lml traceroute parsing does not handle hops with timeouts

[jesteria]

🐞 Bug Report:

The last-mile latency (lml) measurement's logic to parse the output of the traceroute command fails to handle non-responses – requests which timed out – and which are represented in the output as asterisks *.

Current Behavior:

Any hops containing a timeout are skipped. A warning is logged. An incorrect hop may be treated as the last mile.

Expected Behavior:

Timeouts should be permitted, and reported appropriately, as measurements. Hops should not be skipped (as feasible).

Steps to Reproduce:

n/a

Screenshots/Code Snippets:

Consider the following real traceroute output (with real values scrambled arbitrarily):

traceroute to 1.1.1.1 (1.1.1.1), 30 hops max, 60 byte packets
 1  laptop.container.docker (172.20.0.1)  0.043 ms  0.013 ms  0.012 ms
 2  dsldevice.attlocal.net (192.168.1.1)  3.742 ms  3.952 ms *
 3  111-195-164-1.lightspeed.cicril.sbcglobal.net (111.195.164.1)  15.590 ms  15.572 ms  16.258 ms
 4  71.151.17.196 (71.151.17.196)  16.192 ms  16.177 ms  16.160 ms
 5  * * *
 6  32.130.24.203 (32.130.24.203)  16.106 ms  4.901 ms  5.226 ms
 7  * 12.116.153.210 (12.116.153.210)  9.985 ms *
 8  141.101.73.222 (141.101.73.222)  10.537 ms 141.101.73.214 (141.101.73.214)  10.520 ms 141.101.73.210 (141.101.73.210)  10.497 ms
 9  one.one.one.one (1.1.1.1)  10.650 ms  10.363 ms  10.605 ms

In the above, line 2 is skipped, as its pattern does not match the code's expectations:

 2  dsldevice.attlocal.net (192.168.1.1)  3.742 ms  3.952 ms *

Environment:

ghcr.io/internet-equity/netrics:0.0.2-post2

Additional Information:

n/a