OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Documentsearch

Documentsearch is a Python application that loads PDF and DOCX files from a folder and uses Sentence Transformers and ChatGLM-6B to answer questions, with a command-line example and a Streamlit web interface.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Documentsearch — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to explore question answering over a folder of PDF and DOCX documents and can work with its Python examples or Streamlit interface.

Who it is for

  • Developers who want to search and ask questions about a folder of PDF and DOCX documents.
  • People exploring a document-search workflow for collections such as policy archives.

Documented uses

  • Load PDF and DOCX files from a selected folder.
  • Submit a question about the collection through the command-line example.
  • Launch a Streamlit interface configured for a document folder.
Before you choose

Requirements & limits

The repository is public, open source, and licensed under Apache-2.0.

  • The README supports PDF and DOCX files, but explicitly excludes DOC files.
  • The web interface requires changing the global_dir setting in web_ui.py before launch.
  • The repository provides a CLI example and a Streamlit launch command.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set a document folder and launch the Streamlit interface

Configure the web interface to use a folder of PDF and DOCX documents, then launch Streamlit.

Before you start

  • The Documentsearch project files.
  • A folder containing PDF and/or DOCX files.

Steps

  1. Edit the global_dir setting in web_ui.py to point to the document folder.

  2. Run streamlit run web_ui.py --server.fileWatcherType none from the project environment.

Expected outcome

The project’s Streamlit web interface is launched with the configured document folder.

Guide limitations

  • The README lists PDF and DOCX support and says DOC files are not supported.
Documentation 1 ↗
Source-based practical guide

Query a document folder from the command-line example

Use the provided Python example to ask a question about documents in a configured folder.

Before you start

  • The Documentsearch project files.
  • A folder of PDF and/or DOCX documents.

Steps

  1. In demo.py, set global_dir to the document folder.

  2. Create a KnowLedge instance with KnowLedge(global_dir=global_dir).

  3. Call search_result with a question string and print the returned result and data.

Expected outcome

The example prints the result and data returned for the supplied question.

Guide limitations

  • The README lists PDF and DOCX support and says DOC files are not supported.
  • The README provides an example question but does not detail the returned data format.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
yuanzhoulvpi2017/DocumentSearch · Official homepage title · retained in its source language
How we review sourcesSuggest a correction