ENH: RST support

[R5dan]

Feature Type

• Adding new functionality to pandas

• Changing existing functionality in pandas

• Removing existing functionality in pandas

Problem Description

I wish I could use ReStructured Text with pandas

Feature Description

The end users code:

import pandas as pd
df=pd.read_rst(rst)
df.to_rst()

I believe tabulate has a way to do this.

Alternative Solutions

I also built a way to make rst tables.

Additional Context

• The RST docs

I think Grid Tables would be best for pandas (or Simple Tables)

I did not use sudo-code in the examples due to complexity and that examples of how to do this can be seen in the above packages. See RST docs for what they look like.

[haiyashah]

Assign

[haiyashah]

import pandas as pd
from tabulate import tabulate
from docutils.parsers.rst import tableparser
from docutils.statemachine import StringList

def read_rst(rst_string):
parser = tableparser.GridTableParser()
table_data = parser.parse(StringList(rst_string.splitlines()))
header = ["".join(cell) for cell in table_data[0]]
rows = [[cell[0] for cell in row] for row in table_data[1]]
return pd.DataFrame(rows, columns=header)

def to_rst(df):
return tabulate(df, headers="keys", tablefmt="grid")

rst_table = """
+-----+-------+
| ID | Name |
+=====+=======+
| 1 | Alice |
+-----+-------+
| 2 | Bob |
+-----+-------+
"""

df = read_rst(rst_table)
print(df)

rst_output = to_rst(df)
print(rst_output)

[R5dan]

I'm not saying its not possible to transform them, but its not as simple as it could be. It should be able to be done inside pandas

rst_table = """ +-----+-------+ | ID | Name | +=====+=======+ | 1 | Alice | +-----+-------+ | 2 | Bob | +-----+-------+ """

Thats an invalid RST table, so shouldn't be parsable.

[R5dan]

Does anyone else have any suggestions?