feature request: new command csv2rst

[peterjc]

Prerequisites

• make sure you're are using the latest version by csvtk version
• read the usage

Describe your issue

• describe the problem
• provide a reproducible example

Description

I would like to be able to turn CSV or TSV files etc into reStructuredText (RST) tables.
See https://docs.anaconda.com/restructuredtext/detailed/#tables or https://docutils.sourceforge.io/docs/user/rst/quickref.html#tables

Example

Consider a simple input table, I personally prefer tabs:

$ csvtk tab2csv names.tsv 
id,first_name,last_name,username
11,Rob,Pike,rob
2,Ken,Thompson,ken
4,Robert,Griesemer,gri
1,Robert,Thompson,abc
NA,Robert,Abel,123

You already support markdown output:

$ csvtk csv2md -t names.tsv 
id |first_name|last_name|username
:--|:---------|:--------|:-------
11 |Rob       |Pike     |rob
2  |Ken       |Thompson |ken
4  |Robert    |Griesemer|gri
1  |Robert    |Thompson |abc
NA |Robert    |Abel     |123

Rendered here with GitHub,

id	first_name	last_name	username
11	Rob	Pike	rob
2	Ken	Thompson	ken
4	Robert	Griesemer	gri
1	Robert	Thompson	abc
NA	Robert	Abel	123

My desired functionality is just the simplest RST table markup:

$ csvtk csv2rst-t names.tsv 
== ========== ========= ========
id first_name last_name username
== ========== ========= ========
11 Rob        Pike      rob
2  Ken        Thompson  ken
4  Robert     Griesemer gri
1  Robert     Thompson  abc
NA Robert     Abel      123
== ========== ========= ========

That is close to the existing pretty output:

$ csvtk pretty -t names.tsv -s " "
...

I've not looked at the rest of csvtk to say if there is value in making an RST grid table, it is more verbose but allows more rich entries:

+----+------------+-----------+----------+
| id | first_name | last_name | username |
+====+============+===========+==========+
| 11 | Rob        | Pike      | rob      |
+----+------------+-----------+----------+
| 2  | Ken        | Thompson  | ken      |
+----+------------+-----------+----------+
| 4  | Robert     | Griesemer | gri      |
+----+------------+-----------+----------+
| 1  | Robert     | Thompson  | abc      |
+----+------------+-----------+----------+
| NA | Robert     | Abel      | 123      |
+----+------------+-----------+----------+

[shenwei356]

Here it is.

• csvtk_linux_amd64.tar.gz
• csvtk_darwin_amd64.tar.gz

But it can't handle unicode (utf-f8) well, tested at http://rst.aaroniles.net/

Usage

convert CSV to readable aligned table

Attention:

  1. row span is not supported.

Usage:
  csvtk csv2rst [flags]

Flags:
  -k, --cross string               charactor of cross (default "+")
  -s, --header string              charactor of separator between header row and data rowws (default "=")
  -h, --help                       help for csv2rst
  -b, --horizontal-border string   charactor of horizontal border (default "-")
  -p, --padding string             charactor of padding (default " ")
  -B, --vertical-border string     charactor of vertical border (default "|")

Example

1. With header row

    $ csvtk csv2rst testdata/names.csv 
    +----+------------+-----------+----------+
    | id | first_name | last_name | username |
    +====+============+===========+==========+
    | 11 | Rob        | Pike      | rob      |
    +----+------------+-----------+----------+
    | 2  | Ken        | Thompson  | ken      |
    +----+------------+-----------+----------+
    | 4  | Robert     | Griesemer | gri      |
    +----+------------+-----------+----------+
    | 1  | Robert     | Thompson  | abc      |
    +----+------------+-----------+----------+
    | NA | Robert     | Abel      | 123      |
    +----+------------+-----------+----------+
   

2. No header row

    $ csvtk csv2rst -H -t  testdata/digitals.tsv 
    +---+-------+---+
    | 4 | 5     | 6 |
    +---+-------+---+
    | 1 | 2     | 3 |
    +---+-------+---+
    | 7 | 8     | 0 |
    +---+-------+---+
    | 8 | 1,000 | 4 |
    +---+-------+---+
   

3. Misc

    $ cat testdata/names.csv | head -n 1 | csvtk csv2rst 
    +----+------------+-----------+----------+
    | id | first_name | last_name | username |
    +====+============+===========+==========+
    
    $ cat testdata/names.csv | head -n 1 | csvtk csv2rst -H
    +----+------------+-----------+----------+
    | id | first_name | last_name | username |
    +----+------------+-----------+----------+
    
    $ echo | csvtk csv2rst -H
    [ERRO] xopen: no content
    
    $ echo "a" | csvtk csv2rst -H
    +---+
    | a |
    +---+
    
    # some online rst render reports error
    $ echo "沈伟" | csvtk csv2rst -H
    +--------+
    | 沈伟 |
    +--------+
   

[peterjc]

Wow - that was quick. Thank you.

The "rich" RST table syntax does seem more flexible, and will cope with single rows etc. So I understand why you might prefer that.

I was not aware of the RST table issue with unicode, it also happens on https://livesphinx.herokuapp.com/ which is another online test system for RST using Sphinx. Sometimes it works, e.g. emoji or a kanji:

+------------+
| Hello.     |
+----+-------+
| 😀 | Smile |
+----+-------+

+-----------+
| Hello.    |
+----+------+
| 金 | Gold |
+----+------+

I tried some Japanese hiragana, and like your example failed to get it to work regardless of the number of spaces (in case it was a length issue). I think this is probably a bug in the docutils library, perhaps the same thing this sphinx user was asking about: sphinx-doc/sphinx#6702

i.e. The unicode issue is out of your hands.

[shenwei356]

The unicode issue is fixed thanks to the go-runewidth package.

$ csvtk csv2rst testdata/unicode.csv 
+-------+---------+
| value | name    |
+=======+=========+
| 1     | 沈伟    |
+-------+---------+
| 2     | 沈伟b   |
+-------+---------+
| 3     | 沈小伟  |
+-------+---------+
| 4     | 沈小伟b |
+-------+---------+