I am building a network automation script in python, and I got into a problem that I don't know how to resolve. I have the following output
CPE-MGT {
instance-type virtual-router;
interface ge-0/0/0.0;
routing-options {
static {
route 192.168.253.115/32 next-hop 192.168.100.1;
route 0.0.0.0/0 next-hop 192.168.100.1;
}
}
}
DATA {
instance-type virtual-router;
interface ge-0/0/1.0;
}
MGMT_BK {
instance-type virtual-router;
interface ge-0/0/2.0;
routing-options {
static {
route 192.168.253.115/32 next-hop 192.168.100.1;
route 0.0.0.0/0 next-hop 192.168.100.1;
}
}
}
I need to obtain the name of there I find the name of specific primary key interfaces; perhaps it will be clearer if I give an example.
I have a regex that looks into the text and finds that ge-0/0/0.0 is in the text now; what I don't know is how to obtain CPE-MGT.
cmd = net_connect.send_command(
juniper_cmds["sh_routing_instance"])
logger.debug(f'{ip_dict[config_dict["csv_ip_colum"]][ips]}:Device routing instances {cmd}')
regex_routing_instance = re.compile(f'{interface[0]}')
routing_instance = regex_routing_instance.findall(cmd)```
You could use two patterns and the regex
module which offers recursion:
import regex as re
data = """
CPE-MGT {
instance-type virtual-router;
interface ge-0/0/0.0;
routing-options {
static {
route 192.168.253.115/32 next-hop 192.168.100.1;
route 0.0.0.0/0 next-hop 192.168.100.1;
}
}
}
DATA {
instance-type virtual-router;
interface ge-0/0/1.0;
}
MGMT_BK {
instance-type virtual-router;
interface ge-0/0/2.0;
routing-options {
static {
route 192.168.253.115/32 next-hop 192.168.100.1;
route 0.0.0.0/0 next-hop 192.168.100.1;
}
}
}
"""
rx_block = re.compile(r"""
^(?P<name>[A-Z_-]+)\s+
(?P<body>\{
(?:(?:[^{}]+)|(?2))+
\})
""", re.M | re.X)
rx_inner = re.compile(r"interface ge-0/0/0.0;")
for block in rx_block.finditer(data):
m = rx_inner.search(block.group('body'))
if m:
print(block.group('name'))
This would yield
CPE-MGT
Two notes here: you wouldn't really need a regular expression for the inner search as it is a static string and two - probably use a parser solution instead.
See a demo for the expression on regex101.com.