ぱわふる

my-mbp% ipython
Python 3.4.2 (default, Jan 12 2015, 11:46:28) 
Type "copyright", "credits" or "license" for more information.

IPython 2.3.1 -- An enhanced Interactive Python.
?         -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help      -> Python's own help system.
object?   -> Details about 'object', use 'object??' for extra details.

In [1]: from homura import download

In [2]: download('http://aozora-word.hahasoha.net/utf8/newnew.csv.gz')
   100%    615.8 MiB      11.1 MiB/s            0:00:00 ETA

#!/usr/bin/env python

import madoka

sketch = madoka.Sketch()

with open('newnew.csv', 'r') as f:
    for line in f:
        morpheme = line.split(',')[3]
        sketch[morpheme] += 1
        print(morpheme)

print(sketch['人'])
print(sketch['面倒'])
print(sketch['蕩尽'])

1418808
57427
1910

#!/usr/bin/env python

from collections import Counter

cnt_of = Counter()

with open('newnew.csv', 'r') as f:
    for line in f:
        morpheme = line.split(',')[3]
        cnt_of[morpheme] += 1
        print(morpheme)

print(cnt_of['人'])
print(cnt_of['面倒'])
print(cnt_of['蕩尽'])

217837
1738
23

#!/usr/bin/env python

from benchmarker import Benchmarker

with Benchmarker(1000*100, width=20) as bench:
    code_point = range(0x3040, 0x30A0)

    @bench('1')
    def _(bm):
        for i in bm:
            moji_list = []
            for cp in code_point: moji_list.append( chr(cp) )

    @bench('2')
    def _bm(bm):
        for i in bm:
            moji_list = [ chr(cp) for cp in code_point ]

    @bench('3')
    def _bm(bm):
        for i in bm:
            moji_list = list(chr(cp) for cp in code_point)

## benchmarker:         release 4.0.1 (for python)
## python version:      3.4.2
## python compiler:     GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.56)
## python platform:     Darwin-14.0.0-x86_64-i386-64bit
## python executable:   /Users/username/.anyenv/envs/pyenv/versions/3.4.2/bin/python
## cpu model:           Intel(R) Core(TM) i5-4258U CPU @ 2.40GHz 
## parameters:          loop=100000, cycle=1, extra=0

##                        real    (total    = user    + sys)
1                       3.1367    3.0500    3.0200    0.0300
2                       2.4863    2.4300    2.4200    0.0100
3                       2.8027    2.7300    2.7100    0.0200

## Ranking                real
2                       2.4863  (100.0) ********************
3                       2.8027  ( 88.7) ******************
1                       3.1367  ( 79.3) ****************

## Matrix                 real    [01]    [02]    [03]
[01] 2                  2.4863   100.0   112.7   126.2
[02] 3                  2.8027    88.7   100.0   111.9
[03] 1                  3.1367    79.3    89.3   100.0

#!/usr/bin/env python

from collections import OrderedDict

code_point_list = range(0x3040, 0x30A0)
chara_list      = [ chr(code_point) for code_point in code_point_list ]

moji_dict = OrderedDict(); # 登録順序を記憶
moji_dict.update([ tuple([ code_point, chara ]) for code_point, chara in zip(code_point_list, chara_list) ])
moji_dict.update({ 0x303F: chr(0x3020) }) # 順序が保持されているか確認用

for key, value in moji_dict.items():
    print('U+{0:X} {1}'.format(key, value))

U+3040 ぀
U+3041 ぁ
U+3042 あ
U+3043 ぃ
U+3044 い
U+3045 ぅ
U+3046 う
U+3047 ぇ
U+3048 え
U+3049 ぉ
U+304A お
U+304B か
U+304C が
U+304D き
U+304E ぎ
U+304F く
U+3050 ぐ
U+3051 け
U+3052 げ
U+3053 こ
U+3054 ご
U+3055 さ
U+3056 ざ
U+3057 し
U+3058 じ
U+3059 す
U+305A ず
U+305B せ
U+305C ぜ
U+305D そ
U+305E ぞ
U+305F た
U+3060 だ
U+3061 ち
U+3062 ぢ
U+3063 っ
U+3064 つ
U+3065 づ
U+3066 て
U+3067 で
U+3068 と
U+3069 ど
U+306A な
U+306B に
U+306C ぬ
U+306D ね
U+306E の
U+306F は
U+3070 ば
U+3071 ぱ
U+3072 ひ
U+3073 び
U+3074 ぴ
U+3075 ふ
U+3076 ぶ
U+3077 ぷ
U+3078 へ
U+3079 べ
U+307A ぺ
U+307B ほ
U+307C ぼ
U+307D ぽ
U+307E ま
U+307F み
U+3080 む
U+3081 め
U+3082 も
U+3083 ゃ
U+3084 や
U+3085 ゅ
U+3086 ゆ
U+3087 ょ
U+3088 よ
U+3089 ら
U+308A り
U+308B る
U+308C れ
U+308D ろ
U+308E ゎ
U+308F わ
U+3090 ゐ
U+3091 ゑ
U+3092 を
U+3093 ん
U+3094 ゔ
U+3095 ゕ
U+3096 ゖ
U+3097 ゗
U+3098 ゘
U+3099 ゙
U+309A ゚
U+309B ゛
U+309C ゜
U+309D ゝ
U+309E ゞ
U+309F ゟ
U+303F 〠

#!/usr/bin/env python
for cp in range(0x3040, 0x30A0): print(chr(cp), end=' ')

぀ ぁ あ ぃ い ぅ う ぇ え ぉ お か が き ぎ く ぐ け げ こ ご さ ざ し じ す ず せ ぜ そ ぞ た だ ち ぢ っ つ づ て で と ど な に ぬ ね の は ば ぱ ひ び ぴ ふ ぶ ぷ へ べ ぺ ほ ぼ ぽ ま み む め も ゃ や ゅ ゆ ょ よ ら り る れ ろ ゎ わ ゐ ゑ を ん ゔ ゕ ゖ ゗ ゘ ゙ ゚ ゛ ゜ ゝ ゞ ゟ

#!/usr/bin/env perl
use open qw/:utf8 :std/;
print chr . ' ' for 0x3040 .. 0x309F;

#!/usr/bin/env python
 [ print(chr(cp), end=' ') for cp in range(0x3040, 0x30A0) ]

#!/usr/bin/env python
print(' '.join(chr(cp) for cp in range(0x3040, 0x30A0)), end='')

#!/usr/bin/env perl

use strict;
use warnings;
use utf8;
use Graphviz::DSL;
use open IO => qw/:utf8 :std/;

my $graph = graph {
    name 'Miscellaneous Symbols';

    global rankdir => 'LR';

    my $cnt = 0;

    for my $dec ( hex('2600') .. hex('2609') )
    {
        $cnt++;
        my $char = chr $dec;
        node "node${cnt}", shape => 'record', label => "{<p>$char|<n>}";
    }

    while ($cnt > 1)
    {
        edge [ "node" . ($cnt - 1) . ":n" => "node" . $cnt . ":p" ];
        $cnt--;
    }
};

$graph->save(path => 'symbols', type => 'png', encoding => 'utf-8');