'Lingkungan Hidup'}

    tmp_file = generate_unique_filename()
    await asyncio.to_thread(process_dataframe_synchronous, result, tmp_file)

    response = {
        "message": "File berhasil dibaca dan dianalisis.",
        "file_name": filename,
        "file_type": ext,
        "rows": int(len(result)),
        "columns": list(map(str, result.columns)),
        "geometry_valid": int(valid_count),
        "geometry_empty": int(empty_count),
        "geometry_valid_percent": float(round(match_percentage, 2)),
        "geometry_type": geom_type,
        "warnings": warnings,
        "warning_rows": warning_safe,
        "preview": preview_safe,
        "metadata_suggest": ai_suggest,
        "tmp_path": tmp_file
    }
    
    # return successRes(content=response)
    return response










async def handle_upload_file(file: UploadFile = File(...), page: Optional[str] = Form(""), sheet: Optional[str] = Form(""), fileDesc: Optional[str] = Form("")):
    fname = file.filename
    ext = os.path.splitext(fname)[1].lower()
    contents = await file.read()
    size_mb = len(contents) / (1024*1024)
    if size_mb > MAX_FILE_MB:
        raise errorRes(status_code=413, message="Ukuran File Terlalu Besar")
    tmp_path = UPLOAD_FOLDER / fname
    with open(tmp_path, "wb") as f:
        f.write(contents)
    try:
        df = None
        print('ext', ext)

        if ext == ".csv":
            df = read_csv(str(tmp_path))
        elif ext == ".xlsx":
            df = read_csv(str(tmp_path), sheet)
        elif ext == ".mpk":
            df = read_mpk(str(tmp_path))
        elif ext == ".pdf":
            tbl = read_pdf(tmp_path, page)
            if len(tbl) == 0:
                res = {
                    "message": "Tidak ditemukan tabel valid pada halaman yang dipilih",
                    "tables": {},
                    "file_type": ext
                }
                return successRes(message="Tidak ditemukan tabel valid pada halaman yang dipilih", data=res)
            elif len(tbl) > 1:
                res = {
                    "message": "File berhasil dibaca dan dianalisis.",
                    "tables": tbl,
                    "file_type": ext
                }
                return successRes(data=res, message="File berhasil dibaca dan dianalisis.")
            else:
                df = convert_df(tbl[0])
        elif ext == ".zip":
            zip_type = detect_zip_type(str(tmp_path))

            if zip_type == "shp":
                print("[INFO] ZIP terdeteksi sebagai Shapefile.")
                df = read_shp(str(tmp_path))

            elif zip_type == "gdb":
                print("[INFO] ZIP terdeteksi sebagai Geodatabase (GDB).")
                df = read_gdb(str(tmp_path))

            else:
                return successRes(message="ZIP file tidak mengandung SHP / GDB valid.")
        else:
            raise errorRes(status_code=400, message="Unsupported file type")

        if df is None or (hasattr(df, "empty") and df.empty):
            return successRes(message="File berhasil dibaca, Tetapi tidak ditemukan tabel valid")
        
        res = await process_data(df, ext, fname, fileDesc)

        tmp_path.unlink(missing_ok=True)
        
        return successRes(data=res)

    except Exception as e:
        print(f"[ERROR] {e}")
        return errorRes(
            message="Internal Server Error",
            details=str(e),
            status_code=500
        )
    
    # finally:
    #     db_session.close()





class PdfRequest(BaseModel):
    title: str
    columns: List[str]
    rows: List[List]
    fileName: str
    fileDesc: str

async def handle_process_pdf(payload: PdfRequest):
    try:
        df = convert_df(payload.model_dump())
        if df is None or (hasattr(df, "empty") and df.empty):
            return errorRes(message="Tidak ada tabel")

        res = await process_data(df, '.pdf', payload.fileName, payload.fileDesc)        
        return successRes(data=res)
    
    except Exception as e:
        print(f"[ERROR] {e}")

        return errorRes(message="Internal Server Error", details= str(e), status_code=500)
    
    # finally:
    #     db_session.close()








class UploadRequest(BaseModel):
    title: str
    path: str
    rows: List[dict]
    columns: List[str]
    author: Dict[str, Any]
    style: str

# generate _2 if exist
async def generate_unique_table_name(base_name: str):
    base_name = base_name.lower().replace(" ", "_").replace("-", "_")
    table_name = base_name
    counter = 2

    async with engine.connect() as conn:
        while True:
            result = await conn.execute(
                text("SELECT to_regclass(:tname)"), 
                {"tname": table_name}
            )
            exists = result.scalar()

            if not exists:
                return table_name

            table_name = f"{base_name}_{counter}"
            counter += 1

def str_to_date(raw_date: str):
    if raw_date:
        try:
            return datetime.strptime(raw_date, "%Y-%m-%d").date()
        except Exception as e:
            print("[WARNING] Tidak bisa parse dateCreated:", e)
            return None




def generate_unique_filename(folder="tmp", ext="parquet", digits=6):
    os.makedirs(folder, exist_ok=True)
    while True:
        file_id = file_id = uuid.uuid4().int
        filename = f"{folder}/{file_id}.{ext}"
        
        if not os.path.exists(filename):
            return filename

def generate_job_id(user_id: str) -> str:
    timestamp = datetime.now().strftime("%Y%m%d%H%M%S")
    return f"{user_id}_{timestamp}"



def save_xml_to_sld(xml_string, filename):
    folder_path = 'style_temp'
    os.makedirs(folder_path, exist_ok=True)

    file_path = os.path.join(folder_path, f"{filename}.sld")

    with open(file_path, "w", encoding="utf-8") as f:
        f.write(xml_string)

    return file_path


async def process_parquet_upload(filename: str, table_name: str):
    from main import db_pool
    file_path = os.path.join("tmp", filename)
    
    if not os.path.exists(file_path):
        print(f"File {file_path} tidak ditemukan")
        return

    try:
        loop = asyncio.get_running_loop()
        df = await loop.run_in_executor(None, pd.read_parquet, file_path)

        # =====================================================================
        # 1. CLEANING NAMA KOLOM (PENTING!)
        # =====================================================================
        df.columns = [str(col).strip().upper() for col in df.columns]

        # Cek kolom GEOM (bisa GEOM atau geom setelah upper)
        # Kita standarkan nama kolom GEOM di DF menjadi "GEOM" untuk memudahkan logic
        if "GEOM" in df.columns:
            df.rename(columns={"GEOM": "GEOM"}, inplace=True)
            
        if "GEOM" not in df.columns:
            raise Exception("Kolom GEOM tidak ditemukan")

        # =====================================================================
        # 2. PERSIAPAN DATA (Row Processing)
        # =====================================================================
        clean_rows = []
        geom_types = set()
        
        # Ambil semua kolom atribut selain GEOM
        # Pastikan list ini yang dipakai untuk CREATE TABLE dan COPY (SINKRON)
        attr_columns = [col for col in df.columns if col != "GEOM"]
        
        for row in df.itertuples(index=False):
            # --- Handle GEOM ---
            raw_geom = getattr(row, "GEOM", None)
            if not raw_geom: continue

            try:
                geom = None
                if isinstance(raw_geom, str):
                    geom = wkt.loads(raw_geom)
                elif isinstance(raw_geom, bytes):
                    from shapely import wkb
                    geom = wkb.loads(raw_geom)
                
                if not geom: continue
                if not geom.is_valid: geom = geom.buffer(0)
                
                gtype = geom.geom_type.upper()
                if gtype == "POLYGON": geom = MultiPolygon([geom])
                elif gtype == "LINESTRING": geom = MultiLineString([geom])
                
                geom_types.add(geom.geom_type)
                ewkt = f"SRID=4326;{geom.wkt}"
                
            except Exception:
                continue

            # --- Handle Attributes (FORCE STRING) ---
            row_data = []
            for col in attr_columns:
                # getattr menggunakan nama kolom uppercase dari attr_columns
                val = getattr(row, col, None)
                if val is not None:
                    row_data.append(str(val)) # Convert int/float ke string
                else:
                    row_data.append(None)
            
            row_data.append(ewkt)
            clean_rows.append(tuple(row_data))

        if not clean_rows:
            raise Exception("Data valid kosong")

        # =====================================================================
        # 3. DATABASE OPERATIONS
        # =====================================================================
        final_geom_type = list(geom_types)[0].upper() if geom_types else "GEOM"
        if "MULTI" not in final_geom_type and final_geom_type != "GEOM":
             final_geom_type = "MULTI" + final_geom_type

        # A. BUILD DDL (CREATE TABLE)
        # Kita pakai f-string quotes f'"{col}"' agar di DB jadi UPPERCASE ("ID", "NAMA")
        col_defs = [f'"{col}" TEXT' for col in attr_columns]
        
        create_sql = f"""
            CREATE TABLE {table_name} (
                _id SERIAL PRIMARY KEY,    -- lowercase default
                {', '.join(col_defs)},     -- UPPERCASE (Hasil loop attr_columns)
                geom TEXT                  -- lowercase
            );
        """

        async with db_pool.acquire() as conn:
            # Drop table jika ada (untuk safety dev, production hati-hati)
            # await conn.execute(f"DROP TABLE IF EXISTS {table_name}")

            # 1. Create Table
            await conn.execute(create_sql)
            
            # 2. COPY Data
            # target_cols harus PERSIS sama dengan attr_columns
            # asyncpg akan meng-quote string ini otomatis ("ID", "NAMA", "geom")
            target_cols = attr_columns + ['geom']
            await conn.copy_records_to_table(
                table_name,
                records=clean_rows,
                columns=target_cols
            )
            
            # 3. Alter ke GEOM 2D
            alter_sql = f"""
                ALTER TABLE {table_name} 
                ALTER COLUMN geom TYPE geometry({final_geom_type}, 4326) 
                USING ST_Force2D(geom::geometry)::geometry({final_geom_type}, 4326);
                
                CREATE INDEX idx_{table_name}_geom ON {table_name} USING GIST (geom);
            """
            await conn.execute(alter_sql)

        print(f"Sukses upload {len(clean_rows)} baris ke {table_name}.")
        os.remove(file_path)

    except Exception as e:
        print(f"Error processing parquet: {e}")
        # Log error


async def handle_to_postgis(payload: UploadRequest, user_id: int = 2):
    try:
        table_name = await generate_unique_table_name(payload.title)
        # DataFrame
        df = pd.DataFrame(payload.rows)
        df.columns = [col.upper() for col in df.columns]
        if "GEOMETRY" not in df.columns:
            raise HTTPException(400, "Kolom GEOMETRY tidak ditemukan")

        # =====================================================================
        # 1. LOAD WKT → SHAPELY
        # =====================================================================
        def safe_load_wkt(g):
            if not isinstance(g, str):
                return None
            try:
                geom = wkt.loads(g)
                return geom
            except:
                return None

        df["GEOMETRY"] = df["GEOMETRY"].apply(safe_load_wkt)
        df = df.rename(columns={"GEOMETRY": "geom"})

        # =====================================================================
        # 2. DROP ROW geometry NULL
        # =====================================================================
        df = df[df["geom"].notnull()]
        if df.empty:
            raise HTTPException(400, "Semua geometry invalid atau NULL")

        # =====================================================================
        # 3. VALIDATE geometry (very important)
        # =====================================================================
        df["geom"] = df["geom"].apply(lambda g: g if g.is_valid else g.buffer(0))

        # =====================================================================
        # 4. SERAGAMKAN TIPE GEOMETRY (Polygon→MultiPolygon, Line→MultiLine)
        # =====================================================================
        def unify_geometry_type(g):
            gtype = g.geom_type.upper()
            if gtype == "POLYGON":
                return MultiPolygon([g])
            if gtype == "LINESTRING":
                return MultiLineString([g])
            return g  # sudah MULTI atau POINT
        df["geom"] = df["geom"].apply(unify_geometry_type)

        # =====================================================================
        # 5. DETEKSI CRS DARI METADATA / INPUT / DEFAULT
        # =====================================================================
        detected_crs = payload.author.get("crs")

        detected = payload.author.get("crs")
        print('crs', detected)

        if not detected_crs:
            detected_crs = "EPSG:4326"

        detected_crs = 'EPSG:4326'
        # Buat GeoDataFrame
        gdf = gpd.GeoDataFrame(df, geometry="geom", crs=detected_crs)
        row_count = len(gdf)

        # =====================================================================
        # 6. VERIFY CRS (SRID) VALID di PROJ / PostGIS
        # =====================================================================
        try:
            _ = gdf.to_crs(gdf.crs)   # test CRS valid
        except:
            raise HTTPException(400, f"CRS {detected_crs} tidak valid")

        # =====================================================================
        # 7. SIMPAN KE POSTGIS
        # =====================================================================
        job_id = generate_job_id(str(user_id))
        await process_parquet_upload(payload.path, table_name)

        # =====================================================================
        # 9. SIMPAN METADATA (geom_type, author metadata)
        # =====================================================================
        unified_geom_type = list(gdf.geom_type.unique())
        author = payload.author
        async with engine.begin() as conn:
            await conn.execute(text("""
                INSERT INTO backend.author_metadata (
                    table_title,
                    dataset_title,
                    dataset_abstract,
                    keywords,
                    topic_category,
                    date_created,
                    dataset_status,
                    organization_name,
                    contact_person_name,
                    contact_email,
                    contact_phone,
                    geom_type,
                    user_id,
                    process,
                    geometry_count
                ) VALUES (
                    :table_title,
                    :dataset_title,
                    :dataset_abstract,
                    :keywords,
                    :topic_category,
                    :date_created,
                    :dataset_status,
                    :organization_name,
                    :contact_person_name,
                    :contact_email,
                    :contact_phone,
                    :geom_type,
                    :user_id,
                    :process,
                    :geometry_count
                )
            """), {
                "table_title": table_name,
                "dataset_title": payload.title,
                "dataset_abstract": author.get("abstract"),
                "keywords": author.get("keywords"),
                "topic_category": ", ".join(author.get("topicCategory")),
                "date_created": str_to_date(author.get("dateCreated")),
                "dataset_status": author.get("status"),
                "organization_name": author.get("organization"),
                "contact_person_name": author.get("contactName"),
                "contact_email": author.get("contactEmail"),
                "contact_phone": author.get("contactPhone"),
                "geom_type": json.dumps(unified_geom_type),
                "user_id": user_id,
                "process": 'CLEANSING',
                "geometry_count": row_count
            })

        # =====================================================================
        # 10. LOGGING
        # =====================================================================
        await log_activity(
            user_id=user_id,
            action_type="UPLOAD",
            action_title=f"Upload dataset {table_name}",
            details={"table_name": table_name, "rows": len(gdf)}
        )

        result = {
            "job_id": job_id,
            "job_status": "wait",
            "table_name": table_name,
            "status": "success",
            "message": f"Tabel '{table_name}' berhasil dibuat.",
            "total_rows": len(gdf),
            "geometry_type": unified_geom_type,
            "crs": detected_crs,
            "metadata_uuid": ""
        }
        save_xml_to_sld(payload.style, job_id)

        cleansing = await query_cleansing_data(table_name)
        result['job_status'] = cleansing

        publish = await publish_layer(table_name, job_id)
        result['metadata_uuid'] = publish['uuid']

        mapset = {
            "name": payload.title,
            "description": author.get("abstract"),
            "scale": "1:25000",
            'projection_system_id': '0196c746-d1ba-7f1c-9706-5df738679cc7', 
            "category_id": author.get("mapsetCategory"),
            "data_status": "sementara",
            'classification_id': '01968b4b-d3f9-76c9-888c-ee887ac31ce4', 
            'producer_id': '01968b54-0000-7a67-bd10-975b8923b93e', 
            "layer_type": unified_geom_type[0],
            'source_id': ['019c03ef-35e1-738b-858d-871dc7d1e4d6'], 
            "layer_url": publish['geos_link'],
            "metadata_url": f"{GEONETWORK_URL}/srv/eng/catalog.search#/metadata/{publish['uuid']}",
            "coverage_level": "provinsi",
            "coverage_area": "kabupaten",
            "data_update_period": "Tahunan",
            "data_version": "2026",
            "is_popular": False,
            "is_active": True,
            'regional_id': '01968b53-a910-7a67-bd10-975b8923b92e', 
            "notes": "Mapset baru dibuat",
            "status_validation": "on_verification",
        }

        await upload_to_main(mapset)

        return successRes(data=result)

    except Exception as e:
        await log_activity(
            user_id=user_id,
            action_type="ERROR",
            action_title="Upload gagal",
            details={"error": str(e)}
        )
        raise HTTPException(status_code=500, detail=str(e))