views:

49

answers:

3

I am downloading a file from the net, and it fails even though I am doing:

for p in query:

try:

except IOError as e:
   print e;

If there is an error, I want to log it, and then continue on with the next file.

In this loop, I am trying to download an image, if for some reason the filename was bad, or the website was down, etc., I want to continue with the next item in the for loop.

Is there a more generic error that won't fail and continue processing?

Also, how can I log errors to a file?

+2  A: 

You could use the logging module:

import logging
logging.basicConfig(filename='/tmp/myapp.log', level=logging.DEBUG, 
                    format='%(asctime)s %(levelname)s %(name)s %(message)s')
logger=logging.getLogger(__name__)

try:
    1/0
except ZeroDivisionError as err:
    logger.error(err)

Running the script writes in /tmp/myapp.log:

% cat /tmp/myapp.log 
2010-08-01 17:50:45,960 ERROR __main__ integer division or modulo by zero
unutbu
Nice. But the download is still broken. This won't magically make it "[not] fail and continue processing".
S.Lott
+1  A: 

As pointed out by Lott, if a download is failing, unless the problem is fixed upstream (or with your download address), the best you can do is to try again. However, if the situation is that you have a list of downloads, and simply want to skip over failed downloads instead of exiting, then:

logf = open("download.log", "w")
for download in download_list:
    try:
        # code to process download here
    except Exception as e:     # most generic exception you can catch
        logf.write("Failed to download {0}: {1}\n".format(str(download), str(e)))
        # optional: delete local version of failed download
    finally:
        # optional clean up code
        pass

Things to note:

(1) Use of the "logging" module, as suggested by ~unutbu, gives you much more flexibility and power with your logging output, including time stamp, simultaneously writing to different channels (e.g., stderr, file) depending on error levels, etc. etc.

(2) You might consider implementing the above logic using the "with" construct.

Jeet
what is logf, can't find it anywhere when searching for 'python logf' in google.
Blankman
ah, its just a variable thought it was using logging sorry.
Blankman
A: 

This catches everything. But it is much, much better to catch the exact exception. python <= 2.7

while True
  try:
    doStuff()
  except Exception, e:
    f = open('log.txt', 'w')
    f.write('An exceptional thing happed - %s' % e)
    f.close()
nate c