# Path Configuration
from tools.preprocess import *

# Processing context
trait = "Sarcoma"

# Input paths
tcga_root_dir = "../DATA/TCGA"

# Output paths
out_data_file = "./output/preprocess/1/Sarcoma/TCGA.csv"
out_gene_data_file = "./output/preprocess/1/Sarcoma/gene_data/TCGA.csv"
out_clinical_data_file = "./output/preprocess/1/Sarcoma/clinical_data/TCGA.csv"
json_path = "./output/preprocess/1/Sarcoma/cohort_info.json"

import os
import pandas as pd

# List of subdirectories provided in the instructions:
subdirectories = [
    'CrawlData.ipynb', '.DS_Store', 'TCGA_lower_grade_glioma_and_glioblastoma_(GBMLGG)',
    'TCGA_Uterine_Carcinosarcoma_(UCS)', 'TCGA_Thyroid_Cancer_(THCA)', 'TCGA_Thymoma_(THYM)',
    'TCGA_Testicular_Cancer_(TGCT)', 'TCGA_Stomach_Cancer_(STAD)', 'TCGA_Sarcoma_(SARC)',
    'TCGA_Rectal_Cancer_(READ)', 'TCGA_Prostate_Cancer_(PRAD)', 'TCGA_Pheochromocytoma_Paraganglioma_(PCPG)',
    'TCGA_Pancreatic_Cancer_(PAAD)', 'TCGA_Ovarian_Cancer_(OV)', 'TCGA_Ocular_melanomas_(UVM)',
    'TCGA_Mesothelioma_(MESO)', 'TCGA_Melanoma_(SKCM)', 'TCGA_Lung_Squamous_Cell_Carcinoma_(LUSC)',
    'TCGA_Lung_Cancer_(LUNG)', 'TCGA_Lung_Adenocarcinoma_(LUAD)', 'TCGA_Lower_Grade_Glioma_(LGG)',
    'TCGA_Liver_Cancer_(LIHC)', 'TCGA_Large_Bcell_Lymphoma_(DLBC)', 'TCGA_Kidney_Papillary_Cell_Carcinoma_(KIRP)',
    'TCGA_Kidney_Clear_Cell_Carcinoma_(KIRC)', 'TCGA_Kidney_Chromophobe_(KICH)', 'TCGA_Head_and_Neck_Cancer_(HNSC)',
    'TCGA_Glioblastoma_(GBM)', 'TCGA_Esophageal_Cancer_(ESCA)', 'TCGA_Endometrioid_Cancer_(UCEC)',
    'TCGA_Colon_and_Rectal_Cancer_(COADREAD)', 'TCGA_Colon_Cancer_(COAD)', 'TCGA_Cervical_Cancer_(CESC)',
    'TCGA_Breast_Cancer_(BRCA)', 'TCGA_Bladder_Cancer_(BLCA)', 'TCGA_Bile_Duct_Cancer_(CHOL)',
    'TCGA_Adrenocortical_Cancer_(ACC)', 'TCGA_Acute_Myeloid_Leukemia_(LAML)'
]

# We want to find the directory for "Sarcoma" in the most specific manner
matches = []
for subdir in subdirectories:
    if subdir.lower() in ['crawldata.ipynb', '.ds_store']:
        continue
    subdir_lower = subdir.lower()
    # Check if 'sarcoma' or 'sarc' is present
    if ("sarcoma" in subdir_lower) or ("sarc" in subdir_lower):
        matches.append(subdir)

# Among all matches, prefer the one that explicitly has "(SARC)" in its name if available
selected_subdirectory = None
for m in matches:
    if "(sarc)" in m.lower():
        selected_subdirectory = m
        break

# If no "(SARC)" match is found but we do have matches, pick the first
if not selected_subdirectory and matches:
    selected_subdirectory = matches[0]

if not selected_subdirectory:
    # If no matching directory is found, mark dataset as unavailable
    is_final = False
    is_gene_available = False
    is_trait_available = False
    _ = validate_and_save_cohort_info(
        is_final=is_final,
        cohort="TCGA",
        info_path=json_path,
        is_gene_available=is_gene_available,
        is_trait_available=is_trait_available
    )
    print(f"No suitable directory found for '{trait}'. Skipped this trait.")
else:
    # Step 2: Identify clinicalMatrix file and PANCAN file
    cohort_dir = os.path.join(tcga_root_dir, selected_subdirectory)
    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)

    # Step 3: Load both files as dataframes
    clinical_df = pd.read_csv(clinical_file_path, index_col=0, sep='\t')
    genetic_df = pd.read_csv(genetic_file_path, index_col=0, sep='\t')

    # Step 4: Print the column names of the clinical data
    print("Clinical data columns:")
    print(list(clinical_df.columns))
candidate_age_cols = ["age_at_initial_pathologic_diagnosis", "days_to_birth", "year_of_initial_pathologic_diagnosis"]
candidate_gender_cols = ["gender"]

print("candidate_age_cols =", candidate_age_cols)
print("candidate_gender_cols =", candidate_gender_cols)

if candidate_age_cols:
    age_preview = clinical_df[candidate_age_cols].head(5).to_dict(orient="list")
    print(age_preview)

if candidate_gender_cols:
    gender_preview = clinical_df[candidate_gender_cols].head(5).to_dict(orient="list")
    print(gender_preview)
# Based on the inspection of candidate columns and sample values,
# choose 'age_at_initial_pathologic_diagnosis' for age and 'gender' for gender.

age_col = "age_at_initial_pathologic_diagnosis"
gender_col = "gender"

print("Chosen age_col:", age_col)
print("Chosen gender_col:", gender_col)
# 1. Extract and standardize the clinical features
selected_clinical_df = tcga_select_clinical_features(
    clinical_df=clinical_df,
    trait=trait,
    age_col=age_col,
    gender_col=gender_col
)

# 2. Normalize gene symbols in the expression data and save
gene_df = normalize_gene_symbols_in_index(genetic_df)
gene_df.to_csv(out_gene_data_file)

# 3. Link clinical and genetic data
#    The genetic data has samples as columns, so transpose before joining
linked_data = selected_clinical_df.join(gene_df.T, how='inner')

# 4. Handle missing values
processed_data = handle_missing_values(linked_data, trait_col=trait)

# 5. Determine and remove biased features
biased_trait, processed_data = judge_and_remove_biased_features(processed_data, trait)

# 6. Final quality validation
gene_cols = [col for col in processed_data.columns if col not in [trait, "Age", "Gender"]]
is_gene_available = len(gene_cols) > 0
is_trait_available = trait in processed_data.columns

is_usable = validate_and_save_cohort_info(
    is_final=True,
    cohort="TCGA",
    info_path=json_path,
    is_gene_available=is_gene_available,
    is_trait_available=is_trait_available,
    is_biased=biased_trait,
    df=processed_data,
    note="Final data processing for Kidney_Chromophobe"
)

# 7. Save linked data if usable
if is_usable:
    processed_data.to_csv(out_data_file)