book-summary-comparator

Top 1000 Books Summary Comparison

actual dataset is contains 518 books.

Install & Use

npm i -g book-summary-comparator
book-summary-comparator

How

Firstly find which document is which books.

Steps

• Step 1: Gather which ref we have.

ls |grep "" > ref.txt # For unix 
ls | grep "" > ref.txt # For linux 

• Step 2: Gather which book list we have.

// open https://www.edebiyatogretmeni.org/etiket/1000-roman-ozeti-indir/
// Execute this JS in browser console:
// Gather content.
const data = document.querySelectorAll('.entry-content p')[1].innerHTML.split('<br>').map(row => row.toLowerCase()).join('\n')
// Generate hidden input for copy to clipboard.
const dummyInput = document.createElement('input')
dummyInput.value = data;
 
/* Select the text field */
dummyInput.select();
dummyInput.setSelectionRange(0, 99999);
 
/* Copy the text inside the text field */
document.execCommand("copy"); // Ta-da! Copied your clipboard.

Paste into a ref.txt

echo `pbpaste` > list.txt

You have both list.txt and ref.txt

• Step 2: Match the refs and list.

node match.js; # Match the books and refs. Left pad 00001 to 1.doc 
node extract.js # Extract data accordingly 1.doc to 1.txt 

• Step 3: Generate db.json

Just loop over database then fs.writeFileSync..

• Step 4: Compare books with other books. Big O (n^2)

Thankfully Node.js and JS, any loop is parallel executed by default. Does not needed any parallelisation process.

node magic.js # Comparison algorithm is: Dice's Coefficient. 
# Generates two indexed JSON. case_1, case_2 

Other staffs are for build CLI :)

Cheers, Cagatay Cali.