For Python Developers

Native Python client for accessing the UnDatasIO Platform

Python Client

Overview

The UnDatasIO Client for Python enables easy interaction with the UnDatasIO API, allowing developers to execute queries and mutations against the UnDatasIO platform.

Prerequisites

Before you begin, ensure you have the following:

• Python 3.8 or later is installed on your system.

• An active account on the UnDatasIO Platform with access to the API Settings page.

Get your TOKEN

Installation

To install the UnDatasIO Client, use pip

pip install UnDatasIO

Configuration

Please obtain your token in the User Center. If task_name is not filled in, PdfParserDemo will be used as the default task_name.

from undatasio.undatasio import UnDatasIO

token = 'Your API token'
task_name = 'your task name'

# 1. Initialize the UnDatasIO client
client = UnDatasIO(token=token, task_name=task_name)

Upload file

# 2. Upload files
upload_response = client.upload(file_dir_path='./example_files')
if upload_response.code == 200:
    print("File upload successful!")
else:
    print(f"File upload failed: {upload_response.msg}")

View all uploaded files

# 3. View all uploaded files
upload_filename_response = client.show_upload()
if upload_filename_response.code == 200:
    print(upload_filename_response.data)
else:
    print(f"File upload failed: {upload_filename_response.msg}")

Parse the specified PDF file

# 4. Parse files
# All parameter:['fast', 'accurate']
# All language:{'English': 'en', '한국어': 'korean', '日本語': 'japan', '中文': 'ch', 
# '繁體中文': 'chinese_cht', 'Deutsch': 'de', 'Español': 'es', 'Bahasa Indonesia': 'id',
# 'Français': 'fr', 'Português': 'pt', 'Italiano': 'it', 'Türkçe': 'tr', 
# 'Polski': 'pl', 'Čeština': 'cs', 'Dansk': 'da', 'हिंदी': 'hi', 'தமிழ்': 'ta', 
# 'తెలుగు': 'te', 'ಕನ್ನಡ': 'ka', 'Afrikaans': 'af', 'Azərbaycan': 'az', 
# 'Bosanski': 'bs', 'Cymraeg': 'cy', 'Eesti': 'et', 'Gaeilge': 'ga', 
# 'Hrvatski': 'hr', 'Magyar': 'hu', 'Íslenska': 'is', 'Kurdî': 'ku', 'Latina': 'la', 
# 'Lietuvių': 'lt', 'Latviešu': 'lv', 'Māori': 'mi', 'Bahasa Melayu': 'ms', 
# 'Malti': 'mt', 'Nederlands': 'nl', 'Norsk': 'no', 'Occitan': 'oc', 'पालि': 'pi', 
# 'Română': 'ro', 'Srpski': 'rs_latin', 'Slovenčina': 'sk', 'Slovenščina': 'sl', 
# 'Shqip': 'sq', 'Svenska': 'sv', 'Kiswahili': 'sw', 'Tagalog': 'tl', "O'zbek": 'uz', 
# 'Tiếng Việt': 'vi'}
parse_response = client.parser(
                        file_name_list=['example_file1.pdf', 'example_file2.pdf'],
                        lang='en',
                        parameter='fast'
                        )
if parse_response.code == 200:
    print("File parsing successful")
else:
    print(f"File parsing request failed: {parse_response.msg}")

View version information for all parsed results

# 5. View historical parsing results
parse_filename_response = client.show_version()
if parse_filename_response.code == 200:
    print(parse_filename_response.data)
else:
    print(f"File upload failed: {parse_filename_response.msg}")

View the parsed result

# 6. View parsing results (assuming you know the version number is 'v1' and want to get the title and table information in the parsing results)
# All types：['title', 'table', 'text', 'image', 'interline_equation']
results = client.get_result_type(type_info=['title', 'table'], file_name='example_file.pdf', version='v1')
if results.code == 200:
    print(f"Parsing results: {results.data}")
else:
    print(f"Failed to get parsing results: {results.msg}")