Question on how to parse this output

[evilmonkey19]

Hi!

I'm currently using TextFSM in NTC_Templates and so far has been an awesome experience. Thanks for the awesome project!

I would like to ask about a specific problem i came across from a CLI output. Basically I want to be able to parse the data using different templates or whatever. The thing I would love is to be able to parse this output in just one single object:

  -----------------------------------------------------------------------------
  FEC upstream switch :Disable
  OMCC Encrypt switch :On
  Qos mode            :Pq
  Mapping mode        :Vlan
  Tr069 management    :Disable
  -----------------------------------------------------------------------------
  Notes: * indicates Discrete TCONT(TCONT Unbound)
  -----------------------------------------------------------------------------
  <T-CONT   0>          DBA Profile-ID:1
  <T-CONT   1>          DBA Profile-ID:2
  <T-CONT   4>          DBA Profile-ID:5
    <Gem Index 126>
    ------------------------------------------------------------------------
    |Serv-Type:ETH |Encrypt:off |Cascade:off |GEM-CAR:-            |
    |Upstream-priority-queue:-  |Downstream-priority-queue:-       |
    ------------------------------------------------------------------------
     Mapping VLAN  Priority Port   Port  Bundle  Flow  Transparent
     index                  type   ID    ID      CAR
    ------------------------------------------------------------------------
     0       500   -        -      -     -       -     -         
    ------------------------------------------------------------------------

The output should be something as follows:

[{
   "fec_upstream_switch": "Disable",
   "omcc_encrypt_switch": "On",
   "qos_mode": "pq",
   "mapping_mode": "vlan",
   "tr069_management": "disable",
   "t-conts": [
      {
          "t_cont_id": 0,
          "dba_profile_id": 1,
      },
      {
          "t_cont_id": 1,
          "dba_profile_id": 2,
      },
      {
           "t_cont_id":4,
           "dba_profile_id":5,
       }
   ],
   "gems": [
      {
         "gem_index": 126,
         "t_cont_id": 4,
         "serv-type": "eth",
         "encrypt": "off",
         "gem_car": "",
         "upstream_priority_queue": "",
         "downstream_priority_queue": "",
      } 
  "gem_mappings": [
     {
        "gem_index": 126,
       "mapping_index": 0,
       "vlan": 500,
       "priority": "",
        "port_type": "",
         "bundle_id": "",
        "flow_car": "",
        "transparent": "",
     }
   ]
}]

I accept any other suggestion as well! I suggest to reuse the id and index as a way to connect each part, but if it can nested inside even better instead of relating using the indexes. Thanks!

[mjbear]

To close the loop for the broader community, there is a thread over at ntc-templates.

The summary of that thread being TextFSM is not set up for providing (deeply?) nested data. If nested data is required, maybe modify it after textfsm parsing (post-parse) or there might be another project that does this out-of-the-box. There's an awesome network automation list that might help show possibilities.