Need help parsing Cisco output

Question

I am having some issue trying to parse the mrib table of a router. I have been able to parse some of it out but having issue. For example I have the following output:

(192.168.1.1,232.0.6.8) RPF nbr: 55.44.23.1 Flags: RPF
  Up: 4w1d
  Incoming Interface List
    TenGigE0/0/0/1 Flags: A, Up: 4w1d
  Outgoing Interface List
    TenGigE0/0/0/10 Flags: A, Up: 4w1d

(192.168.55.3,232.0.10.69) RPF nbr: 66.76.44.130 Flags: RPF
  Up: 4w1d
  Incoming Interface List
    TenGigE0/0/0/0 Flags: A, Up: 4w1d
    TenGigE0/1/0/0 Flags: A, Up: 4w1d
    TenGigE0/2/0/0 Flags: A, Up: 4w1d
  Outgoing Interface List
    TenGigE0/0/0/10 Flags: A, Up: 4w1d
    TenGigE0/3/0/0 Flags: A, Up: 4w1d
    TenGigE0/4/0/0 Flags: A, Up: 4w1d

I am trying to build a data structure with the above output. For clarity I want it to look like this:

{'192.168.1.1,232.0.6.8': {'incoming': ['TenGigE0/0/0/1'],
                           'outgoing': ['TenGigE0/0/0/10']}}

The above is sounds simple. One of the main issues I am having is with the second block. I am trying to figure a way to iterate through the interfaces after the Incoming and Outgoing interface.

Not necessarily asking for the code but what would be the best way to do something like this?

I'm just going to say that "the best way to do something like this" in python is with [`TextFSM`](https://pypi.org/project/textfsm/ "TextFSM") — Mike Pennington, Jun 14 '19 at 03:09

Jan · Accepted Answer · 2016-06-22T12:36:13.083

Well, if you're able to use the newer regex module in Python, you can define subpatterns and use the following approach:

Define subpatterns for the IP address in the beginning
... as well as the Incoming and Outgoing Interface
Parse the Interfaces separately
See a demo on regex101.com.

Define subpatterns

Define subpatterns for the Incoming and Outgoing Interface strings, the IP adress and the end.

(?(DEFINE)
    (?<ips>[^()]+)
    (?<incoming>Incoming\ Interface \ List)
    (?<outgoing>Outgoing\ Interface \ List)
    (?<end>^$|\Z)
)

Put the regex together

Anchor the IP part to the beginning of the line and use a tempered greedy token with negative lookaheads for the incoming/outgoing part.

    ^\((?P<ip>(?&ips))\)
    (?:(?!(?&incoming))[\s\S]+?)
    (?&incoming)[\r\n]
    (?P<in>(?!(?&outgoing))[\s\S]+?) # tempered greedy token
    (?&outgoing)[\r\n]
    (?P<out>(?!^$)[\s\S]+?)
    (?&end)

Parse the incoming/outgoing parts

As you only need the interfaces types/names, you can simply come up with:

TenGig\S+ # TenGig, followed by anything NOT a whitespace

Hints

You do not really need to define subpatterns but then you'll need to repeat yourself a lot (because of the neg. lookaheads). So if you need to stick with the original re module, you can very well use this as well.

Glued together

All glued together in code, this will be:

import regex as re

string = """
(192.168.1.1,232.0.6.8) RPF nbr: 55.44.23.1 Flags: RPF
  Up: 4w1d
  Incoming Interface List
    TenGigE0/0/0/1 Flags: A, Up: 4w1d
  Outgoing Interface List
    TenGigE0/0/0/10 Flags: A, Up: 4w1d

(192.168.55.3,232.0.10.69) RPF nbr: 66.76.44.130 Flags: RPF
  Up: 4w1d
  Incoming Interface List
    TenGigE0/0/0/0 Flags: A, Up: 4w1d
    TenGigE0/1/0/0 Flags: A, Up: 4w1d
    TenGigE0/2/0/0 Flags: A, Up: 4w1d
  Outgoing Interface List
    TenGigE0/0/0/10 Flags: A, Up: 4w1d
    TenGigE0/3/0/0 Flags: A, Up: 4w1d
    TenGigE0/4/0/0 Flags: A, Up: 4w1d
"""

rx = re.compile(r"""
            (?(DEFINE)
                (?<ips>[^()]+)
                (?<incoming>Incoming\ Interface \ List)
                (?<outgoing>Outgoing\ Interface \ List)
                (?<end>^$|\Z)
            )
            ^\((?P<ip>(?&ips))\)
            (?:(?!(?&incoming))[\s\S]+?)
            (?&incoming)[\r\n]
            (?P<in>(?!(?&outgoing))[\s\S]+?)
            (?&outgoing)[\r\n]
            (?P<out>(?!^$)[\s\S]+?)
            (?&end)
""", re.MULTILINE|re.VERBOSE)

rxiface = re.compile(r'TenGig\S+')

result = dict()
for match in rx.finditer(string):
    key = match.group('ip')
    incoming = rxiface.findall(match.group('in'))
    outgoing = rxiface.findall(match.group('out'))

    result[key] = {'incoming': incoming, 'outgoing': outgoing}

print result
# {'192.168.1.1,232.0.6.8': {'outgoing': ['TenGigE0/0/0/10'], 'incoming': ['TenGigE0/0/0/1']}, '192.168.55.3,232.0.10.69': {'outgoing': ['TenGigE0/0/0/10', 'TenGigE0/3/0/0', 'TenGigE0/4/0/0'], 'incoming': ['TenGigE0/0/0/0', 'TenGigE0/1/0/0', 'TenGigE0/2/0/0']}}

Wow! This is awesome. This is exactly what I was needing and thank you for the direction (and code). I will play with this today. I definitely learned something with the newer RE that you stated. Will let update you, or the code, if any issues arise. — kaceymusgraves, Jun 22 '16 at 13:27
Worked perfectly. I had to add some different interface types to the rxiface expression, BUT I didn't state that in the question. Thank you again. — kaceymusgraves, Jun 22 '16 at 13:40

score 0 · Answer 2 · answered Jun 22 '16 at 07:44

Assumes your input is complete and well formatted:

matcher = re.compile(
    r'\((?P<range>[^\)]+)\)|'
    r'(?P<incoming>\s+Incoming Interface List)|'
    r'(?P<outgoing>\s+Outgoing Interface List)|'
    r'\s+(?P<interface>TenGigE0[^\s]+)'
)

with open('router_table.txt', 'r') as f:
    routing_table = []
    current_range = ''
    direction = ''
    for line in f:
        match = matcher.search(line)
        if match:
            if match.group('interface'):
                routing_table[-1][current_range][direction].append(match.group('interface'))
            if match.group('range'):
                current_range = match.group('range')
                routing_table.append(
                    {
                        current_range: {
                            'incoming': [],
                            'outgoing': []
                        }
                    }
                )
            if match.group('incoming'):
                direction = 'incoming'
            if match.group('outgoing'):
                direction = 'outgoing'