python problems with gff3_fix

[jd3234]

When I am trying to run gff3_fix I am getting the following error:

Traceback (most recent call last):
File "./gff3toolkit_env/bin/gff3_fix", line 7, in <module>
sys.exit(script_main())
~~~~~~~~~~~^^
File ./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/bin/gff3_fix.py", line 95, in script_main
gff3_fix.fix.main(gff3=gff3, output_gff=args.output_gff, error_dict=error_dict, line_num_dict=line_num_dict, logger=logger_stderr)
~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/lib/gff3_fix/fix.py", line 686, in main
fix_phase(gff3=gff3, error_list=error_dict[error_code], line_num_dict=line_num_dict, logger=logger)
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/lib/gff3_fix/fix.py", line 424, in fix_phase
phase = list(map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006']))[1])
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
TypeError: 'map' object is not subscriptable

This seems to be related to a parathesis placement and can be fixed like this:

# line 424
# phase = list(map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006']))[1])
phase = list(map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006'])))[1]

When that is fixed I ran into an other error (same line 424, so previous fix didn't fix everything):

Traceback (most recent call last):
  File "./gff3toolkit_env/bin/gff3_fix", line 7, in <module>
    sys.exit(script_main())
             ~~~~~~~~~~~^^
  File "./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/bin/gff3_fix.py", line 95, in script_main
    gff3_fix.fix.main(gff3=gff3, output_gff=args.output_gff, error_dict=error_dict, line_num_dict=line_num_dict, logger=logger_stderr)
    ~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/lib/gff3_fix/fix.py", line 686, in main
    fix_phase(gff3=gff3, error_list=error_dict[error_code], line_num_dict=line_num_dict, logger=logger)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "./gff3toolkit_env/lib/python3.14/site-packages/gff3tool/lib/gff3_fix/fix.py", line 424, in fix_phase
    phase = list(map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006'])))[1]
            ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
IndexError: list index out of range

This seems to be a regex problem and could be fixed like this:

#phase = list(map(int,re.findall(r'\d',line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006'])))[1]
digits = list(map(int, re.findall(r'\d+', str(line_num_dict.get(sorted_CDS_list[0]['line_index']+1, {}).get('Ema0006', '0 0'))))); phase = digits[1] if len(digits) >= 2 else (digits[0] if digits else 0)

or more stable

if 'Ema0006' in line_num_dict[sorted_CDS_list[0]['line_index']+1]:
    error_msg = line_num_dict[sorted_CDS_list[0]['line_index']+1]['Ema0006']
    should_be_match = re.search(r'should be (\d+)', error_msg)
    if should_be_match:
        phase = int(should_be_match.group(1))
    else:
        # Fallback: try to get second digit from message
        digits = list(map(int, re.findall(r'\d+', error_msg)))
        if len(digits) >= 2:
            phase = digits[1]
        else:
            phase = 0

[MonicaPoelchau-USDA]

@jd3234 Sorry that you're having this issue, and thanks for providing a fix! So I understand the issue a bit better, could you post (or DM me) the line from the QC report that is causing the problem? That will help me understand why the fix program is not picking up the correct phase from the QC output.