Thinky

Run

Purpose

Thinky is a query tool for analyzing the source code of 188 games in the js13kgames 2024 competition

ever wonder who uses speechsynth?
https://bacionejs.github.io/stuff/thinky.html?word=speechSynthesis

or who doesn't use a while/for loop?
https://bacionejs.github.io/stuff/thinky.html?word=-while,-for

audio?
https://bacionejs.github.io/stuff/thinky.html?word=AudioContext

zzfx?
https://bacionejs.github.io/stuff/thinky.html?word=zzfx

find your buddies and other strange oddities at Thinky
Brought to you by Stuff

Limitations

• The data only contains words used by more than one game. Thinky is designed to compare 188 games and uses consensus to reduce 20,000 words to 5,000 words. If you want, I can change it.
• Only evaluates html/js. No json, etc.

Misc

Here are some scripts to analyze your own code.

Sort all words by length

cat minified.js|sed 's/base64[^=]*=//g'|tr -cs '[:alnum:]_' '\n'|awk '{print length,$0}'|sort -u|sort -n|cut -d' ' -f2

Developer Notes

THESE ARE NOT INSTRUCTIONS

I posted these notes so other people can

• see the specific logic used
• make suggestions for doing it right

These scripts were run on Android Termux

• The unroll.js (un-roadroller) script is a homegrown hack. If you know the right way, like a tool, or better unroll logic, please let me know.
• The extract.sh script extracts the data to be used by thinky.js

git.sh

Manually paste into page.txt, the text of the https://js13kgames.com/2024/games website
Make a folder called games
Run ./git.sh

#!/bin/bash
#get game names
cat page.txt | sed '1,4d' | head -n -20 | awk 'NR % 3 == 1' | \
#fix game names
tr '[:upper:]' '[:lower:]' | \
sed "s/['.]//g" | \
sed 's/[^a-zA-Z0-9]/-/g' | sed 's/-\{2,\}/-/g' | sed 's/^-//;s/-$//' | \
#download submitted zips
while read folder; do
  mkdir games/$folder
  if curl -s -o /dev/null -w "%{http_code}" -L https://github.com/js13kGames/games/raw/main/games/$folder/.src/g.zip | grep "200"; then
    curl -L -o                     $folder.zip https://github.com/js13kGames/games/raw/main/games/$folder/.src/g.zip
    mv $folder.zip games/$folder/g.zip
  else
    echo "error: "$folder
  fi
done
#unzip
find games -name "*.zip" -exec sh -c 'unzip -q "{}" -d "$(dirname "{}")" || echo "$PWD/{}"' +
#delete zip
find games -name "*.zip" -exec rm -f {} +
#fix folder spaces
find games -name '* *' -exec bash -c 'mv -v "$0" "`echo $0 | tr " " "-"`"' {} +

unroll.js

Run node unroll.js games

let fs = require('fs');
let path = require('path');
let vm = require('vm');

function extract(s) {
if      (new RegExp(/Function\s*\(\s*"\[M/gs).test(s)){
  let regex=/Function\s*\(\s*"\[M/gs;
  let match=regex.exec(s);
  let start=match.index;
  let p=1;
  let i=start+match[0].length;
  while(i<s.length && p>0){if(s[i]=='(')p++;if(s[i]==')')p--;i++;}
  while(i<s.length && s[i]!=='('){i++;}
  let args=1;
  i++;
  while(i<s.length && args>0){if(s[i]=='(')args++;if(s[i]==')')args--;i++;}
  s=s.substring(start,i);
  s=s.replace(/document.*"/,'return String.fromCharCode(...c)"');
  return s;
}else if(new RegExp(/<script>\s*M\s*=/g)     .test(s)){
  s=s.replace(new RegExp(/document.*</g),'String.fromCharCode(...n)<');
  s=s.replace(new RegExp(/eval\(r\)/g),  'eval(JSON.stringify(r))');
  return s.replace(/[\s\S]*<script\b[^>]*>([\s\S]*?)<\/script>[\s\S]*/,'$1');
}else{
  return undefined;
}
}

function main(file) {
fs.readFile(file, 'utf8', (err, data) => {
if (err) { console.error("Error reading file:", err); return; }
let script,unrolled,s;
try {
  s = extract(data);
  if (s) {
    script = new vm.Script(s);
    unrolled = script.runInContext(vm.createContext({}));
    overwrite(file,unrolled);
    console.log(`Successfully updated: ${file}`);
  }
} catch (error) {
  console.error("Error:", `${file} ${error.message}`);
}
});
}

function overwrite(file,unrolled){
fs.writeFile(file, unrolled, (writeErr) => {
  if (writeErr) {
    console.error("Error writing file:", writeErr);
  }
});
}

function processDirectory(dir) {
fs.readdir(dir, { withFileTypes: true }, (err, entries) => {
if (err) { console.error("Error reading directory:", err); return; }
entries.forEach(entry => {
  let fullPath = path.join(dir, entry.name);
  if (entry.isDirectory()) { // Recursively process subdirectory
    processDirectory(fullPath);
  } else if (entry.isFile() && (entry.name.endsWith('.js') || entry.name.endsWith('.html'))) {
    main(fullPath);
  }
});
});
}

let startDir = path.resolve(process.argv[2]);
processDirectory(startDir);

extract.sh

Run ./extract.sh

#!/bin/bash
# Canonicalize zzfx
find games -type f \( -name '*.html' -o -name '*.js' \) -exec sed -i -r 's/module:\[\[\[/zzfx/gi;s/zzfx\w*/zzfx/gi' {} +
# Find all .js and .html files
# Remove base64
# Split the remaining content into tokens based on non-alphanumeric except _
find games -type f \( -name "*.js" -o -name "*.html" \) -exec \
  awk --posix '{
      gsub(/base64[^=]+=/, "");
      n = split($0, tokens, /[^[:alnum:]_]/);
      for (i=1; i<=n; i++) if (tokens[i] != "") print FILENAME, tokens[i];
  }' {} + | \
# Extract the game name and token
awk -F'[/ ]' '{print $2, $NF}' | \
# Remove tokens starting with a digit
awk '$2 !~ /^[0-9]/' | \
# Remove tokens < 3 characters
awk 'length($2) > 2' | \
# Sort by token
sort -k2,2 | \
# Remove duplicates
uniq | \
# Remove tokens used by < 2 games
awk '{count[$2]++; games[$2] = (games[$2] ? games[$2] ORS : "") $1 " " $2} END {for (word in count) if (count[word] > 1) print games[word]}' | \
cat > data.txt
# Create foreignkeys for tokens
cat data.txt | awk '{if (!($2 in b)) { b[$2] = i++; } a[$1] = a[$1] ? a[$1] "," b[$2] : $1 "," b[$2]; } END { for (k in a) print a[k] }' > games.txt
# Create lookup table for tokens
cat data.txt | awk '{print $2}' | uniq | tr '\n' ',' | sed 's/,$//' > words.txt