print dataframe to be used in markdownhere

Question

I would like to print dataframe to markdown and use it in markdownhere.

I saw to_markdown() api but the format not work in markdownhere.

So I try below code, it works fine:

import pandas as pd

def dumpmd(df):
    out = []
    out.append("|".join(df.columns))
    r = []
    for name in df.columns:
        r.append(":-")
    out.append("|".join(r))    
    for i,row in df.iterrows():
        r = []
        r1 = []
        for e in row:
            r.append(str(e))
        out.append("|".join(r))
    out = "\n".join(out)
    return out


data = [['1','B','C'],
    ['2','E','F'],
    ['3','H','I']]


df = pd.DataFrame(data,columns=["C1" ,"C2","C3"])

print(df.to_markdown(tablefmt='pipe',index=False))
print(dumpmd(df))

Output:

| C1 | C2 | C3 |
|----+----+----|
|  1 | B  | C  |
|  2 | E  | F  |
|  3 | H  | I  |
C1|C2|C3
:-|:-|:-
1|B|C
2|E|F
3|H|I

Would anyone help simplify this code or any better solution to output expected markdown table from dataframe?

Answer 1

I proposed you three methods to do this :

(My preference is to_markdown method)

Source dataframe

import pandas as pd

data = [['1','B','C'],
        ['2','E','F'],
        ['3','H','I']]

df = pd.DataFrame(data,columns=["C1" ,"C2","C3"])

Method 1 : using a specialized function

Could be done the hard way with a writing function :

def dumpmd(df):
    out = ["|" + "|".join(df.columns) + "|"]
    out.append("|" + "|".join([":-" for _ in df.columns]) + "|")
    out.extend("|" + "|".join(map(str, row)) + "|" for _, row in df.iterrows())
    return "\n".join(out)


print(dumpmd(df))

|C1|C2|C3|
|:-|:-|:-|
|1|B|C|
|2|E|F|
|3|H|I|

Method 2 : Pandas markdown method

Use to_markdown() Pandas function directly, specifying the appropriate array format for Markdown.

df.to_markdown(tablefmt='pipe')

'|    |   C1 | C2   | C3   |\n|---:|-----:|:-----|:-----|\n|  0 |    1 | B    | C    |\n|  1 |    2 | E    | F    |\n|  2 |    3 | H    | I    |'

Another interesting tablefmt parameter is gridfor instance for rendering :

df.to_markdown(tablefmt='grid')

'+----+------+------+------+\n|    |   C1 | C2   | C3   |\n+====+======+======+======+\n|  0 |    1 | B    | C    |\n+----+------+------+------+\n|  1 |    2 | E    | F    |\n+----+------+------+------+\n|  2 |    3 | H    | I    |\n+----+------+------+------+'

cool grid rendering :

+----+------+------+------+
|    |   C1 | C2   | C3   |
+====+======+======+======+
|  0 |    1 | B    | C    |
+----+------+------+------+
|  1 |    2 | E    | F    |
+----+------+------+------+
|  2 |    3 | H    | I    |
+----+------+------+------+

Method 3 : tabulate library

Use the tabulate library to convert the DataFrame to a tabulated string, then convert it to Markdown.

from tabulate import tabulate

# Convert the DataFrame to a tabbed string
tabular_data = tabulate(df, headers='keys', tablefmt='pipe')

# Add headers and footers to the table
markdown_data = f"| {' '.join(df.columns)} |\n| {'-' * len(' '.join(df.columns))} |\n{tabular_data}\n"

print(markdown_data)

'| C1 C2 C3 |\n| -------- |\n|    |   C1 | C2   | C3   |\n|---:|-----:|:-----|:-----|\n|  0 |    1 | B    | C    |\n|  1 |    2 | E    | F    |\n|  2 |    3 | H    | I    |\n'