Skip to main content

Make pandas dataframe looking pretty again

Project description

pretty_html_table - Beautiful html tables made easy

pretty_html_table exists to convert a pandas DataFrame into a pretty html table for use in email. The intended target audience is anyone who needs to send reports via email and would like to make their tables look more attractive.

12 different color themes are available. The output of the package embeds nicely with other packages used to send html emails, such as email or O365. The html formatting is set at the DataFrame row level, which allows nearly every email provider to parse it. This obviates the need to grok out how the CSS may interact with the sending/recieving email provider.

Use pip to install the package:

pip install pretty_html_table

A simple example to load an Excel file to a pandas DataFrame, convert it to html, and then save to an html file:

from pretty_html_table import build_table

df = pd.read_excel('df.xlsx')
html_table_blue_light = build_table(df, 'blue_light')

# Save to html file
with open('pretty_table.html', 'w') as f:
    f.write(html_table_blue_light)

# Compare to the pandas .to_html method:
with open('pandas_table.html', 'w') as f:
    f.write(df.to_html())

Use any browser to open pretty_table.html to see how the table would appear in an html email.

Why choose pretty_html_table?

Output is ready to be sent via any Python package used to send emails. Insert the result of this package to the body of the email and voila.

List of colors available

Name font style Header Rows
'blue_light' Century Gothic Bold: yes / Background color: white / Font color: dark blue Odd background color: light blue / Even background color: white
'blue_dark' Century Gothic Bold: yes / Background color: dark blue / Font color: white Odd background color: light blue / Even background color: white
'grey_light' Century Gothic Bold: yes / Background color: white / Font color: dark grey Odd background color: light grey / Even background color: white
'grey_dark' Century Gothic Bold: yes / Background color: dark grey / Font color: white Odd background color: light grey / Even background color: white
'orange_light' Century Gothic Bold: yes / Background color: white / Font color: dark orange Odd background color: light orange / Even background color: white
'orange_dark' Century Gothic Bold: yes / Background color: dark orange / Font color: white Odd background color: light orange / Even background color: white
'yellow_light' Century Gothic Bold: yes / Background color: white / Font color: dark yellow Odd background color: light yellow / Even background color: white
'yellow_dark' Century Gothic Bold: yes / Background color: dark yellow / Font color: white Odd background color: light yellow / Even background color: white
'green_light' Century Gothic Bold: yes / Background color: white / Font color: dark green Odd background color: light green / Even background color: white
'green_dark' Century Gothic Bold: yes / Background color: dark green / Font color: white Odd background color: light green / Even background color: white
'red_light' Century Gothic Bold: yes / Background color: white / Font color: dark red Odd background color: light red / Even background color: white
'red_dark' Century Gothic Bold: yes / Background color: dark red / Font color: white Odd background color: light red / Even background color: white

Example of an integration with the O365 package

First, create a function to send an email:

from O365 import Account

# Never hard code credentials or store them in a repo
# Use environmental variables instead

credentials = (o365credid, o365credpwd)
account = Account(credentials)

def send_email(account, to, subject, start, body, end):
    m = account.new_message()
    m.to.add(to)
    m.subject = subject
    m.body = start + body + end
    m.send()

Then create the start and end of an email in html:

start = """<html>
                <body>
                    <strong>Data table here:</strong><br />"""


end = """       </body>
            </html>"""

Finally we can utilize pretty_table_html to convert our Excel file and send the email:

from pretty_html_table import build_table

html_table_blue_light = build_table(pd.read_excel('df.xlsx'), 'blue_light')

send_email(account
           , 'test@any.com'
           , 'test table'
           , start
           , html_table_blue_light
           , end
           )

Here are all of the currently available colors:

Light Dark

Additional arguments

Several optional arguments now exist that allow the user to control the table's font, font size, and alignment:

  • font_size - accepts absolute keywords (medium) and pixel values (20px)
  • font_family - best practice is to include a generic font family in case a recipient's client cannot render the chosen font. The example below designates Open Sans as a font, but designates the generic sans-serif family as a fallback. It's possible that the fallback font may be utilized in case a recipient has web fonts blocked for security reasons, or if they are viewing the email on a client that does not have acces to Google Fonts.
  • text_align - accepts standard html property values such as left, right, center, justify.
  • width - accepts string representation of pixels. For instance, for the columns to have a width of 100px, you would write: width="100px".
  • width_dict - accepts list of string representation of pixels. It will only work if the length of the list matches the number of columns of your pandas dataframe. You can for example provide the following argument to the fonction: width_dict=['300px','auto', 'auto', 'auto','auto', 'auto'] only the first column would be resized to 300px, the other would be "auto".
  • index - bolean. False by default - If you write index=True, index of the dataframe will then be visible in your table.
  • even_color - accepts string representation of colors (either "white" or "FFFFF"). For instance, for the font color of the even lines to be white, you would write: even_color='white'.
  • color - accepts string representation of colors (either "white" or "FFFFF"). For instance, for the background color of the even lines to be black, you would write: even_color='black'.
html_table = build_table(df
                        , 'yellow_dark'
                        , font_size='medium'
                        , font_family='Open Sans
                        , sans-serif'
                        , text_align='left'
                        , width='auto'
                        , index=False
			, even_color='black'
			, even_bg_color='white'
                        )

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pretty_html_table-0.9.12.tar.gz (5.7 kB view details)

Uploaded Source

Built Distribution

pretty_html_table-0.9.12-py3-none-any.whl (6.1 kB view details)

Uploaded Python 3

File details

Details for the file pretty_html_table-0.9.12.tar.gz.

File metadata

  • Download URL: pretty_html_table-0.9.12.tar.gz
  • Upload date:
  • Size: 5.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/56.0.0 requests-toolbelt/0.9.1 tqdm/4.32.2 CPython/3.7.2

File hashes

Hashes for pretty_html_table-0.9.12.tar.gz
Algorithm Hash digest
SHA256 6585a618b4c3f16db8bab6278a1c0c0990f0f3a856ada29ce7e261fe1c41c63c
MD5 14df3173f29674a37e7caae8b47f60d5
BLAKE2b-256 a47def2ff19498a6f811a47ac60657ea21020e16aeebe7e4fc4a3867650ccf28

See more details on using hashes here.

File details

Details for the file pretty_html_table-0.9.12-py3-none-any.whl.

File metadata

  • Download URL: pretty_html_table-0.9.12-py3-none-any.whl
  • Upload date:
  • Size: 6.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.22.0 setuptools/56.0.0 requests-toolbelt/0.9.1 tqdm/4.32.2 CPython/3.7.2

File hashes

Hashes for pretty_html_table-0.9.12-py3-none-any.whl
Algorithm Hash digest
SHA256 dc2be726d72320e40ee056eb3bf41ca577f2cd5661f4ebb15ae43a79ff91f922
MD5 9d54f400ecf13bcd3533696a69462ece
BLAKE2b-256 4508a5f5a10b6f5356c4a8aaf19b106966e845f0898aca56183f3630ff75ff40

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page