Question

I have a class that reads a file of a particular format. These files tend to be greater than 8Gb in size so are usually compressed. When reading the file in I wanted to catch the error of the file not being compressed but neither except IOError: nor except: will do so, for some reason I don’t understand.

There are a few classes defined together in the file VCF.py, though the offending class is vcfReader(). The file from which the object is instantiated is below test.py, and lastly the Traceback.

Anyone have any ideas as to why it isn’t working?

VCF.py

import gzip
import sys

class Call():
    '''
    Class to handle the sample genotypes and associated information
    '''

    def __init__(self,site,sample,format,data):
        #do stuff here#

class Variant():
    '''
    Class for a single row from a VCF file.
    '''
    def __init__(self, entry, samples):
       #do other stuff here


class vcfReader():
    '''
    read a compressed vcf file ignoring the meta-information, but parsing the header             for sample names
    '''  
    def __init__(self, file):
        try:
            self.vcfFile = gzip.open(file, 'rb')
        except IOError:
            print "Not a gzipped file"
            sys.exit()

        self.samples = self.readHeader()

    def readHeader(self):
        line = self.vcfFile.next()
        while line.startswith('#'):
            if line[1]!='#':
                #lines that start with ##, i.e. meta tags are ignored. Header line starting with '#', sample names are extracted.
                return line.rstrip().rsplit('\t')[9:]
            else:           
                line = self.vcfFile.next()

    def __iter__(self):
        return self

    def next(self):
        row =  self.vcfFile.next()
        return Variant(row, self.samples)

and then test.py

import VCF
from collections import Counter

if __name__=='__main__':
    vcfreader = VCF.vcfReader('all_samples.vcf')

    filters = []
    for i in vcfreader:
        filters.extend(i.FILTERS)

    filters = Counter(filters)

    for k,v in filters.iteritems():
        print "{0}: {1}".format(k,v)

Here is the traceback:

Traceback (most recent call last):
  File "C:\Users\Davy\Documents\Programming\VCF_stuff\src\test.py", line 10, in <module>
    vcfreader = VCF.vcfReader('all_samples.vcf')
  File "C:\Users\Davy\Documents\Programming\VCF_stuff\src\VCF.py", line 95, in __init__
    self.samples = self.readHeader()
  File "C:\Users\Davy\Documents\Programming\VCF_stuff\src\VCF.py", line 98, in readHeader
    line = self.vcfFile.next()
  File "C:\Python27\lib\gzip.py", line 450, in readline
    c = self.read(readsize)
  File "C:\Python27\lib\gzip.py", line 256, in read
    self._read(readsize)
  File "C:\Python27\lib\gzip.py", line 291, in _read
    self._read_gzip_header()
  File "C:\Python27\lib\gzip.py", line 185, in _read_gzip_header
    raise IOError, 'Not a gzipped file'
IOError: Not a gzipped file

Answer 1

The reason your except block doesn’t catch the exception is that it happens outside the try block:

def __init__(self, file):
    try:
        self.vcfFile = gzip.open(file, 'rb')
    except IOError:
        print "Not a gzipped file"
        sys.exit()

    self.samples = self.readHeader() # <<<<<<<< exception is raised here