[Tutor] extracting a column from many files

[Kent Johnson]

On Thu, Feb 19, 2009 at 5:41 AM, Bala subramanian
<bala.biophysics at gmail.com> wrote:
> Dear friends,
>
> I want to extract certain 6 different columns from a many files and write it
> to 6 separate output files.
>
>  #!/usr/bin/env python
> from sys import argv
> lst_files=argv[1:]
>
> sh=[];st=[];sta=[];buc=[];pro=[];ope=[]
>
> def extract(fname):
>     A=[];B=[];C=[];D=[];E=[];F=[]
>     data=open(fname).readlines()
>     for number, line in enumerate(data):
>         if "  Duplex" and " Shear" in line:
>             number=number+3
>         for x in range(0,8):
>                 new=data[number]
>                 A.append(new[19:26])
>                 B.append(new[27:34])
>                 C.append(new[37:42])
>                 D.append(new[44:54])
>                 E.append(new[56:63])
>                 F.append(new[69:75])
>                 number = number + 1
>     sh.append(A)
>     st.append(B)
>     sta.append(C)
>     buc.append(D)
>     pro.append(E)
>     ope.append(F)

I think you want to use extend() rather than append() here so you end
up with flat lists of values. Using append() creates lists of lists
which I don't think you want.
>
> for x in lst_files:
>           extract(x)
>
> list=[sh,st,sta,buc,pro,ope]

Don't use the names of built-ins (e.g. 'list') as variable names, it
shadows the built-in name.

> for value in list:
>     row=map(None,*value)

Don't do this. You just want one column of data. The program you took
this from was writing one column *per input file* which I don't think
you want.

>     out=open(str(value) + '.txt','w')

Value is a list, not a file name. I think you want something like this:

# List of pairs of (value list, name)
lists = [ (sh, 'sh'), (st, 'st'), ...]
for values, name in lists:
  out = open(name + '.txt', 'w')
  for value in values:
    out.write(str(value))
    out.write('\n')
  out.close()

Kent

>     for num in row:
>           out.write('\t'.join(num))
>           out.write('\n')
>     out.close()
>
>
>
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> http://mail.python.org/mailman/listinfo/tutor
>
>