逐行读取gzip压缩文本文件，以便在python 3.2.6中进行处理

Question

I'm a complete newbie when it comes to python, but I've been tasked with trying to get a piece of code running on a machine which has a different version of python (3.2.6) than that which the code was originally built for.

I've come across an issue with reading in a gzipped-text file line-by-line (and processing it depending on the first character). The code (which obviously is written in python > 3.2.6) is

for line in gzip.open(input[0], 'rt'):
    if line[:1] != '>':
        out.write(line)
        continue

    chromname = match2chrom(line[1:-1])
    seqname = line[1:].split()[0]

    print('>{}'.format(chromname), file=out)
    print('{}\t{}'.format(seqname, chromname), file=mappingout)

(for those who know, this strips gzipped FASTA genome files into headers (with ">" at start) and sequences, and processes the lines into two different files depending on this)

I have found https://bugs.python.org/issue13989 , which states that mode 'rt' cannot be used for gzip.open in python-3.2 and to use something along the lines of:

import io

with io.TextIOWrapper(gzip.open(input[0], "r")) as fin:
     for line in fin:
         if line[:1] != '>':
             out.write(line)
             continue

         chromname = match2chrom(line[1:-1])
         seqname = line[1:].split()[0]

         print('>{}'.format(chromname), file=out)
         print('{}\t{}'.format(seqname, chromname), file=mappingout)

but the above code does not work:

UnsupportedOperation in line <4> of /path/to/python_file.py:
read1

How can I rewrite this routine to give out exactly what I want - reading the gzip file line-by-line into the variable "line" and processing based on the first character?

EDIT: traceback from the first version of this routine is (python 3.2.6):

Mode rt not supported  
File "/path/to/python_file.py", line 79, in __process_genome_sequences  
File "/opt/python-3.2.6/lib/python3.2/gzip.py", line 46, in open  
File "/opt/python-3.2.6/lib/python3.2/gzip.py", line 157, in __init__

Traceback from the second version is:

UnsupportedOperation in line 81 of /path/to/python_file.py:
read1
File "/path/to/python_file.py", line 81, in __process_genome_sequences

with no further traceback (the extra two lines in the line count are the import io and with io.TextIOWrapper(gzip.open(input[0], "r")) as fin: lines

Answer 1

I have actually appeared to have solved the problem.

In the end I had to use shell("gunzip {input[0]}") to ensure that the gunzipped file could be read in in text mode, and then read in the resulting file using

for line in open(' *< resulting file >* ','r'):
    if line[:1] != '>':
        out.write(line)
        continue

    chromname = match2chrom(line[1:-1])
    seqname = line[1:].split()[0]

    print('>{}'.format(chromname), file=out)
    print('{}\t{}'.format(seqname, chromname), file=mappingout)