[CentralOH] parsing Windows command line output

[Eric Lake]

Neil,

That seems to do exactly what I need it to do. Now I just need to get
to a point where I can understand what the heck it is doing :)

On Wed, Jul 23, 2008 at 11:27 AM, Neil Ludban <nludban at osc.edu> wrote:
> On Wed, 23 Jul 2008 07:52:49 -0400
> "Eric Lake" <ericlake at gmail.com> wrote:
>> I have been looking for a way to monitor a Windows cluster that we
>> have at work. I was really hoping for a good wmi class that I could
>> use from my XP machine to do it but I have not been lucky finding one.
>> I did however come across the 'cluster' command. I am thinking that I
>> could just call that from within my python code and parse the output
>> to get what I need. I want to get all of the data from the output and
>> then put it in a database. I can not see how to split the lines up
>> though. It would be easy if it was comma delimited but it seems to use
>> a varying number of spaces. My only thought is that I could do a split
>> on anything that is more than 2 spaces.
>
> #!/usr/local/bin/python
>
> import re
>
> fin = open('cluster.txt', 'r')
>
> fin.readline()  # Listing status for all available resources:
> fin.readline()  #
> fin.readline()  # Resource Group Node Status
> nom_len = [ len(x)      # - - - -
>            for x in re.findall('-+', fin.readline().strip()) ]
> #print nom_len  # [20, 20, 15, 6]
>
> def combinations(items, n):
>    if len(items) < n:
>        return
>    if n == 1:
>        for i in items:
>            yield [ i ]
>        return
>    for i in range(len(items)):
>        for rest in combinations(items[i+1:], n-1):
>            yield [ items[i] ] + rest
>    return
>
> #for c in combinations([ 'a', 'b', 'c', 'd' ], 2):
> #    print c
>
> def score(cols):
>    s = 100.0
>    # small penalty if too long
>    s -= 0.20 * sum([ max(len(c) - n, 0.0)
>                      for c, n in zip(cols, nom_len) ])
>    # large penalty if too short
>    s -= 0.80 * sum([ max(n - len(c), 0.0)
>                      for c, n in zip(cols, nom_len) ])
>    # bonus for trailing whitespace?
>    return s
>
> for line in fin.readlines():
>    line = line.strip()
>    cols = re.findall('\s+|[^\s]+', line)
>    #print cols
>    seps = range(1, len(cols), 2)
>    #print [ cols[i] for i in seps ]
>    best_cols = None
>    best_score = None
>    for seps in combinations(seps, 3):
>        tmp = [ ]
>        a = 0
>        for b in seps:
>            tmp.append(''.join(cols[a:b+1]))
>            a = b + 1
>        tmp.append(''.join(cols[a:]))
>        #print tmp
>        s = score(tmp)
>        if ((best_cols is None) or (s > best_score)):
>            best_cols = tmp
>            best_score = s
>    print [ c.strip() for c in best_cols ]
>
> #--#
>



-- 
Thanks,

Eric Lake