[Tutor] read text file in zip archive, process, plot

[Kent Johnson]

Washakie Wyoming wrote:
> Greetings all!
> 
> I'm including here a first python program! Very nice. It's written to
> read in a text file which resides in a .zip archive, extract two fields
> and plot them. It uses some bits from pylab for the plotting.
> 
> I'm writing to ask for ways to improve it. I feel like the writing and
> reading from 'jnk' files is definitely a hack! Any suggestions? I would
> be greatly appreciative of anyone provided comments on how to more
> efficiently tackle this problem. (NOTE: this is not a school
> assignment... just some side learning out of interest in what seems to
> be a great language!).
> 
> Thanks!
> 
> Here's my program (a zip is attached - if that's allowed???):
> 
> import zipfile
> import os
> import pylab as P
> 
> z = zipfile.ZipFile("flight_data.zip", "r")
> for filename in z.namelist():
>         print filename
>         contents = z.read(filename)
>         f = open('jnk','w')
>         f.write(contents)
>         f.close()
>         print "file closed, now reading"
>         #now open files to process
>         f = open('jnk','r')
>         fn = open('newjnk','w') 
>         data = f.readlines()

Alan has shown you a better way to do this.

>         firstline = data[0].strip().split(' ')
>         stind = int(firstline[0])      
>         hdrline = stind - 1
>         #print data[stind:len(data)]
>         for l in data[stind:len(data)]:

You don't need the end index when it is the length of the data:
   for l in data[stind:]:

>                 #l = l.replace('  ',',')
>                 fn.writelines(l)
>         f.close()
>         fn.close()
>         #print 'file closed now'
>         
> jnk2 = P.load('newjnk')
> t = jnk2[:,0]
> x = jnk2[:,24]

It looks like you are using newjnk just as a way to get P.load() to 
parse the data for you. You can do this yourself with something like
   x = []
   t = []
   for l in data[stind:]:
     l = l.split()  # Split on whitespace
     tval = float(l[0])
     t.append(tval)
     xval = float(l[24])
     x.append(xval)

Kent