import os
from batchgenerators.utilities.file_and_folder_operations import join, load_json, isfile, save_json, maybe_mkdir_p
import csv

with open('/storage/chenqi/X_data/annotatedtumor_txt/colon_tumor_cases.txt', 'r') as f:
    all_files=f.readlines()
all_files = [os.path.basename(i.split(' ')[0]).split('.')[0] for i in all_files]
all_files = [i.split('\n')[0] for i in all_files]

with open('/storage/chenqi/code/DiffTumor/STEP3.SegmentationModel/cross_eval/colon_tumor_data_fold/real_tumor_val_1.txt', 'r') as f:
    val_liver=f.readlines()
val_liver = [os.path.basename(i.split(' ')[0]).split('.')[0] for i in val_liver]

cases_info = [*csv.DictReader(open('final_mapping_qichen.csv'))]

val_bdmap=[]
for i in cases_info:
    if i['source_dataset'] == 'MSD-Colon':
        for j in val_liver:
            if j == i['source_id']: 
                val_bdmap.append(i['AbdomenAtlas_id'])


breakpoint()
write_content = []
train_liver = list(set(all_files) - set(val_bdmap))
write_content.append({"train":train_liver, "val":val_bdmap})

# breakpoint()

splits_file = 'splits_final_colon_bdmap.json'
save_json(write_content, splits_file)