Template Documentation
A parser template documentation is provided along with the parser examples in
parser directory. The template doc defines the common
format/structure/guidelines which helps to guide developers to start and
complete their parser development.
Template content:
#***************************************************************************
#* Parser Template
#* -------------------------------------------------------------------------
#* ABOUT THIS TEMPLATE - Please read
#*
#* - Any comments with "#*" in front of them (like this entire comment box)
#* are for template clarifications only and should be removed from the
#* final product.
#*
#* - Anything enclosed in <> must be replaced by the appropriate text for
#* your application
#*
#* Author:
#* Ke Liu, Automation Strategy - Core Software Group (CSG)
#*
#* Support:
#* asg-genie-support@cisco.com
#*
#* Description:
#* This template file describes how to write a specific parser class by
#* inheriting the MetaParser object.
#*
#* Read More:
#* For the complete and up-to-date user guide on parser template, visit:
#* URL= http://wwwin-pyats.cisco.com/documentation/html/parser/index.html
#*
#***************************************************************************
#***************************************************************************
#* DOCSTRINGS
#*
#* All test scripts should use the built-in Python docstrings
#* functionality to define script/class/method headers.
#*
#* Format:
#* Docstring format should follow:
#* URL= http://sphinxcontrib-napoleon.readthedocs.org/en/latest/index.html
#*
#* Read More:
#* Python Docstrings, PEP 257:
#* URL= http://legacy.python.org/dev/peps/pep-0257/
#***************************************************************************
'''template.py
< describe your parser >
Arguments:
<name> (<type>): <description of your parser argument>
Examples:
< provide examples on how to use this parser: init and call. >
References:
< provide references here. >
Notes:
< provide notes if needed >
'''
#***************************************************************************
#* OPTIONAL AUTHOR INFORMATION
#*
#* format:
#* __author__ = '<first> <last> <email>'
#* __copyright__ = 'Copyright 2016, Cisco Systems'
#* __credits__ = ['<list>', '<of>', '<names>']
#* __maintainer__ = '<team owning/maintaining this script>'
#* __email__ = '<email of owners>''
#* __date__= '<last modified date>'
#* __version__ = <decimal version string>
#*
#***************************************************************************
# optional author information
__author__ = 'Ke Liu <kel2@cisco.com>'
__copyright__ = 'Copyright 2016, Cisco Systems'
__credits__ = ["Sedy Yadollahi",
"Siming Yuan",
"Jean-Benoit Aubin"]
__maintainer__ = 'ASG/ATS team'
__email__ = 'asg-genie-support@cisco.com'
__date__= 'May 01, 2016'
__version__ = 1.0
#***************************************************************************
#* IMPORTS
#*
#* import all modules that are needed in your test script here. Use some
#* form of sorting to make it easy to read.
#*
#* Convention:
#* - one module per import for clarity
#* - sort imports either alphabetically or per length to give ease of
#* reading, also try to differentiate by functionality/distributor
#*
#* Example:
#* import os
#* import sys
#* import xmltodict
#* from genie.metaparser import MetaParser
#*
#* Read More:
#* Python Import System
#* URL= https://docs.python.org/3/reference/import.html
#***************************************************************************
#
# imports statements
#
from genie.metaparser.util.schemaengine import Any
from genie.metaparser import MetaParser
#***************************************************************************
#* ShowParser: parser class
#*
#* Each module contains at least one parser class which provides the
# implementation details of all supported parsing mechanisms (cli(), xml(),
#* yang()). Each parser class must inherit from `MetaParser`.
#*
#* Class name should be the first 2 words of the
#* corresponding cli command or equivalent. For example: class 'ShowVersion'
#* to represent 'show version'.
#*
#* If the first 2 words contain strong ambiguity (e.g.: show ip),
#* extend the next word (e.g.: show ip ospf) to clarify the parser purpose.
#*
#* For variable phrases within the parser name (e.g.: show interface Eth3/4),
#* use _WORD_ to present the phrase (e.g.: ShowInterface_WORD_).
#*
class ShowParser(MetaParser):
'''class ShowParser
parser class - implement detailed parsing mechanisms for cli, xml, and
yang output.
Arguments:
<name> (<type>): <description of your parser argument>
Examples:
< provide examples on how to initialize this parser. >
'''
#*************************
#* class constructor (optional) __init__():
#*
#* In case of redefining __init__ in parser class to overwrite the super
#* class MetaParser __init__() to support extra attributes, here is an
#* example:
#* def __init__(self, name, **kwargs):
#* super().__init__(name=name, **kwargs)
# <define your own __init__ here, or skip it by using superclass definition>
#*************************
#* schema - class variable
#*
#* schema defines the common data structure among all types of device
#* output (cli, xml, yang) the current parser supports. The typical
#* scenario is: the first user who defines the first parsing mechanism
#* (e.g.: cli ()) in parser class will also define the schema for the
#* output structure. At the end of the parsing process, parser engine
#* (MetaParser) will do schema checking to make sure the parser always
#* returns the output (nested dict) that has the same data structure
#* across all supported parsing mechanisms (cli(), yang(), xml()).
#*
#* Example of schema (show version) - nested dict
#* schema = {'cmp': {
#* 'module': {
#* Any(): {
#* 'bios_compile_time': str,
#* 'bios_version': str,
#* 'image_compile_time': str,
#* 'image_version': str,
#* 'status': str},}},
#* 'hardware': {
#* 'bootflash': str,
#* 'chassis': str,
#* 'cpu': str,
#* 'device_name': str,
#* 'memory': str,
#* 'model': str,
#* 'processor_board_id': str,
#* 'slots': str,
#* Any(): Any(),},}
#*
#* Here Any() in schema acting like a wildcard character, usually used
#* to presenting the variable keys within the dictionary.
#* For more info on how to use scheme: please read schemaengine API doc.
#*
# schema = <dict>
#******************************
#* parsing mechanism: cli
#* Function cli() defines the cli type output parsing mechanism which
#* typically contains 3 steps: executing, transforming, returning
#*
#* Step1 - executing
#* User has choices of calling the existing cli parsers from known
#* libraries, or implementing new parsing mechanism here
#* (eg.: regular expression).
#*
#* Example 1 - user implementing parsing mechanism
#* parsed_output = {}
#* output = self.device.execute("show version | inc 'cisco '")
#* m = re.match(r"cisco ([a-zA-Z0-9 ]+)", output.strip(' \t\n\r'))
#* if m:
#* parsed_output['model'] = m.group(0).strip()
#*
#* Step2 - transforming
#* This step might be optional for the first parser mechanism writer.
#* The purpose of this step is to enforce the final output structure
#* from all different parsing mechanisms (cli(), xml(), yang()) to be
#* same. User can greatly leverage all the functionalities provided in
#* metaparser.util class.
#*
#* Useful tools to do the transformation:
#* dict.update() --> adding missing key-value pairs
#* metaparser.util.keynames_convert() --> nested key names converting
#*
#* Step3: - returning
#* return the final result - the structure of the result has to be
#* (nested)dictionary
def cli(self, **kwargs):
# executing parser
# <step1 executing: get parsing resullt by calling existing backend
# parser function or write user own parsing code here>
# converting the result to compliance with schema
# <step2: transform the datastructure to be compliance with the
# schema defined on top of the class>
# <step3: return the final parsing result>
return
#******************************
#* parsing mechanism: xml
#* Function xml() defines the xml type output parsing mechanism which
#* typically contains 3 steps: executing, transforming, returning
#*
#* User has choices of calling the existing xml parsers from known
#* libraries, or implementing new parsing mechanism here.
#*
#* Example - Building xml parser using "xml.etree"
import xml.etree.ElementTree as ET
def xml(self, vrf='all'):
cmd = 'show bgp vrf {} all summary'.format(vrf)
out = self.device.execute(cmd + ' | xml')
etree_dict = {}
# Remove junk characters returned by the device
out = out.replace("]]>]]>", "")
root = ET.fromstring(out)
# top table root
show_root = Common.retrieve_xml_child(root=root, key='show')
# get xml namespace
# {http://www.cisco.com/nxos:7.0.3.I7.1.:bgp}
try:
m = re.compile(r'(?P<name>\{[\S]+\})').match(show_root.tag)
namespace = m.groupdict()['name']
except Exception:
return etree_dict
# compare cli command
Common.compose_compare_command(root=root, namespace=namespace,
expect_command=cmd)
# find Vrf root
root = Common.retrieve_xml_child(root=root, key='TABLE_vrf')
if not root:
return etree_dict
# ----- loop vrf -----
for vrf_tree in root.findall('{}ROW_vrf'.format(namespace)):
# vrf
try:
vrf = vrf_tree.find('{}vrf-name-out'.format(namespace)).text
except Exception:
break
# <vrf-router-id>19.0.0.6</vrf-router-id>
try:
route_identifier = vrf_tree.find('{}vrf-router-id'.format(namespace)).text
except Exception:
route_identifier = None
# <vrf-local-as>333</vrf-local-as>
try:
local_as = vrf_tree.find('{}vrf-local-as'.format(namespace)).text
except Exception:
local_as = None
# Address family table
af_tree = vrf_tree.find('{}TABLE_af'.format(namespace))
if not af_tree:
continue
for af_root in af_tree.findall('{}ROW_af'.format(namespace)):
# Address family table
saf_tree = af_root.find('{}TABLE_saf'.format(namespace))
if not saf_tree:
continue
# ----- loop address_family -----
for saf_root in saf_tree.findall('{}ROW_saf'.format(namespace)):
# neighbor
try:
af = saf_root.find('{}af-name'.format(namespace)).text
af = af.lower()
# initial af dictionary
af_dict = {}
if route_identifier:
af_dict['route_identifier'] = route_identifier
if local_as:
af_dict['local_as'] = int(local_as)
except Exception:
continue
# <tableversion>7</tableversion>
try:
af_dict['bgp_table_version'] = int(
saf_root.find('{}tableversion'.format(namespace)).text)
except Exception:
# for valide entry, table version should be there
continue
# <configuredpeers>3</configuredpeers>
af_dict['config_peers'] = \
int(saf_root.find('{}configuredpeers'.format(namespace)).text)
# <capablepeers>2</capablepeers>
af_dict['capable_peers'] = \
int(saf_root.find('{}capablepeers'.format(namespace)).text)
# <totalnetworks>5</totalnetworks>
try:
total_prefix_entries = \
int(saf_root.find('{}totalnetworks'.format(namespace)).text)
if 'prefixes' not in af_dict:
af_dict['prefixes'] = {}
af_dict['prefixes']['total_entries'] = total_prefix_entries
except Exception:
pass
# <totalpaths>10</totalpaths>
try:
total_path_entries = \
int(saf_root.find('{}totalpaths'.format(namespace)).text)
if 'path' not in af_dict:
af_dict['path'] = {}
af_dict['path']['total_entries'] = total_path_entries
except Exception:
pass
# <memoryused>1820</memoryused>
try:
memory_usage = \
int(saf_root.find('{}memoryused'.format(namespace)).text)
af_dict['path']['memory_usage'] = memory_usage
af_dict['prefixes']['memory_usage'] = memory_usage
except Exception:
pass
try:
# <numberattrs>1</numberattrs>
entries_1 = \
saf_root.find('{}numberattrs'.format(namespace)).text
# <bytesattrs>160</bytesattrs>
entries_2 = \
saf_root.find('{}bytesattrs'.format(namespace)).text
af_dict['attribute_entries'] = '[{0}/{1}]'.format(entries_1, entries_2)
except Exception:
pass
try:
# <numberpaths>1</numberpaths>
entries_1 = \
saf_root.find('{}numberpaths'.format(namespace)).text
# <bytespaths>34</bytespaths>
entries_2 = \
saf_root.find('{}bytespaths'.format(namespace)).text
af_dict['as_path_entries'] = '[{0}/{1}]'.format(entries_1, entries_2)
except Exception:
pass
try:
# <numbercommunities>0</numbercommunities>
entries_1 = \
saf_root.find('{}numbercommunities'.format(namespace)).text
# <bytescommunities>0</bytescommunities>
entries_2 = \
saf_root.find('{}bytescommunities'.format(namespace)).text
af_dict['community_entries'] = '[{0}/{1}]'.format(entries_1, entries_2)
except Exception:
pass
try:
# <numberclusterlist>0</numberclusterlist>
entries_1 = \
saf_root.find('{}numberclusterlist'.format(namespace)).text
# <bytesclusterlist>0</bytesclusterlist>
entries_2 = \
saf_root.find('{}bytesclusterlist'.format(namespace)).text
af_dict['clusterlist_entries'] = '[{0}/{1}]'.format(entries_1, entries_2)
except Exception:
pass
# <dampening>Enabled</dampening>
dampening = saf_root.find('{}dampening'.format(namespace)).text.lower()
if 'enabled' in dampening or 'true' in dampening:
af_dict['dampening'] = True
# <historypaths>0</historypaths>
try:
af_dict['history_paths'] = int(saf_root.find('{}historypaths'.format(namespace)).text)
except Exception:
pass
# <dampenedpaths>0</dampenedpaths>
try:
af_dict['dampened_paths'] = int(saf_root.find('{}dampenedpaths'.format(namespace)).text)
except Exception:
pass
# <softreconfigrecvdpaths>10</softreconfigrecvdpaths>
try:
af_dict['soft_reconfig_recvd_paths'] = int(
saf_root.find('{}softreconfigrecvdpaths'.format(namespace)).text)
except Exception:
pass
# <softreconfigidenticalpaths>10</softreconfigidenticalpaths>
try:
af_dict['soft_reconfig_identical_paths'] = int(
saf_root.find('{}softreconfigidenticalpaths'.format(namespace)).text)
except Exception:
pass
# <softreconfigcombopaths>0</softreconfigcombopaths>
try:
af_dict['soft_reconfig_combo_paths'] = int(
saf_root.find('{}softreconfigcombopaths'.format(namespace)).text)
except Exception:
pass
# <softreconfigfilteredrecvd>0</softreconfigfilteredrecvd>
try:
af_dict['soft_reconfig_filtered_recvd'] = int(
saf_root.find('{}softreconfigfilteredrecvd'.format(namespace)).text)
except Exception:
pass
# <softreconfigbytes>0</softreconfigbytes>
try:
af_dict['soft_reconfig_bytes'] = int(
saf_root.find('{}softreconfigbytes'.format(namespace)).text)
except Exception:
pass
# Neighbor table
nei_tree = saf_root.find('{}TABLE_neighbor'.format(namespace))
if not nei_tree:
continue
# Construct the returned structure
# ----- loop neighbors -----
for nei_root in nei_tree.findall('{}ROW_neighbor'.format(namespace)):
# neighbor
try:
nei = nei_root.find('{}neighborid'.format(namespace)).text
except Exception:
continue
if 'vrf' not in etree_dict:
etree_dict['vrf'] = {}
if vrf not in etree_dict['vrf']:
etree_dict['vrf'][vrf] = {}
if 'neighbor' not in etree_dict['vrf'][vrf]:
etree_dict['vrf'][vrf]['neighbor'] = {}
if nei not in etree_dict['vrf'][vrf]['neighbor']:
etree_dict['vrf'][vrf]['neighbor'][nei] = {}
if 'address_family' not in etree_dict['vrf'][vrf]['neighbor'][nei]:
etree_dict['vrf'][vrf]['neighbor'][nei]['address_family'] = {}
if af not in etree_dict['vrf'][vrf]['neighbor'][nei]['address_family']:
etree_dict['vrf'][vrf]['neighbor'][nei]['address_family'][af] = {}
sub_dict = etree_dict['vrf'][vrf]['neighbor'][nei]['address_family'][af]
# --- AF attributes -------
update_dict = deepcopy(af_dict)
sub_dict.update(update_dict)
# --- Neighbors attributes -------
# <neighborversion>4</neighborversion>
sub_dict['neighbor_table_version'] = int(
nei_root.find('{}neighborversion'.format(namespace)).text)
return etree_dict
#******************************
#* parsing mechanism: yang
#* Function yang() defines the yang type output parsing mechanism which
#* typically contains 3 steps: executing, transforming, returning
#*
#* Step1 - executing
#* User has choices of calling the existing yang parsers from known
#* libraries, or implementing new parsing mechanism here.
#*
#* Example - yang parsing mechnism implementation
from parser.yang.bgp_openconfig_yang import BgpOpenconfigYang
def yang(self):
# Initialize empty dictionary
map_dict = {}
# Execute YANG 'get' operational state RPC and parse the XML
bgpOC = BgpOpenconfigYang(self.device)
yang_dict = bgpOC.yang()
# Map keys from yang_dict to map_dict
# bgp_pid
map_dict['bgp_pid'] = yang_dict['bgp_pid']
# vrf
for vrf in yang_dict['vrf']:
if 'vrf' not in map_dict:
map_dict['vrf'] = {}
if vrf not in map_dict['vrf']:
map_dict['vrf'][vrf] = {}
for vrf_attr_key in yang_dict['vrf'][vrf]:
# Set router_id
if vrf_attr_key == 'router_id':
map_dict['vrf'][vrf][vrf_attr_key] = yang_dict['vrf'][vrf][vrf_attr_key]
# Set address_family
if vrf_attr_key == 'address_family':
map_dict['vrf'][vrf][vrf_attr_key] = yang_dict['vrf'][vrf][vrf_attr_key]
if vrf_attr_key == 'neighbor':
for nbr in yang_dict['vrf'][vrf]['neighbor']:
for key in yang_dict['vrf'][vrf]['neighbor'][nbr]:
# Set cluster_id
if key == 'route_reflector_cluster_id':
cluster_id = '0.0.0' + str(yang_dict['vrf'][vrf]['neighbor'][nbr]['route_reflector_cluster_id'])
map_dict['vrf'][vrf]['cluster_id'] = cluster_id
# Return to caller
return map_dict