javascript promise, is it needed here?

Question

I'm slightly confused about how to use promises. I've read a few things on promises mainly because it seems like I have to use them. I'm working on a small application which is supposed to search through some pdfs using pdfjs, and they use promises. I knocked up something in nodejs, by looking at various examples on the net but I run into a problem.

Let's look at the code first:

require('pdfjs-dist');
var fs = require('fs');

//var searchTerm = "course";
var searchTerm = "designee";
//var searchTerm = "document";
var wordCounter = 0;
var searchResultJSON = [];
//var data = new Uint8Array(fs.readFileSync('testPdf.pdf'));
//var data = new Uint8Array(fs.readFileSync('advanced-javascript.pdf'));
var data = new Uint8Array(fs.readFileSync('iss4.pdf'));
PDFJS.getDocument(data).then(function (pdfDocument) {
  console.log('Number of pages: ' + pdfDocument.numPages );
  //var div = document.getElementById('viewer');
  for(var i = 1; i<=pdfDocument.numPages; i++ ){//loops thru pages
      console.log("i is " + (i));
      pdfDocument.getPage((i)).then(function(page){//get page(i), 
         // console.log("page is " + (i));
          //console.log("inside getPage()");
          page.getTextContent().then( function(textContent){//get content of pdf
            //console.log("inside getTextContent()");  
            //if( null != textContent.items ){
                var page_text = "";
                var last_block = null;
                var lineWithResult = "";

                for( var k = 0; k < textContent.items.length; k++ ){
                    var block = textContent.items[k];
                    //console.log("word " + textContent.items.length + " k is " + k );
                    /* if( last_block != null && last_block.str[last_block.str.length-1] != ' '){
                        if( block.x < last_block.x )
                            page_text += "\r\n"; 
                        else if ( last_block.y != block.y && ( last_block.str.match(/^(\s?[a-zA-Z])$|^(.+\s[a-zA-Z])$/) == null ))
                            page_text += ' ';
                    } */

                    page_text += block.str;

                    last_block = block;
                    lineWithResult = searchPdf(block.str);
                    if(lineWithResult != null){
                        console.log(lineWithResult + " wordCounter is " + wordCounter);

                    }

                }//end of for(var k...)
                    //console.log(" page_text " + page_text);
                    //console.log(searchResultJSON);

            //}
          });//end of textContent.items

      });//end of getPage

  }//end of loop      
});
function searchPdf(toSearch){//searching pdf for searchTerm
    var result = toSearch.toLowerCase().indexOf(searchTerm);
    if(result >=0){//if match is found
        wordCounter++;
        //console.log("toSearch " + toSearch + " result is " + result + " wordCounter " + wordCounter);
        constructResult(toSearch, result);//build the result object
        return toSearch;
    }
    else{//if match not found
        return null;
    }

}
function constructResult(toSearch, result){//construct array of objects containing: search term, search result and index of search term
    searchResultJSON.push({
        "TextLine":toSearch,
        "SearchTerm":searchTerm,
        "Result": result,               
    });     
} 

The purpose of this code is to:

• loop through the pdf's pages

• loop through the content

• get the pdf text in a variable line by line

• search the pdf content with a keyword

• if the keyword finds a match, print the match

• get the matches in a javascript object

So, it all works OK but you'll notice that from inside the second for loop (where I get the text of the pdf that is) I call a function, searchPdf() which basically performs the search and from within that function I call another function constructResult(...) which is supposed to create the javascript object with the results.

I have some problems printing this object though, because if I print it outside the scope of the for loop, it is empty because the print call (in my case the console.log) executes before the loop has actually copied and analysed (read process and found a match) the text. So, promises see the way to resolve the problem. Thing is, I'm not sure how to code this in such a way that I can chain the promises and print my object after everything has executed. Any idea?

EDIT: so to clarify, what I need in sequence is this: 1)loop through pdf (I will have to amend the code to loop through a collection of pdfs at some point soon) 2)get each line of text 3)check that there is a match 4)if so, copy the line of text in the javascript object 5)print the javascript object

Answer 1

Try something like this:

function search(doc, s) {
    var allPages = [],
        i;

    for (var i = 1; i <= doc.numPages; i++) {
        allPages.push(doc.getPage(i));
    }

    // Promise.all returns a promise that resolves once 
    // each promise inside allPages has been resolved
    return Promise.all(allPages)
    // pages now contains an array of pages, loop over them
    // using map, return the promise to get the content for each page
    // return it through Promise.all so we can run the next phase
    // once the text is resolved for each page
    .then(pages => Promise.all(pages.map(p => p.getTextContent())))
    // We now have an array of contents for each page, filter based
    // on the passed string
    .then(content => content.filter(c => c.indexOf(s) > -1));
}

// This is a mock of the pdf API used in your question
var pdfDocument = {
    numPages: 3,
    getPage: function(i) {
        return Promise.resolve({
            getTextContent: function() {
                return Promise.resolve('Page ' + i);
            }
        });
    }
}