app.py

import gradio as gr
import subprocess
import pandas as pd

'''
import gradio as gr

def greet(name):
    return "Hello " + name + "!"

demo = gr.Interface(fn=greet, inputs="text", outputs="text")
demo.launch()
'''
def greet(name1, name2):
    # Storing each input in a variable, you can process or save them as you like
    str1_openai = name1 ## openai
    str2_bioportal = "213e22ba-4c3b-402b-bd36-6e9d4e86b1b5"   #bioportal
    str3_huggingface = "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"   # huggingface
    str4_input = name2
    
    
    with open('abstractsave.txt', 'w') as f:
        f.write(str4_input)
    
    def run_command(command):
        result = subprocess.run(command, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)

        output_string = result.stdout
        error_string = result.stderr

        return output_string, error_string

#####     output_string1, error_string1= run_command("pip install optogpt")
    output_string1, error_string1 = run_command("curl -sSL https://install.python-poetry.org | python3 -")
    output_string2, error_string2 = run_command(f"poetry run runoak set-apikey -e openai {str1_openai}")
    run_command(f"poetry run runoak set-apikey -e bioportal {str2_bioportal}")
    run_command(f"poetry run runoak set-apikey -e hfhub-key {str3_huggingface}")
#####     output = run_command(f"ontogpt extract -t gocam.GoCamAnnotations -i ./abstract.txt")
    output = run_command(f"cancerontogpt extract -t cancer.CancerAnnotations -i ./abstractsave.txt")
    
    output = output[0].replace('\\n', '\n')
    
    # Find the positions of the start and end markers
    start_marker = "raw_completion_output: |-"
    end_marker = "prompt: "
    start_position = output.find(start_marker)
    end_position = output.find(end_marker)

    # Extract the text between the start and end positions
    output = output[start_position + len(start_marker):end_position].strip()


    # Splitting the data string into lines and then split each line into key-value pairs
    key_value_pairs = [line.split(": ", 1) for line in output.split("\n")[1:] if line.strip()]

    def format_identifier(identifier: str) -> str:
        # Split the string by underscores
        words = identifier.strip().split('_')

        # Capitalize each word
        capitalized_words = [word.capitalize() for word in words]

        # Join the words with spaces
        formatted_identifier = ' '.join(capitalized_words)

        return formatted_identifier

    key_value_pairs = [[format_identifier(x[0]), *x[1:]] for x in key_value_pairs]
    
    # Convert the key-value pairs into a table format (a list of lists)
    df_pred = pd.DataFrame(key_value_pairs, columns = ["Ontology Attribute", "Value"]).iloc[:19,:]
    
    
    
    
#     data = {
#         "Name": ['fawef', 'fseaf', 'asef'],
#         "Age": [30, 25, 35],
#         "City": ["New York", "San Francisco", "Los Angeles"]
#     }
#     df = pd.DataFrame(data)
    return df_pred.to_html()

####     output_string1, error_string1=run_command("poetry")# ontogpt")

    
#     return location
    # For the purpose of this example, I'm just returning the values concatenated
#     return f"Inputs received: {str1} \n, {str2}, {str3}, {str4}, '--------------',   '--------------', {output_string1},{error_string1},{output_string2},{error_string2},{output}"
#     #     return location
    # For the purpose of this example, I'm just returning the values concatenated
#     return f"{str4_input}"

# Define 5 text input boxes with labels
input_boxes = [
    gr.inputs.Textbox(label="openai api key"),
    gr.inputs.Textbox(lines=20, label="Input cencer report", placeholder='Type text here...'),
]

# iface = gr.Interface(fn=greet, inputs=input_boxes, outputs="text")
iface = gr.Interface(fn=greet, inputs=input_boxes, outputs=gr.outputs.HTML(label="Output Table"),examples=[                                    # Sample text examples
        ["this is a sample text"],
        ["gradio is great"]
    ])

iface.launch()