Benign python Download

langchain-xberg 1.1.0

“LangChain document loader for Xberg — extract 88+ file formats into Documents with async, batching, chunking, and rich metadata”

Legacy LangChain package pathimports Path from pathlib

SHA-25655cd9f37aabc1facdd2f49cae5872322617e9e16c193d6bcbc6ac20e86f28e35

URLhttps://files.pythonhosted.org/packages/f3/de/4ccd6001ab1de0355cdd8bcc627d72834dae710cd2dc7ad2db7889e80fb9/langchain_xberg-1.1.0-py3-none-any.whl

MaleculeH₂(FPo₂)

Evidence

imports Path from pathlib lines 1–20

1"""Xberg document loader for LangChain."""
2
3from __future__ import annotations
4
5import asyncio
6from concurrent.futures import ThreadPoolExecutor
7from pathlib import Path
8from typing import TYPE_CHECKING, Any
9
10from langchain_core.document_loaders import BaseLoader
11from langchain_core.documents import Document
12from xberg import ExtractInput, XbergError, extract, extract_batch
13
14if TYPE_CHECKING:
15    from collections.abc import AsyncIterator, Callable, Coroutine, Iterator
16
17    from xberg import ExtractedDocument, ExtractionConfig, ExtractionResult
18
19# Metadata fields (from xberg.Metadata) that carry JSON-friendly scalar/list values.
20# Complex nested fields (pages, format, image_preproces …

References the PDF signature or media type lines 62–70

29… rgLoader(file_path=["a.pdf", "b.docx", "c.txt"])
          >>> docs = loader.load()
64
      Load from bytes:
          >>> loader = XbergLoader(data=raw_bytes, mime_type="application/pdf")
          >>> docs = loader.load()
68
      Load a directory:
          >>> loader = XbergLoader(file_path="./docs/", glo …

Python getattr dynamic attribute lookup lines 152–163

⋯5 lines
      if isinstance(config, dict):
          return config.get(name)
      return getattr(config, name, None)
160
  @property
  def _per_page(self) -> bool:
      """Whether per-page splitting is enabled in the con …

Calls Python glob for file enumeration lines 183–189

29… lf._glob or "**/*"
              yield from sorted(p for p in file_path.glob(pattern) if p.is_file())
          else:
              yield file_path
187
  def _is_batch(self) -> bool:
      """Whether …

Submits a task to an executor lines 392–400

24… _running_loop()
      except RuntimeError:
          return asyncio.run(factory())
395
      with ThreadPoolExecutor(max_workers=1) as pool:
          return pool.submit(lambda: asyncio.run(factory())).result()
398
  def lazy_load(self) -> Iterator[Document]:
      """Load documents lazily, yiel …

No evidence locations were recorded for this file. Raw result