parsiad/lazy-table

A python-tabulate wrapper for producing tables from generators

A python-tabulate wrapper for producing tables from generators.

Motivation

lazy_table is useful when (i) each row of your table is generated by a possibly expensive computation and (ii) you want to print rows to the screen as soon as they are available.

For example, the rows in the table below correspond to numerically solving an ordinary differential equation with progressively more steps. See examples/euler_vdp.py for the code to generate this table.

Here is the same example with a progress bar.

Installation and usage

Install lazy_table via pip:

$ pip install lazy_table

In your Python code, add

import lazy_table as lt

Now, generating your own lazy table is as simple as calling lt.stream on a generator that yields one or more lists, where each list is a row of your table (see the example below).

Example

As a toy example, consider printing the first 10 Fibonacci numbers in a table. Since this is a relatively cheap operation, we will pretend it is expensive by adding an artificial sleep(1) call beteween Fibonacci numbers.

import time
import lazy_table as lt

def fib_table(n):
     x0, x1 = 0, 1
     yield [0, x0]
     yield [1, x1]
     for i in range(2, n + 1):
         time.sleep(1)  # Simulate work
         x0, x1 = x1, x0 + x1
         yield [i, x1]

lt.stream(fib_table(10), headers=['N', 'F_N'])

Your final table should look like this:

  N    F_N
---  -----
  0      0
  1      1
  2      1
  3      2
  4      3
  5      5
  6      8
  7     13
  8     21
  9     34
 10     55

You can also specify an "artist" that determines how the table is rendered. For example, the ConsoleWithProgress artist displays a progress bar alongside the table indicating the percentage of rows completed:

n = 10
artist = lt.artists.ConsoleWithProgress()
lt.stream(fib_table(n), headers=['N', 'F_N'], n_rows=n + 1, artist=artist)