Skip to content

Instantly share code, notes, and snippets.

@eggplants

eggplants/gen.sh

Last active September 8, 2023 19:59
Show Gist options
  • Save eggplants/1557ea5380cf2bdf97b89a8c73e0e793 to your computer and use it in GitHub Desktop.
Save eggplants/1557ea5380cf2bdf97b89a8c73e0e793 to your computer and use it in GitHub Desktop.
Download ZIP
漢字でGO!問題リストを作る
Raw
gen.sh
#!/usr/bin/env bash
set -euo pipefail
LOG="${LOG:-}"
[[ -n "$LOG" ]] && echo Verbose LOG will not be printed.
SKIP="${SKIP:-}"
[[ -n "$SKIP" ]] && echo Processes of image conversion will be skipped.
if ! command -v awk convert mogrify jq ruby &>/dev/null
then
echo "install: awk convert mogrify jq ruby" >&2
exit 1
fi
if ! [[ -d '漢字でGO!' ]]
then
echo "Download & Extract: <https://plicy.net/GamePlay/155561>" >&2
exit 1
fi
echo "[TSV]"
< '漢字でGO!/www/data/CommonEvents.json' jq '
.[]
| select(. != null)
| .list[]
| select(.code == 122)
| select(
(.parameters[0] == .parameters[1])
and ([.parameters[1]] | inside([9,10,19,20]))
and (.parameters[2] == 0)
and (.parameters[3] == 4)
and ([.parameters[4]] | inside([0]) | not)
)
| .parameters[4]' -r |
tr -d \" |
sed -z 's/芲悧囧谿.0*.//' |
ruby -e '
puts STDIN.read
.split(?\n)
.each_slice(4)
.map{["#{_2 != ?0 * 21 ? [_1, _2] * ?、 : _1}", _3.to_s, _4.to_s].map(&:strip)}
.map{_1 * ?\t}
.filter{_1=~/\p{Hiragana}/}
' > _answer.tsv
{
grep -v 【 _answer.tsv
grep 【 _answer.tsv
} > _
mv _ _answer.tsv
find '漢字でGO!/www/img/pictures' -name 'Lv*' |
grep -E 'Lv[0-9]+_[0-9]+.rpgmvp' |
grep -v 0000 |
awk -F/ '$0=$NF' | sort -V > _problem_a.tsv
find '漢字でGO!/www/img/pictures' -name 'LvCa004_*' |
awk -F/ '$0=$NF' | sort -V > _problem_b.tsv
len_a="$(wc -l <_answer.tsv)"
len_b="$(wc -l <_problem_a.tsv)"
len_c="$(wc -l <_problem_b.tsv)"
echo "_answer.tsv lines: <${len_a}>"
echo "_problem_a.tsv: <${len_b}>"
echo "_problem_b.tsv: <${len_c}>"
if ! [[ "$len_a" == "$((len_b+len_c))" ]]
then
echo "Error: _answer.tsv lines <${len_a}> != _problem_a.tsv <${len_b}> + _problem_b.tsv <${${len_c}}> mismatch.">&2
exit 1
fi
echo "problem answer line1 line2" > kanji_de_go.tsv
paste <(cat _problem_a.tsv _problem_b.tsv) _answer.tsv | sed 's/.rpgmvp//' >> kanji_de_go.tsv
mkdir -p ./kanji_problems/{png,webp}
[[ -n "$SKIP" ]] || {
echo "[Convert RPGMVP images of problems into PNG]"
find '漢字でGO!/www/img/pictures' |
grep -E "Lv(Ca)?[0-9]+_[0-9]+.rpgmvp" |
grep -v '_0000.png' |
while read -r i
do
[[ -n "$LOG" ]] && printf '\e[K\r%s' "$i"
dest_name='./kanji_problems/png/'"$(basename "$i" | sed 's/rpgmvp/png/')"
printf '\x89\x50\x4E\x47\x0D\x0A\x1A\x0A\x00\x00\x00\x0D\x49\x48\x44\x52' > "$dest_name"
tail -c +33 "$i" >> "$dest_name"
done
[[ -n "$LOG" ]] && echo
echo "[Shrink PNG]"
find kanji_problems/png -name '*.png' | {
if [[ -n "$LOG" ]]
then
xargs -P0 -I{} bash -c 'printf "\e[K\r{}" && mogrify -quality 1 -resize 200x {}'
else
xargs -P0 -I{} bash -c 'mogrify -quality 1 -resize 200x {}'
fi
}
[[ -n "$LOG" ]] && echo
echo "[PNG -> WEBP]"
find kanji_problems/png -name '*.png' | sed -r 's_^.*/(Lv.*)\.png$_\1_' | {
if [[ -n "$LOG" ]]
then
xargs -P0 -I{} bash -c 'printf "\e[K\r{}" && convert kanji_problems/png/{}.png kanji_problems/webp/{}.webp'
else
xargs -P0 -I{} bash -c 'convert kanji_problems/png/{}.png kanji_problems/webp/{}.webp'
fi
}
[[ -n "$LOG" ]] && echo
}
echo "[HTML:problem lists]"
levels=()
while read -r i
do
levels+=("$i")
done < <(
find kanji_problems/png -type f -name '*' |
awk -F'[_/]' '$0=$(NF-1)' |
sort -V | uniq
)
levels_len="${#levels[@]}"
for (( idx=0; idx<levels_len; idx++ ))
do
level="${levels[idx]}"
f="kanji_de_go_list_${level}.html"
count="$(find kanji_problems/png -name "$level"'_*.png' | wc -l)"
if [[ "$idx" != 0 ]]
then
prv="<span><a href='kanji_de_go_list_${levels[idx-1]}.html'>${levels[idx-1]}</a></span>"
else
prv=""
fi
if [[ "$((idx+1))" != "$levels_len" ]]
then
nxt="<span><a href='kanji_de_go_list_${levels[idx+1]}.html'>${levels[idx+1]}</a></span>"
else
nxt=""
fi
echo "${level} -> ${f}, ${count} problem(s)"
{ sed 's/^ *//' << EOS;} > "$f"
<!DOCTYPE html>
<html lang="ja" dir="ltr">
<head>
<meta charset="UTF-8">
<title>${f//.*/}</title>
<meta name="viewport" content="width=device-width,initial-scale=1">
</head>
<body>
<h1>${level} -- 全${count}問</h1>
＜＜${prv:-}|<a href='index.html'>●</a>|${nxt:-}＞＞
<table border=1>
<thead>
<tr>
<th>ID</th>
<th>問題</th>
<th>答え</th>
<th>解説</th>
</tr>
</thead>
<tbody>
EOS
grep "$level" kanji_de_go.tsv | while IFS=$'\t' read -r a b c d
do
cat<<EOS
<tr>
<td>${a}</td>
<td><a href='kanji_problems/png/${a}.png'><img alt='${a}' width=200 height=57 loading='lazy' src='kanji_problems/webp/${a}.webp'></a></td>
<td>${b}</td>
<td>${c}${d}</td>
</tr>
EOS
done | sed 's/^ *//' >> "$f"
sed 's/^ *//' <<'EOS'>> "$f"
</tbody>
</table>
</body>
</html>
EOS
done
echo "[HTML:index]"
cat << EOS > index.html
<!DOCTYPE html>
<html lang="ja" dir="ltr">
<head>
<meta charset="UTF-8">
<title>kanji_de_go_list</title>
<meta name="viewport" content="width=device-width,initial-scale=1">
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>
</head>
<body>
<h1><a href="https://plicy.net/GamePlay/155561">漢字でGO!</a>　非公式レベル別問題リスト</h1>
<h2>概要</h2>
<p>このサイトは漢字でGO!の問題と答えをレベル別にリストアップし、対策するための非公式サイトです。</p>
<p>問題画像・答え・解説テキストの権利はすべて製作者さまに帰属します。</p>
<p>最終更新：$(LANG=C date +%Y-%m-%d)</p>
<h2>リスト一覧</h2>
<ol>
<li><a href='kanji_de_go_list_Lv00.html'>Lv00 -- サンプル</a></li>
<li><a href='kanji_de_go_list_Lv01.html'>Lv01 -- 10〜5</a></li>
<li><a href='kanji_de_go_list_Lv02.html'>Lv02 -- 6〜準2</a></li>
<li><a href='kanji_de_go_list_Lv03.html'>Lv03 -- 4〜準1</a></li>
<li><a href='kanji_de_go_list_Lv04.html'>Lv04 -- 2〜1</a></li>
<li><a href='kanji_de_go_list_Lv05.html'>Lv05 -- 準1〜対象外</a></li>
<li><a href='kanji_de_go_list_Lv06.html'>Lv06 -- 対象外</a></li>
<li><a href='kanji_de_go_list_Lv07.html'>Lv07 -- 対象外</a></li>
<li><a href='kanji_de_go_list_LvCa004.html'>LvCa004 -- JLPT N5</a></li>
</ol>
<h2>レベルと漢検の対応</h2>
<blockquote class="twitter-tweet">
<p lang="ja" dir="ltr">
【難易度の範囲について】<br>
（以下は紹介動画で使用した画像）出題される問題は大まかに漢検を基準としております。<br>
但し、配当漢字でも読みが簡単と判断したものはひとつ下のレベルに下げていたり、難しい問題はレベルを上げている場合がございます。
<a href="https://t.co/K2ds85cO41">pic.twitter.com/K2ds85cO41</a>
</p>
&mdash; 『漢字でGO!』開発 (@KanzideGo)
<a href="https://twitter.com/KanzideGo/status/1691044777972944896?ref_src=twsrc%5Etfw">August 14, 2023</a>
</blockquote>
</body>
</html>
EOS
if ! :
then
echo "[OCR]"
if ! command -v python &>/dev/null
then
echo "install: python" >&2
exit 1
fi
if { python -m manga_ocr -h |& grep -q FLAGS;}
then
echo "run: pip install manga_ocr">&2
exit 1
fi
python -c'
from pathlib import Path
import sys
import manga_ocr
m = manga_ocr.MangaOcr()
for i in sorted(Path("kanji_problems").glob("*.png")):
print(i, file=sys.stderr)
print(m(i))
' > ocr_problem
fi
mkdir -p docs/
if [[ -d docs/kanji_problems ]]
then
rm -rf docs/kanji_problems
fi
mv kanji_problems index.html kanji_de_go_list_Lv*.html docs
echo "[DONE!]"
@eggplants
Copy link
Author

eggplants commented Sep 7, 2023 •
edited

Requirements

  • awk
  • imagemagick
  • jq
  • ruby

Generate lists

  1. mkdir kanji && cd kanji
  2. Download & Extract 漢字でGO!.zip from https://plicy.net/GamePlay/155561
  3. Run ./gen.sh

Download & Extract 漢字でGO!.zip

file_id="$(
  curl -s "$(
    < game_page grep -oEm1 'https://drive.google.com/drive/folders/[^<]+'
  )" |
  grep -oE '<div data-id="'".*"'" data-target="doc" draggable="true"' |
  awk -F'"' '$0=$2'
)"
[[ -z "$file_id" ]] && exit 1

git clone --depth 1 "https://gist.github.com/eggplants/f638ba6a6208e4e37f49ccae94cc948e" _
mv _/google-drive-downloader.sh .
rm -rf _
chmod +x google-drive-downloader.sh
./google-drive-downloader.sh "$file_id" '漢字でGO!.zip'
unzip -Ocp932 '漢字でGO!.zip'
rm '漢字でGO!.zip' google-drive-downloader.sh game_page

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment