This repo contains jinjanator
, a CLI tool to render
Jinja2 templates. It is a fork of
j2cli
, which itself was a fork of jinja2-cli
, both of which are no
longer actively maintained.
Open Source software: Apache License 2.0
Features:
- Jinja2 templating
- INI, YAML, JSON data sources supported
- Environment variables can be used with or without data files
- Plugins can provide additional formats, filters, tests, extensions and global functions (see jinjanator-plugins for details)
pip install jinjanator
- jinjanator-plugin-ansible - makes Ansible's 'core' filters and tests available during template rendering
- jinjanator-plugin-format-toml - provides a TOML parser for input data files
- jinjanator-plugin-format-xml - provides an XML parser for input data files
Suppose you have an NGINX configuration file template, nginx.j2
:
server {
listen 80;
server_name {{ nginx.hostname }};
root {{ nginx.webroot }};
index index.htm;
}
And you have a JSON file with the data, nginx.json
:
{
"nginx":{
"hostname": "localhost",
"webroot": "/var/www/project"
}
}
This is how you render it into a working configuration file:
$ jinjanate nginx.j2 nginx.json > nginx.conf
The output is saved to nginx.conf
:
server {
listen 80;
server_name localhost;
root /var/www/project;
index index.htm;
}
Alternatively, you can use the -o nginx.conf
or --output-file nginx.conf
options to write directly to the file.
Suppose, you have a very simple template, person.xml.j2
:
<data><name>{{ name }}</name><age>{{ age }}</age></data>
What is the easiest way to use jinjanator here? Use environment variables in your Bash script:
$ export name=Andrew
$ export age=31
$ jinjanate /tmp/person.xml.j2
<data><name>Andrew</name><age>31</age></data>
Even when you use a data file as the data source, you can always
access environment variables using the env()
function:
Username: {{ login }}
Password: {{ env("APP_PASSWORD") }}
Or, if you prefer, as a filter:
Username: {{ login }}
Password: {{ "APP_PASSWORD" | env }}
jinjanate
accepts the following arguments:
template
: Jinja2 template file to renderdata
: (optional) path to the data used for rendering. The default is-
: use stdin.
There is some special behavior with environment variables:
- When
data
is not provided (data is-
),--format
defaults toenv
and thus reads environment variables.
--format FMT, -f FMT
: format for the data file. The default is?
: guess from file extension. Supported formats are YAML (.yaml or .yml), JSON (.json), INI (.ini), and dotenv (.env), plus any formats provided by plugins you have installed.--format-option OPT
: option to be passed to the parser for the data format selected with--format
(or auto-selected). This can be specified multiple times. Refer to the documentation for the format itself to learn whether it supports any options.--help, -h
: generates a help message describing usage of the tool.--import-env VAR, -e VAR
: import all environment variables into the template asVAR
. To import environment variables into the global scope, give it an empty string:--import-env=
. (This will overwrite any existing variables with the same names!)--output-file OUTFILE, -o OUTFILE
: Write rendered template to a file.--quiet
: Avoid generating any output on stderr.--undefined
: Allow undefined variables to be used in templates (no error will be raised).--version
: prints the version of the tool and the Jinja2 package installed.
For details on the behavior of these options, see the Customization section.
-
--filters PYTHON_FILE
- specify a file of Python source code, containing additional Jinja2 filters as simple functions. You can use this option more than once to include multiple files.- NOTE: While this option's behavior matches the
j2cli
documentation, it does not match thej2cli
implementation. If you are migrating fromj2cli
and use more than one filters file, you will need to specify this option once for each file.
- NOTE: While this option's behavior matches the
-
--tests PYTHON_FILE
- specify a file of Python source code, containing additional Jinja2 tests as simple functions. You can use this option more than once to include multiple files.- NOTE: While this option's behavior matches the
j2cli
documentation, it does not match thej2cli
implementation. If you are migrating fromj2cli
and use more than one tests file, you will need to specify this option once for each file.
- NOTE: While this option's behavior matches the
-
--customize PYTHON_FILE
- specify a file of Python source code containing customization functions. This file can modify the Jinja2 context, add filters/tests, or change Jinja2's configuration. Unlike--filters
and--tests
, this option can only be specified once.
Render a template using INI-file data source:
$ jinjanate config.j2 data.ini
Render using JSON data source:
$ jinjanate config.j2 data.json
Render using YAML data source:
$ jinjanate config.j2 data.yaml
Render using JSON data on stdin:
$ curl http://example.com/service.json | jinjanate --format=json config.j2 -
Render using environment variables:
$ jinjanate config.j2
Or use environment variables from a file:
$ jinjanate config.j2 data.env
Or pipe it: (note that you'll have to use "-" in this particular case):
$ jinjanate --format=env config.j2 - < data.env
Data input from environment variables.
This format does not support any options.
Render directly from the current environment variable values:
$ jinjanate config.j2
Or alternatively, read the values from a dotenv file:
NGINX_HOSTNAME=localhost
NGINX_WEBROOT=/var/www/project
NGINX_LOGS=/var/log/nginx/
And render with:
$ jinjanate config.j2 data.env
Or:
$ env | jinjanate --format=env config.j2
If you're going to pipe a dotenv file into jinjanate
, you'll need to
use "-" as the second argument:
$ jinjanate config.j2 - < data.env
INI data input format.
This format does not support any options.
data.ini:
[nginx]
hostname=localhost
webroot=/var/www/project
logs=/var/log/nginx
Usage:
$ jinjanate config.j2 data.ini
Or:
$ cat data.ini | jinjanate --format=ini config.j2
JSON data input format.
array-name
: accepts a single string (e.g.array-name=foo
), which must be a valid Python identifier and not a Python keyword. If this option is specified, and the JSON data provided is anarray
(sequence, list), the specified name will be used to make the data available to the Jinja2 template. Errors will be generated ifarray
data is provided and this option is not specified, or if this option is specified and the data provided is anobject
.
data.json:
{
"nginx":{
"hostname": "localhost",
"webroot": "/var/www/project",
"logs": "/var/log/nginx"
}
}
Usage:
$ jinjanate config.j2 data.json
Or:
$ cat data.json | jinjanate --format=ini config.j2
YAML data input format.
sequence-name
: accepts a single string (e.g.sequence-name=foo
), which must be a valid Python identifier and not a Python keyword. If this option is specified, and the YAML data provided is asequence
(array, list), the specified name will be used to make the data available to the Jinja2 template. Errors will be generated ifsequence
data is provided and this option is not specified, or if this option is specified and the data provided is amapping
.
data.yaml:
nginx:
hostname: localhost
webroot: /var/www/project
logs: /var/log/nginx
Usage:
$ jinjanate config.j2 data.yml
Or:
$ cat data.yml | jinjanate --format=yaml config.j2
Use an environment variable's value in the template.
This filter is available even when your data source is something other than the environment.
Example:
User: {{ user_login }}
Pass: {{ "USER_PASSWORD" | env }}
You can provide a default value:
Pass: {{ "USER_PASSWORD" | env("-none-") }}
For your convenience, it's also available as a global function:
User: {{ user_login }}
Pass: {{ env("USER_PASSWORD") }}
Notice that there must be quotes around the environment variable name when it is a literal string.
Jinjanator supports customizing Jinja2 template processing using two methods - via simple files containing custom filters or tests, or via a more advanced "customizations" file that allows you to do all of the above as well as modify core configuration of the Jinja2 engine.
The simplest way to add additional filters or tests is via "filters" and "tests" files. These files contain Python source code consisting of simple functions. Each function becomes a filter or test.
Examples:
filters.py
# Simple filters file
def parentheses(message):
""" Put message in parenthesis """
return f"({message})"
tests.py
# Example of simple tests file
def an_odd_number(number):
""" test if number is odd """
return True if (number % 2) else False
And a template that uses them:
{% for x in range(4) %}
{{x}} is: {% if x is an_odd_number %}
{{- "odd" | parentheses }}
{%- else %}
{{- "even" | parentheses }}
{%- endif %}
{%- endfor %}
The output is:
$ jinjanate --filter ./filters.py --test ./tests.py simple.j2
0 is: (even)
1 is: (odd)
2 is: (even)
3 is: (odd)
You can include multiple functions in each file and/or use multiple files as needed.
A more advanced way to customize your template processing is by using a "customizations" file.
Customizations files allow you to:
- Pass additional keywords to the Jinja2 environment
- Modify the context before it is used for rendering
- Register custom filters and tests
This is done through hooks that you implement in a customization file in Python code. Each hook is a plain function at the module level with the exact name as shown below.
The following hooks are available:
-
j2_environment_params() -> dict
: returns adict
of additional parameters for Jinja2 Environment. -
j2_environment(env: Environment) -> Environment
: lets you customize theEnvironment
object. -
alter_context(context: dict) -> dict
: lets you modify the context variables that are going to be used for template rendering. You can do all sorts of pre-processing here. -
extra_filters() -> dict
: returns adict
with extra filters for Jinja2 -
extra_tests() -> dict
: returns adict
with extra tests for Jinja2
All of them are optional.
The example customization.py file
for your reference:
#
# Example customization.py file for jinjanator
# Contains hooks that modify the way Jinja2 is initialized and used
def j2_environment_params():
""" Extra parameters for the Jinja2 Environment """
# Jinja2 Environment configuration
# https://jinja.pocoo.org/docs/2.10/api/#jinja2.Environment
return dict(
# Just some examples
# Change block start/end strings
block_start_string='<%',
block_end_string='%>',
# Change variable strings
variable_start_string='<<',
variable_end_string='>>',
# Remove whitespace around blocks
trim_blocks=True,
lstrip_blocks=True,
# Enable line statements:
# http://jinja.pocoo.org/docs/2.10/templates/#line-statements
line_statement_prefix='#',
# Keep \n at the end of a file
keep_trailing_newline=True,
# Enable custom extensions
# http://jinja.pocoo.org/docs/2.10/extensions/#jinja-extensions
extensions=('jinja2.ext.i18n',),
)
def j2_environment(env):
""" Modify Jinja2 environment
:param env: jinja2.environment.Environment
:rtype: jinja2.environment.Environment
"""
env.globals.update(
my_function=lambda v: 'my function says "{}"'.format(v)
)
return env
def alter_context(context):
""" Modify the context and return it """
# An extra variable
context['ADD'] = '127'
return context
def extra_filters():
""" Declare some custom filters.
Returns: dict(name = function)
"""
return dict(
# Example: {{ var | parentheses }}
parentheses=lambda t: '(' + t + ')',
)
def extra_tests():
""" Declare some custom tests
Returns: dict(name = function)
"""
return dict(
# Example: {% if a|int is custom_odd %}odd{% endif %}
custom_odd=lambda n: True if (n % 2) else False
)
If you'd like to chat with the jinjanator community, join us on Matrix!
This tool was created from j2cli, which itself was created from jinja2-cli. It was created to bring the project up to 'modern' Python coding, packaging, and project-management standards, and to support plugins to provide extensibility.
"Standing on the shoulders of giants" could not be more true than it is in the Python community; this project relies on many wonderful tools and libraries produced by the global open source software community, in addition to Python itself. I've listed many of them below, but if I've overlooked any please do not be offended :-)