Mailing List Archive

On Sat, 24 Apr 1999 15:21:03 GMT, "Gordon McMillan"
<gmcm@hypernet.com> wrote:

>msrisney writes:

>> here are my questions: 1. this prints out not only files with the
>> file extensions ".log" but also any file name that has "log" in it's
>> name. how would I rewrite to avoid??
>
>In a regex, a "." is a wildcard character. If you want a literal "."
>you need to escape it:
> re.compile('\\.log')
>or
> re.compile(r'\.log')

Wouldn't that be re.compile('\\.log$'), so as to avoid things like
tree.log.jpg?

doesn't-yet-grok-python-but-does-grok-regexps-ly y'rs

Steve
--
-- Steve Atkins -- steve@blighty.com

Gordon McMillan wrote:
>
> msrisney writes:

...

> At any rate, os.path.exists('e:/') is an effective way of finding out
> if e: exists. Though specifying 'a:/' will pop up one of those lovely
> Abort/Retry/Fail dialogs if nothing is in the drive.

Not always. On my Win98 box, the floppy drive brushes teeth for
a second, and then I get back *one* from os.path.exists.
This is really bad, since I didn't insert a floppy.

> in-nearly-20-years-I-still-haven't-figured-out-the-difference-between
> -Abort-and-Fail-ly y'rs

Some of my early dos tools aborted immediately from the DOS
function when I hit abort, while they caught the error and
some were able to continue, when I answered with "Fail".
So I think the designed intent was to give the user a chance
to decide wether the program should shut down, or an internal
error handler should get a chance.

Anyway no good solution - ciao - chris

--
Christian Tismer :^) <mailto:tismer@appliedbiometrics.com>
Applied Biometrics GmbH : Have a break! Take a ride on Python's
Kaiserin-Augusta-Allee 101 : *Starship* http://starship.python.net
10553 Berlin : PGP key -> http://wwwkeys.pgp.net
PGP Fingerprint E182 71C7 1A9D 66E9 9D15 D3CC D4D7 93E2 1FAE F6DF
we're tired of banana software - shipped green, ripens at home

msrisney writes:
>
> I'm attemtping to locate log files on my drive(s), and do some
> comparison. here's a simplified snippet where I'm getting
> bottlenecked.
>
> >>>import os,re
> >>>regexp = re.compile('.log')
> >>>def find_log_files(arg, directory, names):
> ...for name in os.listdir(directory):
> if regexp.search(name):
> print directory + "\\" + name
> >>>os.path.walk('D:\\',find_log_files,None)
>
>
> here are my questions: 1. this prints out not only files with the
> file extensions ".log" but also any file name that has "log" in it's
> name. how would I rewrite to avoid??

In a regex, a "." is a wildcard character. If you want a literal "."
you need to escape it:
re.compile('\\.log')
or
re.compile(r'\.log')

> 2. is there a better, faster way of doing this??, my end goal is to
> open the files and compare time sequences to one another.

Any number of ways that would be faster. Since you've already got
os imported, you could use os.path.splitext()
if os.path.splitext(name)[1] == '.log':

> 3. Is ther any way to determine the number of drives on a system,
> obviuosly I am hardcoding the top level drive letter "D:\", is there
> any way to search the entire system much like win32's find file
> search??

Do you know how to search multiple drives with FindxxxFIle()? I've
always thought that you either got the current drive or specified the
drive to search.

At any rate, os.path.exists('e:/') is an effective way of finding out
if e: exists. Though specifying 'a:/' will pop up one of those lovely
Abort/Retry/Fail dialogs if nothing is in the drive.

in-nearly-20-years-I-still-haven't-figured-out-the-difference-between
-Abort-and-Fail-ly y'rs

- Gordon

[code to find and process files ending in .log]

>2. is there a better, faster way of doing this??,

For matching literal text, string.find, .rfind, .index, .rindex
are faster that regexes. Since you want to match text at end only,
if filename[-4:] == '.log'
will be even faster.

TJR

>>>>> "m" == msrisney <msrisney@my-dejanews.com> writes:

m> here are my questions: 1. this prints out not only files with the
m> file extensions ".log" but also any file name that has "log" in
m> it's name. how would I rewrite to avoid??

How about using "fnmatch.fnmatch" instead of a regular expression?

--
===== R.Hooft@EuroMail.net http://www.xs4all.nl/~hooft/rob/ =====
===== R&D, Nonius BV, Delft http://www.nonius.nl/ =====
===== PGPid 0xFA19277D ========================== Use Linux! =========