Repository files navigation

This project is not maintained anymore.

Here are some actively maintained forks:

• https://github.com/kpfleming/jinjanator
• https://github.com/mattrobenolt/jinja2-cli
• https://github.com/m000/jj2cli

j2cli is a command-line tool for templating in shell-scripts,leveraging theJinja2 library.

Features:

• Jinja2 templating
• INI, YAML, JSON data sources supported
• Allows the use of environment variables in templates! HelloDocker :)

Inspired bymattrobenolt/jinja2-cli

pip install j2cli

To enable the YAML support withpyyaml:

pip install j2cli[yaml]

Suppose, you want to 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:

$ j2 -f json nginx.j2 nginx.json> nginx.conf

The output is saved tonginx.conf:

server {  listen 80;  server_name localhost;  root /var/www/project;  index index.htm;}

Alternatively, you can use the-o nginx.conf option.

Suppose, you have a very simple template,person.xml:

<data><name>{{ name }}</name><age>{{ age }}</age></data>

What is the easiest way to use j2 here?Use environment variables in your bash script:

$export name=Andrew$export age=31$ j2 /tmp/person.xml<data><name>Andrew</name><age>31</age></data>

Even when you use yaml or json as the data source, you can always access environment variablesusing theenv() function:

Username: {{ login }}Password: {{ env("APP_PASSWORD") }}

Compile a template using INI-file data source:

$ j2 config.j2 data.ini

Compile using JSON data source:

$ j2 config.j2 data.json

Compile using YAML data source (requires PyYAML):

$ j2 config.j2 data.yaml

Compile using JSON data on stdin:

$ curl http://example.com/service.json | j2 --format=json config.j2

Compile using environment variables (hello Docker!):

$ j2 config.j2

Or even read environment variables from a file:

$ j2 --format=env config.j2 data.env

Or pipe it: (note that you'll have to use the "-" in this particular case):

$ j2 --format=env config.j2 - < data.env

j2 accepts the following arguments:

• template: Jinja2 template file to render
• data: (optional) path to the data used for rendering.The default is-: use stdin. Specify it explicitly when using env!

Options:

• --format, -f: format for the data file. The default is?: guess from file extension.

• --import-env VAR, -e EVAR: 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!)

• -o outfile: Write rendered template to a file

• --undefined: Allow undefined variables to be used in templates (no error will be raised)

• --filters filters.py: Load custom Jinja2 filters and tests from a Python file.Will load all top-level functions and register them as filters.This option can be used multiple times to import several files.

• --tests tests.py: Load custom Jinja2 filters and tests from a Python file.

• --customize custom.py: A Python file that implements hooks to fine-tune the j2cli behavior.This is fairly advanced stuff, use it only if you really need to customize the way Jinja2 is initialized.SeeCustomization for more info.

There is some special behavior with environment variables:

• Whendata is not provided (data is-),--format defaults toenv and thus reads environment variables
• When--format=env, it can read a special "environment variables" file made like this:env > /tmp/file.env

Data input from environment variables.

Render directly from the current environment variable values:

$ j2 config.j2

Or alternatively, read the values from a dotenv file:

NGINX_HOSTNAME=localhostNGINX_WEBROOT=/var/www/projectNGINX_LOGS=/var/log/nginx/

And render with:

$ j2 config.j2 data.env$ env | j2 --format=env config.j2

If you're going to pipe a dotenv file intoj2, you'll need to use "-" as the second argument to explicitly:

$ j2 config.j2 - < data.env

INI data input format.

data.ini:

[nginx]hostname=localhostwebroot=/var/www/projectlogs=/var/log/nginx/

Usage:

$ j2 config.j2 data.ini$ cat data.ini | j2 --format=ini config.j2

JSON data input format

data.json:

{    "nginx":{        "hostname": "localhost",        "webroot": "/var/www/project",        "logs": "/var/log/nginx/"    }}

Usage:

$ j2 config.j2 data.json$ cat data.json | j2 --format=ini config.j2

YAML data input format.

data.yaml:

nginx:  hostname: localhost  webroot: /var/www/project  logs: /var/log/nginx

Usage:

$ j2 config.j2 data.yml$ cat data.yml | j2 --format=yaml config.j2

Given a Docker Link environment variable value, format it into something else.

This first parses a Docker Link value like this:

DB_PORT=tcp://172.17.0.5:5432

Into a dict:

{'proto':'tcp','addr':'172.17.0.5','port':'5432'}

And then usesformat to format it, where the default format is '{addr}:{port}'.

More info here:Docker Links

Use an environment variable's value inside your template.

This filter is available even when your data source is something other that the environment.

Example:

User: {{ user_login }}Pass: {{ "USER_PASSWORD"|env }}

You can provide the default value:

Pass: {{ "USER_PASSWORD"|env("-none-") }}

For your convenience, it's also available as a function:

User: {{ user_login }}Pass: {{ env("USER_PASSWORD") }}

Notice that there must be quotes around the environment variable name

j2cli now allows you to customize the way the application is initialized:

• Pass additional keywords to Jinja2 environment
• Modify the context before it's used for rendering
• Register custom filters and tests

This is done throughhooks that you implement in a customization file in Python language.Just plain functions at the module level.

The following hooks are available:

• j2_environment_params() -> dict: returns adict of additional parameters forJinja2 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 beused 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 customize.py file for j2cli# Contains potional hooks that modify the way j2cli is initializeddefj2_environment_params():""" Extra parameters for the Jinja2 Environment """# Jinja2 Environment configuration# http://jinja.pocoo.org/docs/2.10/api/#jinja2.Environmentreturndict(# Just some examples# Change block start/end stringsblock_start_string='<%',block_end_string='%>',# Change variable stringsvariable_start_string='<<',variable_end_string='>>',# Remove whitespace around blockstrim_blocks=True,lstrip_blocks=True,# Enable line statements:# http://jinja.pocoo.org/docs/2.10/templates/#line-statementsline_statement_prefix='#',# Keep \n at the end of a filekeep_trailing_newline=True,# Enable custom extensions# http://jinja.pocoo.org/docs/2.10/extensions/#jinja-extensionsextensions=('jinja2.ext.i18n',),    )defj2_environment(env):""" Modify Jinja2 environment    :param env: jinja2.environment.Environment    :rtype: jinja2.environment.Environment    """env.globals.update(my_function=lambdav:'my function says "{}"'.format(v)    )returnenvdefalter_context(context):""" Modify the context and return it """# An extra variablecontext['ADD']='127'returncontextdefextra_filters():""" Declare some custom filters.        Returns: dict(name = function)    """returndict(# Example: {{ var | parentheses }}parentheses=lambdat:'('+t+')',    )defextra_tests():""" Declare some custom tests        Returns: dict(name = function)    """returndict(# Example: {% if a|int is custom_odd %}odd{% endif %}custom_odd=lambdan:Trueif (n%2)elseFalse    )#