data_processing.py

import pandas as pd
from typing import List, Tuple, Dict, Any

REQUIRED_BASE_COLUMNS = [
    "NodeID","NodeName","NeighborNodeID"
]

OPTIONAL_NUMERIC_GUESSES = ["TransitionCost","WaitingTime"]


def load_dataset(path: str) -> pd.DataFrame:
    df = pd.read_csv(path)
    # Normalize column names (strip spaces)
    df.columns = [c.strip() for c in df.columns]
    # Type coercion attempts
    for col in ["NodeID","NeighborNodeID"]:
        if col in df.columns:
            df[col] = pd.to_numeric(df[col], errors='coerce')
    for col in df.columns:
        if col not in ("NodeName",):
            # Try numeric
            try:
                df[col] = pd.to_numeric(df[col])
            except Exception:
                pass
    return df


def validate_dataset(df: pd.DataFrame) -> Tuple[bool, List[str]]:
    errors = []
    for col in REQUIRED_BASE_COLUMNS:
        if col not in df.columns:
            errors.append(f"Missing required column: {col}")
    if df['NodeID'].isna().any():
        errors.append("Some NodeID values are NaN after coercion.")
    # Allow NeighborNodeID NaN for terminal nodes
    return len(errors) == 0, errors


def list_numeric_columns(df: pd.DataFrame) -> List[str]:
    numeric_cols = []
    for col in df.columns:
        if col in ("NodeID","NeighborNodeID"):
            continue
        if pd.api.types.is_numeric_dtype(df[col]):
            numeric_cols.append(col)
    return numeric_cols


def build_edge_records(df: pd.DataFrame) -> List[Dict[str, Any]]:
    records = []
    for _, row in df.iterrows():
        src = int(row['NodeID']) if not pd.isna(row['NodeID']) else None
        dst = row.get('NeighborNodeID', None)
        if pd.isna(dst) or dst == '' or dst is None:
            continue
        dst = int(dst)
        rec = {k: row[k] for k in df.columns if k not in ('NodeID','NeighborNodeID')}
        rec['source'] = src
        rec['target'] = dst
        records.append(rec)
    return records


def list_nodes(df: pd.DataFrame):
    node_info = {}
    for _, row in df.iterrows():
        nid = int(row['NodeID']) if not pd.isna(row['NodeID']) else None
        if nid is None:
            continue
        node_info.setdefault(nid, {
            'name': row.get('NodeName', str(nid))
        })
    # Also include neighbor-only nodes
    for _, row in df.iterrows():
        dst = row.get('NeighborNodeID', None)
        if pd.isna(dst) or dst == '' or dst is None:
            continue
        dst = int(dst)
        node_info.setdefault(dst, {'name': str(dst)})
    return node_info