aboutsummaryrefslogtreecommitdiffstats
path: root/README.org
blob: 1ed72f587a20740215d88a822ef476f8f43540dc (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
#+title: Octans
#+subtitle: Octans is a program to solve Algot's Wordplay (Wordsearch) puzzles
#+export_file_name: index
#+setupfile: ~/.emacs.d/org-templates/projects.org

| Website         | https://andinus.nand.sh/octans             |
| Source          | https://git.tilde.institute/andinus/octans |
| GitHub (mirror) | https://github.com/andinus/octans          |

* Demo

This was recorded with ~asciinema~.

[[https://asciinema.org/a/385500][https://asciinema.org/a/385500.png]]

+ Octans v0.1.0 - finds 10 solutions to a puzzle:
  https://asciinema.org/a/385598
+ Octans v0.1.0: https://asciinema.org/a/385500

+ Octans 2021-01-14: https://asciinema.org/a/384464
+ alt-link (download): https://andinus.nand.sh/static/octans

* Installation

Octans is released to CPAN, you can get it from there or install it from
source. In any case, ~zef~ is required to install the distribution.

You can run Octans without ~zef~. Just run ~raku -Ilib bin/octans~ from
within the source directory.

- *Note*: Octans requires ~WWW~ module. You can however chose to not install
  it if you wish to run it locally. Checkout the ~without-www~ branch.

** Release

1. Run ~zef install octans~.

Octans should be installed, try running ~octans --version~ to confirm.

** From Source

You can either download the release archive generated by cgit/GitHub or
clone the project if you have ~git~ installed.

*** Without ~git~

1. Download the release:
   - https://git.tilde.institute/andinus/octans
   - https://github.com/andinus/octans/releases
2. Extract the file.
3. Run ~zef install .~ in source directory.

*** With ~git~

All commits will be signed by my [[https://andinus.nand.sh/static/D9AE4AEEE1F1B3598E81D9DFB67D55D482A799FD.asc][PGP Key]].

#+begin_src sh
# Clone the project.
git clone https://git.tilde.institute/andinus/octans
cd octans

# Install octans.
zef install .
#+end_src

* Documentation

** Implementation

Initially it went over the list of words & checked if they exist in the
grid. This was very slow.

Currently it walks the grid & checks if the current string exist in the
dictionary. This is faster for these reasons:

- The dictionary is sorted, we perform binary range search on the
  dictionary to return the list of all words that start with specific
  string.
- Starting positions are limited.

If the dictionary wasn't sorted then this probably would've been slower
than previous implementation.

The ~neighbors~ subroutine (=lib/Octans/Neighbors.rakumod=) was taken from
my AoC (Advent of Code) 2020's day-11 solution.

** Options

*** dict

Octans's default dictionary file is =/usr/share/dict/words=, use ~--dict~
flag to change the dictionary. The words in dictionary must be seperated
by a newline (=\n=) & sorted alphabetically.

*** length

Minimum word length to look for. Default is 7.

*** path

The path to be passed must be a *readable file* or an *url* in either
format:

- Link when you view it from your local instance:
  https://tilde.zone/web/statuses/105531207939242077

- Link from Algot's profile:
  https://mastodon.art/@Algot/105333136907848390

*** verbose

This will increase verbosity.

* News

** v0.2.0 - 2021-03-04

+ Removed ~sample~ option

  ~--sample~ will not solve the sample puzzle.

+ Removed shorthand for verbose option

  ~-v~ won't work in ~v0.2.0~.

+ Change representation of visited squares

  When ~--verbose~ was passed, it would print the solved puzzle with
  visited squares replaced with fancy characters. Now it marks them by
  adding:

  - ~*~ to visited + gray squares (start positions)
  - ~/~ to visited squares

+ Removed time taken

  Time taken won't be printed anymore.

  #+begin_quote
  It was misleading because the time printed was not the time taken to
  find that specific word, it was the time taken to find all the words
  upto that word. It would reset for each starting position.
  #+end_quote

** v0.1.4 - 2021-02-19

+ Fixed the puzzle parsing regex

  The older regex fails to parse this puzzle:
  https://mastodon.art/@Algot/105690195742318751.

** v0.1.3 - 2021-01-24

+ Added an option to specify minimum word length.

** v0.1.2 - 2021-01-20

+ Input puzzle can now be of any size & not restricted to 4x4 grid.

** v0.1.1 - 2021-01-20

+ Read puzzle from a file.

** v0.1.0 - 2021-01-19

+ Improved performance by using a better algorithm to find words in the
  grid.