Bad JSON to rST tables converter

Inelegantly converts JSON files to tables in rST files.

To run:

• Clone the repository: git clone https://github.com/zeddee/bad-json-to-rst-tables
• Navigate to the cloned dir: cd bad-json-to-rst-tables
• Run the Python package (because I am bad at packaging): python3 json2rst.py --input <inputdir> [--output <outputdir>]

Usage

usage: json2rst.py [-h] --input INFILES [--output OUTDIR] --headers
               PIVOT_HEADERS [--strict] [--sort-by SORT_BY]
               [--sort-order SORT_ORDER] [--csv-out CSV_OUT]
               [pivot]

optional arguments:
  -h, --help            show this help message and exit

General options:
  --input INFILES       (Required) Input JSON file, or a directory containing JSON files.
  --output OUTDIR       Output directory. Defaults to current directory.

Pivot a directory of JSON files:
  pivot                 Specify 'pivot' to pivot JSON files. Collects JSON
                        files from --input,and extract values from fields that
                        match names in --header, and write to a csv-table.
  --headers PIVOT_HEADERS
                        Required for pivot. Add a list of header names as
                        comma-separated values. JSON files from --input will
                        be pivoted against this list. Headers MUST BE UNIQUE
                        (RFC8259 §4). Duplicate headers are discarded.

                        E.g.: --headers='key1,key2,key3'
  --strict              Strict mode for pivot. When set, JSON files must have
                        all fields specified with --headers.
  --sort-by SORT_BY     Sort the pivot table by a given key. Specify only one
                        key.
  --sort-order SORT_ORDER
                        Sort --sort-by in 'ascending' or 'descending' order.
  --csv-out CSV_OUT     Name of output CSV file saved in --output dir.

Page titles

Page titles are crudely implemented for now.

A page title using the filename of the JSON file is added to the top of each rST file.

For example, running python json2rst.py --input sample.json --output _output creates the file _output/sample.rst which has the following first few lines:

sample
********

..  list-table::

    //...

Rich content

You can add rich content to a JSON field as a list of dicts(key-value pairs):

{
  "this rich content field": [
          {"h1" : "Heading 1"},
          {"h2" : "Heading 2"},
          {"ul" : "this is an unordered list item"},
          {"ol" : "this is an ordered list item"},
          {"p" : "this is a paragraph"},
          {"code" : "print(\"this is a block of code (single-line)\")"},
          {"code-block": "print(\"this is a single line in a code block\")"},
          {"image" : "/img/sample.jpg"}
    ]
}

Expected input

{
    "ID": "this is a title",
    "CVE": "\-",
    "Description": "This is a description",
    "Assessment": [
      {"heading": "This is a heading."},
      {"p": "This is a multi\nline field that can contain things like `links <#link>`_\nand code ``snippets``"},
      {"ul": "this is an unordered list item"},
      {"heading": "This is a heading."},
      {"ul": "this is an unordered list item"},
      {"ul": "this is an unordered list item"}
    ]
}

Expected output

..  list-table::
    :stub-columns: 1

    - * ID
      * this is a title

    - * CVE
      * \-

    - * Description
      * This is a description

    - * Assessment
      * **This is a heading.**

        This is a multi
        line field that can contain things like `links <#link>`_
        and code ``snippets``


        - this is an unordered list item

        **This is a heading.**

        - this is an unordered list item

        - this is an unordered list item

This is rendered as:

ID	this is a title
CVE	-
Description	This is a description
Assessment	This is a heading. This is a multi line field that can contain things like links and code snippets this is an unordered list item This is a heading. this is an unordered list item this is an unordered list item

Pivots

If you run python json2rst.py pivot ..., you can pivot a list of JSON files against field names specified with --headers "key1,key2,key3" and write it to a csv file (default: ./pivot.csv).

• DOES NOT SUPPORT Rich content. CSV files are simple creatures and we should let them be.
• You can sort the CSV list by specifying --sort-by "key_name".
• Sort by (--sort-by) descending or ascending order (default: ascending).
• Apply --strict mode so the pivot fails if at least one JSON file does not contain all the keys specified in --headers.

For example, running:

python json2rst.py \
  pivot \
  --input tests/samples \
  --headers="ID,Description" \
  --csv-out "test.csv" \
  --sort-by "ID" \
  --sort-order "descending"

Writes a pivot table named test.csv that looks like this:

ID,Description
EIQ-2021-1235,This is a description
EIQ-2021-1234,This is a description

You can embed it in an rST file like this:

.. csv-table::
   :header-rows: 1
   :file: ./test.csv

That is rendered like this:

ID	Description
EIQ-2021-1235	This is a description
EIQ-2021-1234	This is a description