Changelog

Name changes

Over the course of development the project name has changed. Previous names include:

History

Version 1.2.4 (2025-06-05)

Documentation and README cleanup and fixing errors (spelling, examples, etc.).
Added Basic Auth to serve tool.
Extending find by using paths for relative matching.
Updating links to new repository.
Added new alias rm to obj del.
Removed module_import and class_definition from UnsafeRickle.
Renamed function to python to reflect functionality better, leaving function open to safer sandboxed scripting.
Dismantling ObjectRickler into functions object_to_rickle and rickle_to_object.
RICKLE_OPENING_BRACES and RICKLE_CLOSING_BRACES dropped in favor of single RICKLE_HANDLEBARS with default {{}}.
Remove from_file, from_csv, api_json in favour of file, csv, api.
Dropped add_html as the same could be achieved through api.
Removed Schema and Convert classes from tools.
Created new schema package to contain all schema related functionality.
Renamed CLI tool obj_func to obj_python_func to reflect future changes.
Added PUT method to serve tool to now be able to create new paths with values.
Added x-rickle-output-type header to serve so that different output types can be requested.

Version 1.2.3 (2025-03-25)

Adding ability to load list (array) instead of dict (object), opening way for lists of configs.
New list type internal representation "array".
Added new put method for putting values in Rickle with any path. Where the path does not exist, the path will be created.
Changed set to only mutate paths that already exist. Creating clear distinction between set and put.
Added multi document YAML support.
Added JSONL support.
Added new find CLI tool to find key / value pairs.

Version 1.2.2 (2025-02-17)

Rename of entire project from rickled to rickle!
Fixed bug in CLI tool not able to run
Made provider_access_key optional for secret type.
Can now use list type in path method, for example [0]/path/to will reference the first object in the list.

Version 1.2.1 (2025-02-12)

Major additions and changes, bumped version to 1.2.
Init args now parsed with opening braces {{ and closing braces }}.
Opening and closing braces can be defined using environment variables RICKLE_OPENING_BRACES and RICKLE_CLOSING_BRACES.
RICKLE_OPENING_BRACES and RICKLE_CLOSING_BRACES can also be passed as init arguments.
Added optional INI support.
Added optional .ENV support if extras installed pip install rickled[dotenv].
Added optional XML support if extras installed pip install rickled[xml].
Renamed optional extras for twisted, now installed pip install rickled[net].
With added ability to use XML, keys are now cleansed of non alphabetic characters, but kept for serialisation.
Added more customisation. Path separator can now be define with init arg RICKLE_PATH_SEP or environment variable RICKLE_PATH_SEP.
Added list type index getter to paths, for example /path/to/a_list[0] returns first element (if a_list is of list type)
Added static method flatten_dict to squash a Python dictionary into a thin dictionary with depth of 1.
Undo the flattening of a dictionary with static method inflate_dict.
Now able to dump to INI file or string using to_ini.
When reading INI, dictionaries are automatically inflated, split on RICKLE_INI_PATH_SEP env var or init arg.
Furthermore, RICKLE_INI_OPENING_BRACES and RICKLE_INI_CLOSING_BRACES can be set with env var or init arg.
Removed unsafe code and usage into separate class, UnsafeRickle.
Added optional schema validator if extras installed (only for Python>3.8) pip install rickled[validators].
Adding ability to pipe into CLI tool, i.e. perform cat file.json | rickle conv -x json -t yaml.
Added a helper classify_string to tools to classify a string type.
Renaming optional arguments for CLI in order to be more clear on their meaning.
-i becomes --input
-d becomes --dir
-o becomes --output
-t becomes --output-type
Added full to install extras to install all extras.
Alternative for from_file type is now file. Will be deprecated in the future.
Alternative for from_csv type is now csv. Will be deprecated in the future.
Deprecating add_env_variable for add_env.
Deprecating add_csv_file for add_csv.
Deprecating add_from_file for add_from.
Deprecating add_api_json_call for add_api_json.
Deprecating add_attr for add.
New init arg RICKLE_NAME_CLEAN_UP to skip cleaning key names of disallowed chars.
Fixed logical flaw in dict where cleaned up keys not being used.
Added helper mapping get_native_type_name to map Python type names to format native names.
Renamed schema types to conform more to JSON schema, i.e. dict becomes object, list becomes array etc.
Added min and max to array type in schema validation.
Added ability to use jsonschema instead in Schema.schema_validation by setting use_json_schema=True if jsonschema is installed.
Added --extras to schema generation to add things like "required" etc.
Renamed schema to properties for objects in default rickle schema.
Renamed schema to items for objects in default rickle schema.
Added add_secret to Rickle, now able to add secrets from cloud providers aws, google, azure.
Added add_random_value to Rickle, using generate_random_value in tools. Useful for mock data.
Moved --output-type to the main argument parser rickle.

Version 1.1.3 (2024-05-05)

New addition: TOML now supported

Version 1.1.2 (2024-04-23)

Minor fix in schema tool where bool mistaken as int.
Added CLI tools documentation.
Fixed bug when using CLI tool and twisted not installed, better error handling in CLI tools.
Changed the default serialised=False for the to YAML and JSON file/string methods.
Changed default behaviour of serve to return deserialised form.
Added 404 response to serve when path can not be traversed.
Added -s to serve to serve the file in serialised form.
Added -t to serve to serve the file in output type JSON or YAML (JSON by default).
Changed -f to -i in serve tool for consistency.

Version 1.1.1 (2024-03-19)

Updated documentation and links to pages.

Version 1.1.0 (2024-02-05)

Added set to BaseRickle, as a equivalent to the get method.
Updated the get method to handle document paths.
Added remove method to delete items from object.
Added the __delitem__ dunder, now able to del rickle['key'].
Fixed bug where search_path was exiting loops in dictionary too early (after first find).
Implemented type guessing of params for functions when using path like calling.
Added ability to add params to callable. Example rickle('/path/to/func', x=1, y='str').
Bug fix in trying to parse URL by user deajan, commit 7cb1773.

Version 1.0.2 (2023-12-12)

Added strict boolean parameter for allowing properties that are reserved keywords.

Version 1.0.1 (2023-10-03)

Added the ability to load a Rickle from URL at init.
Added the -b flag to serve CLI tool to open host on browser.
Bug fix in infer_read_file_type when reading unknown file suffix.
Renamed the -i flag to -a in serve CLI tool.

Version 1.0.0 (2023-09-20)

Added the schema validation tool schema in tools.
Added all CLI tools.
Now releasing version 1.0.0

Version 0.3.5 (2023-09-09)

Added the first of the rickled.tools, the converter.
Added first CLI tools rickle conv and rickle serve.

Version 0.3.4 (2023-07-20)

Fixed error when importing from rickled.net when openssl is not installed.

Version 0.3.3 (2023-07-20)

Adding optional installation of twisted library.
Added serve_rickle_http and serve_rickle_https to rickled.net to serve Rickles as REST API.

Version 0.3.2 (2023-04-07)

When calling dict on rickle, hot loaded items were not being serialised. Fixed.

Version 0.3.1 (2023-04-02)

Fixed issue for path based query, where Rickle objects are considered callable (rightfully).
Uses inspect.isfunction instead of callable.
Added meta to base class for getting metadata of a property.

Version 0.3.0 (2023-02-16)

Bumped up to minor 3, close to releasing version 1.0 after http server is implemented.
Added the hot_load property to API calls, making it load on call instead of only on start.
Added the hot_load property to HTML page, making it load on call instead of only on start.
Added the hot_load property to add file, making it load on call instead of only on start.

Version 0.2.7 (2023-02-15)

-Complete revamp of internal versioning.

Version 0.2.6 (2023-02-15)

Fixed the same bug, but the root cause. The fact that modules are imported before proper install.

Version 0.2.5 (2023-01-18)

Fixed a bug where requests is not installed

Version 0.2.4 (2022-09-02)

Added ability to get nodes by using Unix style paths to get to nodes.
Added a safe load environment variable "RICKLE_SAFE_LOAD" to override all lambda loads (as a safety measure).
Added search_path to search for a key in the Rickle.
Removed includes_self_reference due to confusion.
Added a third way to load CSV files, see example documentation.
Added load_as_rick to add_api_json_call.

Version 0.2.3 (2022-03-13)

Merged but cleaned up contributions by Fabian.

Version 0.2.2 (2022-02-14)

Added do_recursive param to .get to optionally do a deeper recursive search.
Do you agree that valentine's day is bullshit? Because my gf doesn't.

Version 0.2.1 (2021-12-08)

Added add_class_definition to define classes.
Created a new class, ObjectRickler, to dump (almost) any object or convert to Rickle.
Added add_module_import to Rickle, with functionality to add global Python module imports.

Version 0.2.0 (2021-12-06)

Renamed project to Rickled to avoid any possible lawsuits from money hungry media execs.
Pickle Rick was a great name, possibly even considered a parody which is protected under copyright law.
But rather safe than sued...

Version 0.1.14 (2021-10-28)

Added new add_html_page to load HTML text.
Added new add_csv_file to load CSV files as either a list of lists, or list of PickleRicks.

Version 0.1.13 (2021-10-07)

-Added ability to load from multiple YAML files or JSON files from start up.

Version 0.1.12 (2021-09-23)

Fixed major bug, YAML was not loaded!
Adding preload arguments for load and replace values within YAML files using _|PARAM|_
Added new API JSON call method, to load and create a Rick from an API response add_api_json_call.
Added new ability to load other YAML, JSON, or text files from within, using add_from_file.
Added add_base64 to load base 64 encoded data.