Version 0.2.0 - Embedding Extension & Base64 Filter update

Version 0.2.0 - Embedding Extension & Base64 Filter update

Added

• The EmbedExtension to the Jinja2 environment used that enables using the
  {% embed 'my_file.txt' %} to embed the requested file "as-is" (so not
  parsed as a template or anything like that) into templates wholesale with
  these optional arguments:
  • indent (default: 0) to indent each line of the embedded file by a certain
    number of spaces, e.g. for embedding into YAML files, where indent is
    syntactically important
  • alviss (default: False) to run the embedded file though the Alviss
    "render_loader" before embedding the results
  • env (default: False) to make the Alviss parser render ${__ENV__:...}
    expressions
  • fidelius (default: False) to make the Alviss parser render
    ${__FID__:...} expressions
• The main use case here was considered things like creating Kubernetes File
  Config maps where e.g. config file content is embedded wholesale into the
  manifest's YAML
• The base64 filter to the Jinja2 environment used to quicly encode context
  variables as base64 encoded strings
  • The main use case here was considered things like rendering of Kubernetes
    Secret manifests where keys need to be base64 encoded
• A bunch of "proper" unittests for the new embedding and filter functions

Changed

• The FileTemplate now uses the Jinja2 environment's FileSystemLoader,
  pre-seeded with search paths of the current working directory (os.getcwd())
  as well as the root directory of the script being executed (sys.argv[0])
  plus whatever paths are defined in the STENCIL_TEMPLATE_PATH environment
  variable, if any (multiple paths seperated by ;)