computeBase.ParseDataset.java Source code

Introduction

Here is the source code for computeBase.ParseDataset.java

Source

/*
 * To change this license header, choose License Headers in Project Properties.
 * To change this template file, choose Tools | Templates
 * and open the template in the editor.
 */
package computeBase;

import static computeBase.Main.path1;
import static computeBase.Main.path2;
import java.io.*;
import java.util.ArrayList;
import org.jdom2.*;
import org.jdom2.input.*;
import java.util.List;
import java.util.Iterator;
import org.openscience.cdk.exception.CDKException;
import org.openscience.cdk.interfaces.IAtomContainer;

/**
 *
 * @author oxis
 */
public enum ParseDataset {

    INSTANCE;

    private final org.jdom2.Document document;
    private final Element drugs;

    private final List<IAtomContainer> ListDrug = new ArrayList<>();

    private ParseDataset() {
        //On cre une instance de SAXBuilder
        /*File file;
        if(new File(path1 + "trainingset_uncomputed.xml").isFile()) {
        file = new File(path1 + "trainingset_uncomputed.xml");
        } else {
        file = new File(path2 + "trainingset_uncomputed.xml");
        }*/

        File file;
        if (new File(path1 + "testingset_uncomputed.xml").isFile()) {
            file = new File(path1 + "testingset_uncomputed.xml");
        } else {
            file = new File(path2 + "testingset_uncomputed.xml");
        }

        SAXBuilder sxb = new SAXBuilder();
        try {
            //On cre un nouveau document JDOM avec en argument le fichier XML
            //Le parsing est termin ;)
            document = sxb.build(file);
        } catch (IOException | JDOMException e) {
            throw new RuntimeException(e);
        }

        //On initialise un nouvel lment racine avec l'lment racine du document.
        drugs = document.getRootElement();
    }

    public List<IAtomContainer> getStructs() {
        //On cre une List contenant tous les noeuds "etudiant" de l'Element racine

        List listDrugs = drugs.getChildren("drug");

        float logp = 0;
        String smiles = "";
        String name;
        String cas;

        DrugStruct tmpDrug;
        int good = 0;
        int bad = 0;
        //On cre un Iterator sur notre liste
        Iterator itDrug = listDrugs.iterator();
        while (itDrug.hasNext()) {
            Element drug = (Element) itDrug.next();

            name = drug.getChild("name").getText();
            cas = drug.getChild("cas-number").getText();

            List listExpProp = drug.getChild("experimental-properties").getChildren("property");

            Iterator itExpProp = listExpProp.iterator();
            while (itExpProp.hasNext()) {
                Element expProp = (Element) itExpProp.next();

                if (expProp.getChild("kind").getText().equals("logP")) {
                    logp = Float.parseFloat(expProp.getChild("value").getText());
                }
            }

            List listCalcProp = drug.getChild("calculated-properties").getChildren("property");

            Iterator itCalcProp = listCalcProp.iterator();
            while (itCalcProp.hasNext()) {
                Element calcProp = (Element) itCalcProp.next();

                if (calcProp.getChild("kind").getText().equals("SMILES")) {
                    smiles = calcProp.getChild("value").getText();
                }
            }
            try {
                tmpDrug = new DrugStruct(smiles, logp, name, cas);
                ListDrug.add(tmpDrug.drug);
                good++;
            } catch (CDKException e) {
                System.out.println("---BUG--- " + name + " -- " + cas + " -- " + smiles + " -- " + logp);
                bad++;
            }
            System.out.println(good + "/" + bad);
        }
        return ListDrug;
    }

    //    ExecutorService executors = Executors.newFixedThreadPool(4);
    //                Future<Object> future = executors.submit(new Callable<Object>(){
    //                    public Object call() throws Exception {
    //                        ..
    //                    }
    //                });
}