#!/usr/bin/env python3

"""Run `sarif-create-aggregate-report` over csvs in the directories produced by
   ./sarif-extract-scans-(runner)
    ...

    and creates the summary file as named by the arg
"""

import argparse
import os
import sys
import pandas as pd
import csv
from sarif_cli import status_writer

#
# Handle arguments
#
parser = argparse.ArgumentParser(description='Run sarif-extract-scans over a directory hierarchy')

parser.add_argument('sarif_files', metavar='sarif-files', type=str,
                    help='File containing list of sarif files that were processed, use - for stdin') 

parser.add_argument('-s','--summary_filename', metavar='summary-filename', type=str, default="summary-report.csv",
                    help='Filename for final summary report') 

parser.add_argument('-in', '--in-dir', metavar='input-dir', type=str, default="",
                    help='Directory containing input set of results (corresponds to --outdir on the runner if supplied')

args = parser.parse_args()

#
# only warn if specified outfile exists, still use
#
if os.path.exists(args.summary_filename):
    print("Summary file provided exists, warning, overwriting.")
      
#
# Collect sarif file information
# 
with open(args.sarif_files, 'r') if args.sarif_files != '-' else sys.stdin as fp: 
    paths = fp.readlines()

#
# If specific input dir specified - format that
# 
if args.in_dir != "":
    args.in_dir+="/"

#
# Traverse all possible individual summary csv containing directory
# 
number_processed= 0
data = []

for path in paths:
    path = path.rstrip()
    #
    # Validate input data directory and content
    #
    csv_infile = os.path.join(args.in_dir+ path + ".csv")
    if not os.path.exists(csv_infile):
        continue
    else:
        number_processed+=1
        data.append(pd.read_csv(csv_infile))
        
all = pd.concat(data)

final_counts = [0]*(status_writer.STATUS_NUM+1)

for i in range(status_writer.STATUS_NUM+1):
    try:
        final_counts[i]=all['levelcode'].value_counts()[i]
    except KeyError: pass

header = ['number_processed', 'number_successfully_created', 'number_zero_results', 
"number_input_sarif_missing", "number_file_load_error", "number_input_sarif_extra", "number_unknown_sarif_parsing_shape",
"number_unknown" ]

final_counts.insert(0, number_processed)

with open(args.summary_filename, 'w') as f:
    csv_writer = csv.writer(f)
    csv_writer.writerow(header)
    csv_writer.writerow(final_counts)