Generate static HTML documentation from JSON schemas
Project description
JSON Schema for Humans
Quickly generate a beautiful static HTML or Markdown page documenting a JSON schema
Documentation (with visual examples)
Features
- Support for JSON Schema Draft-07
- Since the result is static, it is easier to host and faster to load
- HTML and Markdown output support
- Different templates to choose from. More details: HTML version - Markdown version
- Anchor links, allow to send a user to a specific section of the documentation
- Support for references (even circular!)
Installation
pip install json-schema-for-humans
Usage
Options for generation of the doc are documented using the library itself: HTML version - Markdown version
They can be supplied in various ways:
- Using a JSON or YAML configuration file with the CLI option
--config-file
- Using the CLI option
--config
- Using the
ConfigurationOption
object from code
More details are available in the appropriate sections below.
From CLI
generate-schema-doc [OPTIONS] SCHEMA_FILES_OR_DIR [RESULT_FILE_OR_DIR]
SCHEMA_FILES_OR_DIR
can be:
- a path to a single schema file;
- a path to a directory, in this case all files with extensions json, yaml, or yml will be used; or
- a comma-separated list of the above
All schemas provided must be a valid JSON Schema (in JSON or YAML format)
Examples:
my_schema.json
my_folder
my_folder/my_schema.yaml,another_schema.json
The default value for RESULT_FILE_OR_DIR
depends on the context:
- the current working directory if more than one schema as been provided as input
schema_doc.html
if rendering a single schema as HTMLschema_doc.md
if rendering a single schema as Markdown
In a case where more than one schema is provided as input, RESULT_FILE_OR_DIR
must be a directory. The output documentation will have the same name as the input schema, but with a different extension (html
or md
).
CLI options
--config
Supply generation config parameters. The parameters are documented in the JSON schema config_schema.json
at the root of the repo or see the generated doc: HTML version - Markdown version.
Each parameter is in the format --config parameter_name=parameter_value
. Example: --config expand_buttons=true
. The parameter value must be valid JSON.
For flags, you can also omit the value for true
or prefix the parameter name with no_
for false
. Example: --config expand_buttons
or --config no_expand_buttons
.
--config-file
Path to a JSON or YAML configuration file respecting the schema config_schema.json
.
Example: --config-file jsfh-conf.yaml
where jsfh-conf.yaml
is in the current directory and contains the following:
description_is_markdown: false
expand_buttons: true
copy_js: false
From code
To render schema documentation from code, the method to call is generate
with the following signature:
def generate(
schema_file_or_dir: List[Union[str, Path, TextIO]],
result_file_or_dir: Optional[Union[str, Path, TextIO]],
config: GenerationConfiguration = None,
loaded_schemas: Optional[Dict[str, Any]] = None,
) -> Tuple[List[Path], Dict[str, str]]:
...
Where:
schema_file_or_dir
is a list of schema file paths or open file pointerresult_file_or_dir
is the output path. If there is more than one schema to render, this should be a directory. The same rules as the CLI call documented above applies.config
is an instance of theGenerationConfiguration
object, seeThe GenerationConfiguration object
belowloaded_schemas
is a dictionary of already loaded schema files. SeePre-load schemas
below. This is an advanced option.
The output is a tuple where:
- the first element is a list of the paths to documentation files written. This part is only populated when there is an output path and the files are written to disk
- the second is a dictionary of the rendered schemas in the format input_schema_file_name -> rendered_documentation. This is only populated when there is no output path
Notes:
- When providing a file pointer, it must be open in read mode for the input schema and in write mode for the output documentation files
- If there is an output path and it is not disabled in the config, CSS and JS files are copied to the current working directory with names "schema_doc.css" and "schema_doc.min.js" respectively. These are needed to render the documentation HTML page in a browser
Example:
In this example, my_schema.json
will be rendered to my_schema.html
in the output directory
from json_schema_for_humans.generate import generate
from pathlib import Path
output_dir = Path.cwd() / "output_dir"
output_dir.mkdir(parents=True, exist_ok=True)
generated = generate(["my_schema.json"], output_dir)
# Returns ([Path('[cwd]/output_dir/my_schema.html')], {})
Shortcut methods
These methods are kept for retrocompatibility and because they are easier to call for simple use cases.
Method Name | Schema input | Output | CSS and JS copied? |
---|---|---|---|
generate_from_schema | schema_file as str, Path (from pathlib) or a file object |
Rendered HTML as a str | No |
generate_from_filename | schema_file_name as a str or Path |
Rendered HTML written to the file at path result_file_name |
Yes |
generate_from_file_object | schema_file as an open file object (read mode) |
Rendered HTML written to the file at result_file , which must be an open file object (in write mode) |
Yes |
Notes:
- When using file objects, it is assumed that files are opened with encoding "utf-8"
- CSS and JS files are copied to the current working directory with names "schema_doc.css" and "schema_doc.min.js" respectively
- Other parameters of these methods are analogous to the CLI parameters documented above.
The GenerationConfiguration object
To reduce the number of parameters to pass from function to function in the code, there is a GenerationConfiguration
object that should be used for providing options.
Example:
from json_schema_for_humans.generate import generate_from_filename
from json_schema_for_humans.generation_configuration import GenerationConfiguration
config = GenerationConfiguration(copy_css=False, expand_buttons=True)
generate_from_filename("my_schema.json", "schema_doc.html", config=config)
# Your doc is now in a file named "schema_doc.html". Next to it, "schema_doc.min.js" was copied, but not "schema_doc.css"
# Your doc will contain a "Expand all" and a "Collapse all" button at the top
Pre-load schemas
generate_from_schema
has a loaded_schemas
parameter that can be used to pre-load schemas. This must be a dict with the key being the real path of the schema file and the value being the result of loading the schema (with json.load
or yaml.safe_load
, for example).
This should not be necessary in normal scenarios.
What's supported
See the excellent Understanding JSON Schema to understand what are those checks
The following are supported:
- Types
- Regular expressions
- String length
- Numeric types multiples and range
- Constant and enumerated values
- Required properties
- Pattern properties
- Default values
- Array
minItems
,maxItems
,uniqueItems
,items
(schema that must apply to all of the array items), andcontains
- Combining schema with
oneOf
,allOf
,anyOf
, andnot
- Examples
- Conditional subschemas
These are not supported at the moment (PRs welcome!):
- String format
- Property names and size
- Array items at specific index (for example, first item must be a string and second must be an integer)
- Property dependencies
- Media
References
References are supported:
- To another part of the schema, e.g.
{ $ref: "#/definitions/something" }
- To a local file,
{"$ref": "references.json"}
,{"$ref": "references.json#/definitions/something"}
- To a URL,
{"$ref": "http://example.com/schema.json"}
,{"$ref": "http://example.com/schema.json#/definitions/something"}
You can have a description
next to a $ref
, it will be displayed in priority to the description from the referenced element.
If you have several attributes using the same definition, the definition will only be rendered once.
All other usages of the same definition will be replaced with an anchor link to the first render of the definition.
This can be turned off using --config no_link_to_reused_ref
. See With references
in the examples.
Templates
Templates control the style of the generated documentation.
js
This is the default template. It uses Bootstrap along with minimal Javascript to allow for the following:
- Properties are in expandable dynamic sections. You can include a button to expand or collapse all. (See doc: HTML version - Markdown version)
- Conditional subschemas (
anyOf
,oneOf
,allOf
) are in tabbed sections - Anchor links will scroll to, expand, and animate the target section
- Long descriptions are collapsed by default
When using this template, you need to include the Javascript file (schema_doc.min.js
) that is automatically copied next to the output HTML file (schema_doc.html
by default).
flat
Note: This template is a work in progress
It is sometimes not possible or desirable to include custom Javascript in documentation. This template addresses this issue by removing interactive elements in favor of simpler HTML.
At the moment, this means the whole documentation is generated without any collapsible sections, which may make it hard to understand the schema structure. Contributions are welcomed to improve it!
MD (Markdown)
Note: This template is a work in progress
This template allows users to publish the generated documentation without hosting an HTTP server.
On GitHub, this format is rendered directly when browsing code.
A table of content is provided at the beginning of the file for easy navigation.
You can display some important information as badge using an option. See doc: HTML version - Markdown version
Contributions are welcomed to improve it!
Contributing
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file json-schema-for-humans-0.35.3.tar.gz
.
File metadata
- Download URL: json-schema-for-humans-0.35.3.tar.gz
- Upload date:
- Size: 47.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.1.8 CPython/3.9.7 Linux/5.8.0-1042-azure
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 65152ba1dd88c3b96e5ea5b60e3bb4576e64ab2e287f1ff095ca5a4566ef5d8e |
|
MD5 | b47c43ca37b24f1306d64f08ef994d6a |
|
BLAKE2b-256 | df22b6dd7b5a92b23b6df3327cde3099b32897bf7ccba2525634b2f63ef55625 |
File details
Details for the file json_schema_for_humans-0.35.3-py3-none-any.whl
.
File metadata
- Download URL: json_schema_for_humans-0.35.3-py3-none-any.whl
- Upload date:
- Size: 65.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: poetry/1.1.8 CPython/3.9.7 Linux/5.8.0-1042-azure
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | c8cc73a3fb53c44b4b0da3373f4de0c470b03c517bbc3b0c11491b22a7971a58 |
|
MD5 | 1d07f7fd82580ce151e3f27eaf4c2741 |
|
BLAKE2b-256 | 91e8732d48f2e638a1916dc293fe51eee93b1643f341d3163b084a10da4b9be3 |