File search program not efficient enough

Question

I have created a program to search for a file in the computer. But I feel it can be improved. Can anyone help me with this?

filesearch.py

import win32api
import os
import io
from fuzzywuzzy import process
def filesearch(self,*args):
 drives = win32api.GetLogicalDriveStrings()
 drives = drives.split('000円')[:-1]
 searchfile = ''.join((args))
 print("Searching...")
 try:
 q = io.open("index.txt", "r", encoding="utf-8")# I have created an index using index.py
 x = q.readlines()
 z = searchfile.lower()
 matches=[]
 basenames=[]
 for line in x:
 ##################################
 # I created this part so that if the file doesn't exist it skips the file. But I feel this doesn't work properly.
 #if os.path.exists(line) == False:
 #pass
 ##################################
 if z in os.path.basename(line):
 matches.append((line[:-1]))
 for match in matches:
 basenames.append(os.path.basename(match))
 s=process.extractOne(z, basenames)
 s=list(s)
 for item in matches:
 if s[0] == os.path.basename(item):
 print("Chosen file: %s"%(item) + "\nMatch percent: %s"%(s[1]))
 print (item)
 try:
 os.startfile(item)
 break
 except WindowsError:
 print"Sir, I am not able to open the file."
 print("\nMatches:")
 for match in matches:
 print(match)
 q.close()
 # Ih created this part as a failsafe just in case the index isn't up to date
 except TypeError, e:
 print(e)
 searchfile = re.compile(searchfile,re.I|re.DOTALL)
 matches = []
 for drive in drives:
 if 'C' in drive:
 for folderName, subfolders, filenames in os.walk(drive):
 # Skip the windows folder as it has to may files and doesn't have anything the user might need.
 if 'Windows' in folderName:pass
 else:
 for filename in filenames:
 thefile = os.path.join(folderName,filename)
 mo = searchfile.search(thefile)
 try:
 #Update index in case the file didn't exist in the index.
 matches.append(thefile)
 q = io.open("index.txt", "a", encoding="utf-8")
 q.write(("\n"+thefile).lower())
 q.close()
 except AttributeError:
 pass
 else:
 for folderName, subfolders, filenames in os.walk(drive):
 for filename in filenames:
 thefile = os.path.join(folderName,filename)
 mo = searchfile.search(thefile)
 try:
 #Update index in case the file didn't exist in the index.
 matches.append(thefile)
 q = io.open("index.txt", "a", encoding="utf-8")
 q.write(("\n"+thefile).lower())
 q.close()
 except AttributeError:
 pass
 if len(matches) != 0:
 print"\nMatches:"
 for m in matches:
 print("\n"+m)

index.py

from __future__ import unicode_literals
import os, sys
import os
import io
x = open("index.txt", "w")
x.write("")
x.close()
from os.path import join
for (dirname, dirs, files) in os.walk('C:\\'):
for filename in files:
 thefile = os.path.join(dirname,filename)
 print thefile
 x = io.open("index.txt", "a", encoding="utf-8")
 x.write(("\n"+thefile).lower())
 x.close()
for (dirname, dirs, files) in os.walk('D:\\'):
for filename in files:
 thefile = os.path.join(dirname,filename)
 print thefile
 x = io.open("index.txt", "a", encoding="utf-8")
 x.write(("\n"+thefile).lower())
 x.close()
for (dirname, dirs, files) in os.walk('E:\\'):
for filename in files:
 thefile = os.path.join(dirname,filename)
 print thefile
 x = io.open("index.txt", "a", encoding="utf-8")
 x.write(("\n"+thefile).lower())
 x.close()

Also, I believe my program uses depth first search. Is there a better way to search for files on a computer?

• Graipher: Your indentation in index.py is still off (the for loops should be nested).

Answer 1

A few points:

• Your program is neither breadth or depth-first; it uses whatever order os.walk returns.

• In index.py, opening and closing the index file so often when writing it is slow. Instead, put the open and close outside the for loop; also, use the with syntax for auto-closing, something like:

• In index.py, you're repeating code a lot. Consolidate it.

  from itertools import chain
  def drives():
   for drv in "C", "D", "E":
   yield from os.walk("{}:\\".format(drv))
  with io.open("index.txt", "a", encoding="utf-8") as x:
   for dirname, dirs, files in drives():
   for filename in files:
   thefile = os.path.join(dirname,filename)
   print thefile
   x.write(("\n"+thefile).lower())
  

• ferada: Is the second point missing a code example?
• pjz: no, the code example addresses both the second and third point.