tmconfpy

ci-pipeline test coverage container image size Package version releases


tmconfpy provides a simple parser (tmconfpy command) to serialize a tmconf file (eg. /config/bigip.conf) to JSON (or python dict). The produced JSON is printed to STDOUT or a specified output (--output) file. It is also usable as a python module for easy consumption in your own projects.

This project aims to be a minimalistic dependency free tool. It is based on tmconfjs, it’s parsing implementation leans heavily on the community project F5 BIG-IP Automation Config Converter (BIG-IP ACC).

The TMOS configuration parser f5-corkscrew is a more sophisticated alternative with advanced functionality and active development.

Have a look at the example directory, for interactive use of tmconfpy with jupyter notebooks :notebook: or for implementing policy-as-code / audit configuration for compliance :cop:

More details about the relevant BIG-IP configuration files and data formats.

Using tmconfpy with ansible

tmconfpy is available as an ansible module, please see docs: Ansible.

Documentation by example

Installation

pip3 install tmconfpy

Command line usage

When installed globally, tmconfpy can be invoked as a command:

tmconfpy example/test.tmconf 2>/dev/null \
    | jq '."ltm profile client-ssl clientssl-secure"'
{
  "app-service": "none",
  "cert": "/Common/default.crt",
  "cert-key-chain": {
    "default": {
      "cert": "/Common/default.crt",
      "key": "/Common/default.key"
    }
  },
  "chain": "none",
  "ciphers": "ecdhe:rsa:!sslv3:!rc4:!exp:!des",
  "defaults-from": "/Common/clientssl",
  "inherit-certkeychain": "true",
  "key": "/Common/default.key",
  "options": [
    "no-ssl",
    "no-tlsv1.3"
  ],
  "passphrase": "none",
  "renegotiation": "disabled"
}

Errors, warnings or any debug information is written to STDERR:

tmconfpy example/test.tmconf \
    >/dev/null 2> example/test.tmconf.log

cat example/test.tmconf.log
2024-06-30T18:39:16Z - WARNING - tmconfpy.parser - UNRECOGNIZED LINE for object 'sys software update': '     auto-check enabled'
2024-06-30T18:39:16Z - WARNING - tmconfpy.parser - UNRECOGNIZED LINE for object 'sys software update': '     auto-phonehome enabled'
2024-06-30T18:39:16Z - WARNING - tmconfpy.parser - UNRECOGNIZED LINE for object 'fatal-grace-time': '	time 500'
2024-06-30T18:39:16Z - WARNING - tmconfpy.parser - UNRECOGNIZED LINE for object 'fatal-grace-time': '	enabled yes'

Input is also accepted from STDIN:

cat example/imap.tmconf | tmconfpy
{
    "ltm profile imap imap": {
        "activation-mode": "require"
    }
}

The <file_path> argument is preferred over STDIN however:

cat example/imap.tmconf | tmconfpy example/pop3.tmconf
{
    "ltm profile pop3 pop3": {
        "activation-mode": "require"
    }
}

The output can be written to a specified file using --output or -o when STDOUT is not desired:

tmconfpy --output example/pop3.tmconf.json example/pop3.tmconf
cat example/pop3.tmconf.json
{
    "ltm profile pop3 pop3": {
        "activation-mode": "require"
    }
}

tmconfpy supports multiple output formats of the parsed tmconf data, which can be specified via --format.

(cat example/imap.tmconf; echo; cat example/pop3.tmconf) | \
  tmconfpy --format jsonl
{"path": "ltm profile imap", "name": "imap", "object": {"activation-mode": "require"}}
{"path": "ltm profile pop3", "name": "pop3", "object": {"activation-mode": "require"}}
(cat example/imap.tmconf; echo; cat example/pop3.tmconf) | \
  tmconfpy --format tabular
[
  ["ltm profile imap", "imap", {"activation-mode": "require"}],
  ["ltm profile pop3", "pop3", {"activation-mode": "require"}]
]
(cat example/imap.tmconf; echo; cat example/pop3.tmconf) | \
  tmconfpy --format tabular_kv
[
  {"path":"ltm profile imap","name":"imap","object":{"activation-mode":"require"}},
  {"path":"ltm profile pop3","name":"pop3","object":{"activation-mode":"require"}}
]

Sorting the output is also supported since version 1.1.0. This is helpful when comparing data. tmconfpy uses python sorted() and will sort all data within the tmconf (all dicts, and lists).

cat <<EOF | tmconfpy --sort | jq 
ltm profile profile-type zProfile { }
ltm profile profile-type MyProfile {
    b {
        Z { 3 2 A 1 0 }
        a 1
        A 2
    }
    aaa 0
    AA { a c b }
}
EOF
{
  "ltm profile profile-type MyProfile": {
    "AA": [ "a", "b", "c" ],
    "aaa": "0",
    "b": {
      "A": "2",
      "Z": [ "0", "1", "2", "3", "A" ],
      "a": "1"
    }
  },
  "ltm profile profile-type zProfile": {}
}

Use as python module

>>> from tmconfpy import Parser
>>> parsed = Parser('example/imap.tmconf', is_filepath=True)
>>> parsed.dict
{'ltm profile imap imap': {'activation-mode': 'require'}}
>>> tmconf = r"""
... ltm profile pop3 pop3 {
...     activation-mode require
... }
... ltm profile imap imap {
...     activation-mode require
... }
... """
>>> parsed = Parser(tmconf)
>>> parsed.json
'{"ltm profile pop3 pop3": {"activation-mode": "require"}, "ltm profile imap imap": {"activation-mode": "require"}}'
>>> parsed.tabular
[tabularTmconf(path='ltm profile pop3', name='pop3', object={'activation-mode': 'require'}), tabularTmconf(path='ltm profile imap', name='imap', object={'activation-mode': 'require'})]
>>> parsed.tabular_kv
[{'path': 'ltm profile pop3',
  'name': 'pop3',
  'object': {'activation-mode': 'require'}},
 {'path': 'ltm profile imap',
  'name': 'imap',
  'object': {'activation-mode': 'require'}}]
>>> parsed.tabular_json
'[["ltm profile pop3", "pop3", {"activation-mode": "require"}], ["ltm profile imap", "imap", {"activation-mode": "require"}]]'
>>> parsed.jsonl
'{"path": "ltm profile pop3", "name": "pop3", "object": {"activation-mode": "require"}}\n{"path": "ltm profile imap", "name": "imap", "object": {"activation-mode": "require"}}'

Using the (optional) apiserver / container

Run the container, the API listens on port 8000 (http).

docker run --rm -p 8000:8000 simonkowallik/tmconfpy

The container is also available on ghcr.io as an alternative to docker hub.

docker run --rm -p 8000:8000 ghcr.io/simonkowallik/tmconfpy

The apiserver can be reached at http://localhost:8000/ and offers two endpoints which are described by the OpenAPI specification.

API documentation can be reached at / and /redoc for interactive use.

Parsing a single file by using POST, note --data-binary is required to avoid interpretation of the file content:

curl -X POST -s http://localhost:8000/parser/ \
  --data-binary @example/imap.tmconf
{"ltm profile imap imap":{"activation-mode":"require"}}

Parsing multiple files via multipart form:

curl -X POST -s http://localhost:8000/fileparser/ \
  -F 'filename=@example/imap.tmconf' \
  -F 'filename=@example/pop3.tmconf'
[
  {"filename":"imap.tmconf",
   "output":{"ltm profile imap imap":{"activation-mode":"require"}}
  },
  {"filename":"pop3.tmconf",
   "output":{"ltm profile pop3 pop3":{"activation-mode":"require"}}
  }
]

JSONL and tabular data

The /parser/ api-endpoint also supports returning the parsed tmconf as JSONL or tabular data using the query parameter ?response_format=<format>.

(cat example/imap.tmconf; echo; cat example/pop3.tmconf) | \
  curl -X POST -s http://localhost:8000/parser/?response_format=jsonl \
  --data-binary @-
{"path": "ltm profile imap", "name": "imap", "object": {"activation-mode": "require"}}
{"path": "ltm profile pop3", "name": "pop3", "object": {"activation-mode": "require"}}
(cat example/imap.tmconf; echo; cat example/pop3.tmconf) | \
  curl -X POST -s http://localhost:8000/parser/?response_format=tabular \
  --data-binary @-
[
  ["ltm profile imap","imap",{"activation-mode":"require"}],
  ["ltm profile pop3","pop3",{"activation-mode":"require"}]
]

Sorting the output is also supported since version 1.1.0. This is helpful when comparing data. tmconfpy uses python sorted() and will sort all data within the tmconf (all dicts, and lists).

cat <<EOF | tmconfpy --sort | jq 
ltm profile profile-type zProfile { }
ltm profile profile-type MyProfile {
    b {
        Z { 3 2 A 1 0 }
        a 1
        A 2
    }
    aaa 0
    AA { a c b }
}
EOF
{
  "ltm profile profile-type MyProfile": {
    "AA": [ "a", "b", "c" ],
    "aaa": "0",
    "b": {
      "A": "2",
      "Z": [ "0", "1", "2", "3", "A" ],
      "a": "1"
    }
  },
  "ltm profile profile-type zProfile": {}
}

Using the container as a command line tool

Use the --entrypoint argument with tmconfpy to invoke the tmconfpy tool instead of the apiserver (which is the default). Don’t forget to pass --interactive | -i to the container.

cat example/imap.tmconf | docker run --rm --interactive --entrypoint tmconfpy simonkowallik/tmconfpy
{
    "ltm profile imap imap": {
        "activation-mode": "require"
    }
}

Note that you can’t use --output | -o to write the output to a file using the above method unless you mount a volume into the container.

Disclaimer, Support, License

Please read and understand the LICENSE first.

There is no support on this project.

It is maintained on best effort basis without any warranties.

For any software or components used in this project, read their own LICENSE and SUPPORT policies.

If you decide to use this project, you are solely responsible.