Tika REST Client

Python client for Apache Tika REST endpoints.

Overview

tika_rest_client provides a small, typed interface for common Apache Tika operations:

• server health/version checks,
• language identification,
• text extraction (plain text or paginated),
• metadata extraction.

It wraps HTTP/network failures with library-specific exceptions and returns structured response models.

Installation

pip install tika-rest-client

Requirements

• Python 3.12+
• A running Apache Tika server (for example on http://localhost:8080)

Quick Start

import io
from tika_rest_client import TikaRestClient

# IMPORTANT:
# In the current implementation, `host` is used as the full base URL.
# Example: "http://localhost:8080"
client = TikaRestClient(host="http://localhost:8080")

# Health check
print(client.check_connection())  # True / False

# Version
version = client.check_version()
print(version.response)  # e.g. "3.4.4"

# Language detection
lang = client.identify_language("Hello world")
print(lang.response)  # "en"

# Metadata extraction
metadata = client.extract_file_metadata(io.BytesIO(b"example content"))
print(metadata.metadata.general.content_type)

Main API

check_connection()

Returns True if the Tika server responds with HTTP 200 on the version endpoint, otherwise False.

check_version()

Returns a PlainTextResponse containing the server version. Raises TikaRestClientNetworkException on HTTP errors.

identify_language(input_to_test)

Accepts one of:

• str
• bytes
• buffered file-like object (io.BytesIO, BinaryIO)

Returns PlainTextResponse. Raises:

• InputFormatNotAvailableForThisEndpoint for unsupported input types,
• LanguageIdentificationError when a language detection request fails.

extract_file_content(file_content, paginate, boilerplate, ocr)

Extracts textual content from binary file data.

• If paginate=False: returns PlainTextResponse.
• If paginate=True: returns DocumentDividedInMatrixResponse.

Parameters:

• file_content: file bytes wrapped in io.BytesIO
• paginate: enable page-oriented output model
• boilerplate: use boilerplate endpoint variant
• ocr: enable OCR header for Tika

extract_file_metadata(content)

Extracts metadata and returns TikaMetadataResponse.

Response Models

Depending on the method, the client returns:

• PlainTextResponse
• DocumentDividedInMatrixResponse
• TikaMetadataResponse

Each response includes client_info plus method-specific payload.

Error Handling

Library-specific exceptions include:

• InputFormatNotAvailableForThisEndpoint
• LanguageIdentificationError
• TikaRestClientNetworkException

Example:

from tika_rest_client.exception import TikaRestClientNetworkException

try:
    print(client.check_version().response)
except TikaRestClientNetworkException:
    print("Unable to reach Tika server.")

Contributing

Contributions are welcome.

When opening a pull request:

1. Keep style consistent with the project.
2. Add or update tests when behavior changes.
3. Document public API changes.

License

This project is licensed under the Creative Commons CC-BY-SA