Apps quick start

API_TOKEN = '<YOUR-API-TOKEN>' # Enter your API token.
API_ROOT = 'https://aihub.instabase.com/api'
API_HEADERS = {
    'Authorization': f'Bearer {API_TOKEN}',
    'IB-Context': '<ORGANIZATION-ID or USER-ID>' # Enter ORGANIZATION_ID to use your commercial account. Omit or use USER_ID to use your community account.
}

INPUT_FILEPATHS = ['<../sample1.pdf>', '<sample2.docx>', '<inner/folder/sample3.png>']  # Replace with file paths for the input files to be uploaded.
def create_batch(batch_name, workspace_name=None):
  resp = requests.put(f'{API_ROOT}/v2/batches', headers=API_HEADERS, data=json.dumps({'batch_name': batch_name, 'workspace_name': workspace_name }))
  return resp.json()['batch_id']

def upload_input_file(local_filepath, batch_id, filename):
  with open(local_filepath, 'rb') as input_file:
    requests.put(f'{API_ROOT}/v2/batches/{batch_id}/files/{filename}',
                 headers=API_HEADERS,
                 data=input_file.read())

BATCH_ID = create_batch('<YOUR-BATCH-NAME>', '<WORKSPACE-NAME>')  # Define a name for the batch and, optionally, specify a workspace in which to create the batch. If WORKSPACE-NAME is not defined, the default is accepted.
for input_filepath in FILENAMES:
  upload_input_file(input_filepath, BATCH_ID, os.path.basename(input_filepath))

APP_NAME = '<EXAMPLE-APP-NAME>'  # Enter the app name.

# Step 1: Run app
run_app_payload = json.dumps({
      'batch_id': BATCH_ID, # BATCH_ID from previous step
      'name': APP_NAME
      # optional: 'input_dir': '<file path>', If using 'input_dir' to specify a file path, remove 'batch_id': BATCH_ID. 
      # optional: 'output_workspace': '<workspace-name>', if not specified, default is used.
	  # optional: 'owner': '<owner-name>' If not specified, default is your user ID.
    })

url = f'{API_ROOT}/v2/zero-shot-idp/projects/app/run'
run_app_resp = requests.post(url, headers=API_HEADERS, data=run_app_payload)

job_id = run_app_resp.json().get('job_id')

url = f'{API_ROOT}/v1/jobs/status?type=flow&job_id={job_id}'

job_status = ''
while job_status != 'DONE':
    time.sleep(5)
    job_status_resp = requests.get(url, headers=API_HEADERS)
    job_status = job_status_resp.json().get('state')

# Now the job is complete, get the results
output_folder_path = job_status_resp.json()['results'][0]['output_folder']

results_payload = json.dumps({
        'file_offset': 0, # Results are paginated with a limit of 20 files per page. This parameter needs to be updated across requests for a job with >20 files.
        'ibresults_path': f'{output_folder_path}/batch.ibflowresults'
})

url = f'{API_ROOT}/v1/flow_binary/results'
results_resp = requests.post(url, headers=API_HEADERS, data=results_payload)

import requests
import time
import json

API_TOKEN = '<YOUR-API-TOKEN>' # Enter your API token.
APP_NAME = '<EXAMPLE-APP-NAME>' # Enter your app name
INPUT_FILEPATHS = ['<../sample1.pdf>', '<sample2.docx>', '<inner/folder/sample3.png>']  # Replace with file paths for the input files to be uploaded.

API_ROOT = 'https://aihub.instabase.com/api'
API_HEADERS = {
    'Authorization': f'Bearer {API_TOKEN}',
    'IB-Context': '<ORGANIZATION-ID or USER-ID>' # Enter ORGANIZATION_ID to use your commercial account. Omit or use USER_ID to use your community account.
}

def create_batch(batch_name, workspace_name=None):
    resp = requests.put(f'{API_ROOT}/v2/batches', headers=API_HEADERS, data=json.dumps({'batch_name': batch_name, 'workspace_name': workspace_name }))
    return resp.json()['batch_id']

def upload_input_file(local_filepath, batch_id, filename):
    with open(local_filepath, 'rb') as input_file:
        requests.put(f'{API_ROOT}/v2/batches/{batch_id}/files/{filename}',
                        headers=API_HEADERS,
                        data=input_file.read())

def run_app(batch_id, app_name):
    run_app_payload = json.dumps({
      'batch_id': batch_id,
      'name': APP_NAME,
      # optional: 'input_dir': '<file path>, if using 'input_dir' to specify a file path, omit 'batch_id': BATCH_ID.
      # optional: 'output_workspace': '<workspace-name>', if not specified, default is used.
      # optional: 'owner': '<owner-name>', if not specified, default is your user ID.
    })

    url = f'{API_ROOT}/v2/zero-shot-idp/projects/app/run'
    run_app_resp = requests.post(url, headers=API_HEADERS, data=run_app_payload)

    job_id = run_app_resp.json().get('job_id')
    return job_id

def get_results(job_id):
    # Using the Job ID from the app run, gets the current status. If complete, gets the results.
    url = f'{API_ROOT}/v1/jobs/status?type=flow&job_id={job_id}'

    job_status = ''
    while job_status != 'DONE':
        time.sleep(5)
        job_status_resp = requests.get(url, headers=API_HEADERS)
        job_status = job_status_resp.json().get('state')

    # Now the job is complete, gets the results.
    output_folder_path = job_status_resp.json()['results'][0]['output_folder']

    results_payload = json.dumps({
        'file_offset': 0,
        'ibresults_path': f'{output_folder_path}/batch.ibflowresults'
    })

    url = f'{API_ROOT}/v1/flow_binary/results'
    results_resp = requests.post(url, headers=API_HEADERS, data=results_payload)

    return results_resp.json()

if __name__ == "__main__":
    # 1. Creates a new batch.
    batch_id = create_batch('<YOUR-BATCH-NAME>', '<WORKSPACE-NAME>') # Define a name for the batch and, optionally, specify a workspace in which to create the batch. If WORKSPACE-NAME is not defined, the default is accepted.
    # 2. Uploads files into batch.
    for input_filepath in INPUT_FILEPATHS:
        upload_input_file(input_filepath, batch_id, os.path.basename(input_filepath))
    # 3. Runs app using batch as input.
    job_id = run_app(batch_id, APP_NAME)
    # 4. Gets results of app run.
    results = get_results(job_id)