{"id":410319,"date":"2024-06-29T21:18:56","date_gmt":"2024-06-29T21:18:56","guid":{"rendered":"http:\/\/savepearlharbor.com\/?p=410319"},"modified":"-0001-11-30T00:00:00","modified_gmt":"-0001-11-29T21:00:00","slug":"","status":"publish","type":"post","link":"https:\/\/savepearlharbor.com\/?p=410319","title":{"rendered":"Only 39% of the functions in node_modules are unique in the default Angular project<\/span>"},"content":{"rendered":"
<\/div>\n
\n
\n
\n
\n

Only 39%<\/strong> of the functions in node_modules<\/code> are unique in the default Angular project created by ng new my-app<\/code>.<\/p>\n

<\/p>\n

I think the developers of open source solve problems in the same ways, because they study the same algorithms. Well, why be honest, they copy the popular solutions from StackOverflow also.<\/p>\n

<\/p>\n

<\/p>\n

<\/a> <\/p>\n

How to compare functions in Javascript?<\/h2>\n

<\/p>\n

If you want to compare functions in Javascript, then just convert them to string form:<\/p>\n

<\/p>\n

const a = () => 'hi'; a.toString(); \/\/ \"() => 'hi'\"<\/code><\/pre>\n
  <\/p>\n
If variable names are different<\/h3>\n
  <\/p>\n
It is necessary to bring them to the same form using the UglifyJs<\/a> library. UglifyJs<\/strong> will minimizes functions, remove unnecessary strings, and make simple calculations.<\/p>\n
  <\/p>\n
How do you extract functions from a Javascript file?<\/h2>\n
  <\/p>\n
For this I used the Esprima<\/a> library. We parse the file and walk an AST tree. I think that it would be possible only with Uglifyjs<\/strong>, but something was wrong and I was too lazy to figure it out.<\/p>\n
  <\/p>\n
Action plan<\/h2>\n
  <\/p>\n
    \n
  1. Iterate through all the * .js<\/code> files in the directory<\/li>\n
  2. Parse each file and extract functions of types ArrowFunctionExpression<\/code>,FunctionExpression<\/code> and FunctionDeclaration<\/code><\/li>\n
  3. Compress each of the functions using UglifyJs<\/code> and write it to a file whose name is the hash of the function<\/li>\n
  4. Put the id, path and hash to the separate file info.csv<\/code><\/li>\n
  5. Load the file info.csv<\/code> intoSQLite<\/code> and make all kinds of queries, because this database is not a toy<\/a>!<\/li>\n<\/ol>\n
      <\/p>\n
    Implementation details<\/h3>\n
      <\/p>\n