modules/batch_processor.py

"""
Batch Processing Module — NEW feature
Process multiple documents in one run.
Returns aggregated entity list + per-document breakdown.
"""

import os
from modules.doc_reader import read_document, chunk_text
from modules.ner_engine  import run_ner
from modules.analytics   import label_counts


def process_single(path: str) -> dict:
    """Process one document. Returns dict with text, entities, meta."""
    text, meta = read_document(path)
    chunks     = chunk_text(text, max_chars=400)

    all_entities = []
    char_offset  = 0

    for chunk in chunks:
        ents = run_ner(chunk)
        # Adjust character offsets to be relative to full document
        for e in ents:
            e["start"] += char_offset
            e["end"]   += char_offset
        all_entities.extend(ents)
        char_offset += len(chunk) + 1   # +1 for the space between chunks

    return {
        "path":     path,
        "meta":     meta,
        "text":     text,
        "entities": all_entities,
        "counts":   label_counts(all_entities),
    }


def process_batch(paths: list[str]) -> list[dict]:
    """Process a list of file paths. Returns list of results."""
    results = []
    for path in paths:
        if not os.path.exists(path):
            continue
        try:
            result = process_single(path)
            results.append(result)
        except Exception as e:
            results.append({
                "path":  path,
                "error": str(e),
            })
    return results


def aggregate_entities(batch_results: list[dict]) -> list[dict]:
    """Combine entities from all documents in a batch."""
    combined = []
    for res in batch_results:
        if "entities" in res:
            combined.extend(res["entities"])
    return combined