benhoyt

import collections, random, sys, textwrap

# Build possibles table indexed by pair of prefix words (w1, w2)
w1 = w2 = ''
possibles = collections.defaultdict(list)
for line in sys.stdin:
    for word in line.split():
        possibles[w1, w2].append(word)
        w1, w2 = w2, word

benhoyt

# $ python3 count_features.py writings/go-python-features.md 
# go1.25  0 23
# go1.24  1 29
# go1.23 15 27
# go1.22  6 41
# go1.21  8 50
# go1.20  4 33
# go1.19  0 14
# go1.18 15 31
# go1.17  3 15

benhoyt

package main

import (
	"fmt"
	"os"
	"path/filepath"
)

func main() {
	if len(os.Args) != 2 {

benhoyt

"""Quick performance tests comparing "match" with "if...elif".

See:
https://benhoyt.com/writings/python-pattern-matching/
https://news.ycombinator.com/item?id=28601616

# Enum switch with match:
$ python3.10 -m timeit -s 'import mt' -c 'mt.enum_match(mt.FileType.FILE)'
1000000 loops, best of 5: 356 nsec per loop
$ python3.10 -m timeit -s 'import mt' -c 'mt.enum_match(mt.FileType.SYMLINK)'

benhoyt

"""Print most frequent N-grams in given file.

Usage: python ngrams.py filename

Problem description: Build a tool which receives a corpus of text,
analyses it and reports the top 10 most frequent bigrams, trigrams,
four-grams (i.e. most frequently occurring two, three and four word
consecutive combinations).

NOTES

benhoyt

"""An atomic, thread-safe incrementing counter."""

import threading


class AtomicCounter:
    """An atomic, thread-safe incrementing counter.

    >>> counter = AtomicCounter()
    >>> counter.increment()

benhoyt

import csv
import os
import subprocess
import time


print('| Go version | binary size (MB) | countwords (s) | sumloop (s) |')
print('| ---------- | ---------------- | -------------- | ----------- |')

benhoyt

| Go version | binary size (MB) | countwords (s) | sumloop (s) |
| ---------- | ---------------- | -------------- | ----------- |
| 1.0        |             3.27 |           8.20 |        9.47 |
| 1.1        |             3.95 |           7.57 |        9.54 |
| 1.2        |             5.50 |          25.51 |        9.36 |
| 1.3        |             4.27 |           2.85 |        1.99 |
| 1.4        |             4.32 |           3.03 |        2.13 |
| 1.5        |             4.73 |           2.16 |        1.16 |
| 1.6        |             4.70 |           2.15 |        1.18 |
| 1.7        |             3.94 |           1.86 |        0.63 |

benhoyt

"""Speed up os.walk() significantly by using file attributes that
FindFirst/Next give us instead of doing an extra stat(). Can also do the same
thing with opendir/readdir on Linux.

This is doubly useful when the user (caller of os.walk) is doing *another*
stat() to get say the file sizes.

On my tests (Windows 64-bit) our walk() is about 5x as fast as os.walk() for
large directory trees, and 9x as fast if you're doing the file size thing.
Note that these timings are "once it's in the cache", not first-time timings.

benhoyt

"""Efficient sliding-window sorting of time-series data in CSV file.

Demo for http://stackoverflow.com/a/42398981/68707

Tested on Python 3.5.
"""

import collections
import csv
import datetime