从索引修剪字符串

Question

I have now a more or less working example, as you can see below, to kill some pods in openshift.

It's now stuck at this point.

myindx = out.find(key)
print out[myindx:-1].rstrip()

out is the output of oc get pod --all-namespaces ... which looks like this

 -n namespace01 pod-name-9-xd45
 -n namespace01 pod-name-9-xd67
 -n namespace02 pod-name2-9-xd45
 -n namespace02 pod-name2-9-rd45

with out.find(key) I find the right line now I need to get the whole line to call oc delete $line .

In perl I would put all output from oc get pod --all-namespaces ... into an array and search in every line in the array but how can I do this in python?

Thanks for any help.

#!python

import StringIO
import sys
import re
import subprocess
import shlex
from collections import defaultdict

# Since you're keeping counts, we'll initialize this so that the values
# of the dictionary are `int`
myhash = defaultdict(lambda: 0)

if sys.argv[3] != "NodeReady":
     sys.exit(0)

commandline = shlex.split("/bin/oc get pod --all-namespaces")

proc = subprocess.Popen(commandline,stdout=subprocess.PIPE)
out, err = proc.communicate()
buf = StringIO.StringIO(out)


for line in buf.readlines():
#    print line
    """
    Diese Projekte und pod Zeilen werden ignoriert
    """
    if re.match('^NAMESPACE|.*-router|^logging|^default|.*infra-ipfailover|.*-(build|deploy)',line):
        continue
    linesplited = line.split()
    prefix = re.split('(.*)-\d+-\w+$',linesplited[1])
    #print  linesplited[1]
    #print prefix[1]
    myhash[prefix[1]] += 1

commandline = shlex.split('/bin/oc get pod --all-namespaces -o template --template=\'{{range .items }} -n {{ .metadata.namespace }} {{.metadata.name}}\n{{end}}\'')

proc = subprocess.Popen(commandline,stdout=subprocess.PIPE)
out, err = proc.communicate()
#print out
buf = StringIO.StringIO(out)
#print buf

#for line in buf.readlines():
#    print line

for key in myhash.keys():
    print "<<<>>>"
#    print myhash[key] % 2 , "\n"
    print key,":",myhash[key]
    if  myhash[key] % 2 :
        if myhash[key] > 1:
            print "mod 2",key,":",myhash[key]
            myindx = out.find(key)
            print out[myindx].rstrip()
    else:
        print "not mod 2",key,":",myhash[key]
        print out.find(key)

Answer 1

I'm shooting a little in the dark here but I think your problem can be reduced to the following.

After this line, out should have all the output you mentioned.

out, err = proc.communicate() 

out probably looks like this:

-n namespace01 pod-name-9-xd45\n
-n namespace01 pod-name-9-xd67\n
-n namespace02 pod-name2-9-xd45\n
-n namespace02 pod-name2-9-rd45\n

Rather than shove it into a StringIO object, you can cut up this string into an array:

output_lines = [x.strip() for x in output.split('\n')]

And then output_lines looks like this:

['-n namespace01 pod-name-9-xd45',
 '-n namespace01 pod-name-9-xd67',
 '-n namespace02 pod-name2-9-xd45',
 '-n namespace02 pod-name2-9-rd45']

And then you want to find every the string in the array that contains your key.

if  myhash[key] % 2 :
    if myhash[key] > 1:
        print "mod 2",key,":",myhash[key]
        for ol in output_lines:
            myindx = ol.find(key)
            if myindx > -1:
                print ol