def fold()

in src/backend/main.py [0:0]
67 lines of code
9 McCabe index (conditional complexity)

def fold():
    user_info = valid_user()
    if user_info is not None:
        #  check if the post request has the file part
        if 'file' not in request.files:
            flash('No file part')
            return Response('{"status":"uploaded file missing."}',status=400, mimetype='application/json')
        f = request.files['file']

        timestamp = datetime.now().strftime("%d/%m/%Y %H:%M:%S")
        form = dict(request.form)
        print(f"Folding request received on {timestamp}, with parameters {form}")
        # Saving the received file to local disk

        filename = secure_filename(f.filename )
        print(f"The received filename was: {filename}")

        # Copying the received file to cloud storage and saving the path
        gcs_sequence_path = f'fasta/{filename}' # TODO add username subfolder

        # save file locally, for fasta validation purpose
        save_file_locally(f, filename)
        gcs_path = upload_to_bucket(gcs_sequence_path,filename,BUCKET_NAME)
        print(f"Protein file uploaded to: {gcs_path}")

        # is_monomer, sequences = fasta_utils.validate_fasta_file_gcs(gcs_path,gcs_sequence_path,BUCKET_NAME,storage_client)
        is_monomer, sequences = fasta_utils.validate_fasta_file(filename)
        # TODO delete local disk fasta file after validate
        print(f'IS MONOMER {is_monomer}\n{sequences}')

        # Capturing the running parameters of the pipeline
        params = {
        'sequence_path': gcs_path,
        'max_template_date': '2030-01-01',
        'use_small_bfd': True if form["smallBFD"] == "yes" else False,
        'num_multimer_predictions_per_model': int(form["predictionCount"]),
        'is_run_relax': 'relax' if str(form["relaxation"]).lower() == "yes" else '', 
        'model_preset': str(form["proteinType"]).lower(),
        'project': PROJECT_ID,
        'region': REGION
        }

        os.environ['PREDICT_MACHINE_TYPE'] = str(form["predictMachineType"]).lower()
        os.environ['PREDICT_ACCELERATOR_COUNT'] = str(form["acceleratorCount"]).lower()
        os.environ['PREDICT_ACCELERATOR_TYPE'] = decide_accelerator_type(str(form["predictMachineType"]))
        
        os.environ['RELAX_MACHINE_TYPE'] = str(form["relaxMachineType"]).lower()
        os.environ['RELAX_ACCELERATOR_COUNT'] = str(form["relaxAcceleratorCount"]).lower()
        os.environ['RELAX_ACCELERATOR_TYPE'] = decide_accelerator_type(str(form["relaxMachineType"]))

        os.environ['ALPHAFOLD_COMPONENTS_IMAGE'] = IMAGE_URI
        os.environ['NFS_SERVER'] = FILESTORE_IP
        os.environ['NFS_PATH'] = FILESTORE_SHARE
        os.environ['NETWORK'] = FILESTORE_NETWORK
        os.environ['MODEL_PARAMS_GCS_LOCATION'] = MODEL_PARAMS
        os.environ['PARALLELISM'] = '5'
        
        # Compile the pipeline
        from pipelines.alphafold_inference_pipeline import alphafold_inference_pipeline as pipeline
        run_tag = str(form["runTag"]).lower()
        experiment_id = str(form["experimentId"]).lower()
        pipeline_name = f'universal-pipeline-{experiment_id}'
        compiler.Compiler().compile(
            pipeline_func=pipeline,
            package_path=f'{pipeline_name}.json')

        # Run FOLD
        # Running the existing pipeline
        labels = {'run_tag': run_tag
                 ,'experiment_id': experiment_id
                 , 'sequence_id': filename.split(sep='.')[0].lower()
                 , 'user': f'{user_info["given_name"].lower()}_{user_info["family_name"].lower()}'
                 }

        pipeline_job = vertex_ai.PipelineJob(
            display_name=pipeline_name,
            template_path=f'{pipeline_name}.json', # pipeline name need to be unique, and subfolders in GCS
            pipeline_root=f'gs://{BUCKET_NAME}/pipeline_runs/{pipeline_name}',
            parameter_values=params,
            enable_caching=True,
            labels=labels)
        pipeline_job.run(sync=False)
        pipeline_job.wait_for_resource_creation()


        # Log folding job
        message = f"Folding started for experiment ID {experiment_id} with parameters {params}"
        print(f'/fold MESAGE {message}')
        return Response("status: folding is in progress... }", 
                        status=200,mimetype='application/json')
    else:
        return Response("{'status':'Unauthorized'}", status=401, mimetype='application/json')