Initial commit

selenium/pom.xml

@@ -0,0 +1,29 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project xmlns="http://maven.apache.org/POM/4.0.0"
+         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+    <modelVersion>4.0.0</modelVersion>
+    <parent>
+        <groupId>ch.gtache.fro</groupId>
+        <artifactId>fro</artifactId>
+        <version>1.0.0-SNAPSHOT</version>
+    </parent>
+
+    <artifactId>fro-selenium</artifactId>
+
+    <properties>
+        <selenium.version>4.35.0</selenium.version>
+    </properties>
+
+    <dependencies>
+        <dependency>
+            <groupId>ch.gtache.fro</groupId>
+            <artifactId>fro-jsoup</artifactId>
+        </dependency>
+        <dependency>
+            <groupId>org.seleniumhq.selenium</groupId>
+            <artifactId>selenium-java</artifactId>
+            <version>${selenium.version}</version>
+        </dependency>
+    </dependencies>
+</project>

selenium/src/main/java/ch/gtache/fro/selenium/AbstractSeleniumFetcher.java

@@ -0,0 +1,73 @@
+package ch.gtache.fro.selenium;
+
+import ch.gtache.fro.BirdProvider;
+import ch.gtache.fro.Configuration;
+import ch.gtache.fro.Fetcher;
+import ch.gtache.fro.jsoup.AbstractJSoupFetcher;
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
+import org.openqa.selenium.WebDriver;
+import org.openqa.selenium.chrome.ChromeDriver;
+import org.openqa.selenium.chrome.ChromeOptions;
+
+import java.io.IOException;
+
+/**
+ * Abstract implementation of {@link Fetcher} using Selenium
+ */
+public abstract class AbstractSeleniumFetcher extends AbstractJSoupFetcher implements AutoCloseable {
+    private final ChromeDriver driver;
+
+    /**
+     * Instantiates the fetcher
+     *
+     * @param birdProvider  The bird provider
+     * @param configuration The configuration
+     * @throws NullPointerException If any parameter is null
+     */
+    protected AbstractSeleniumFetcher(final BirdProvider birdProvider, final Configuration configuration) {
+        super(birdProvider, configuration);
+        final var options = new ChromeOptions();
+        options.addArguments("--headless=new");
+        this.driver = new ChromeDriver(options);
+    }
+
+    /**
+     * Returns the web driver
+     *
+     * @return The driver
+     */
+    protected WebDriver driver() {
+        return driver;
+    }
+
+    /**
+     * Retrieves a document from an URL
+     *
+     * @param url The URL
+     * @throws IOException If an error occurs
+     */
+    @Override
+    protected Document getDocument(final String url) throws IOException {
+        driver.get(url);
+        waitFor();
+        final var source = (String) driver.executeScript("return document.documentElement.outerHTML");
+        if (source == null) {
+            throw new IOException("Page source is null");
+        } else {
+            return Jsoup.parse(source);
+        }
+    }
+
+    /**
+     * Waits for the page to be loaded
+     *
+     * @throws IOException If an error occurs
+     */
+    protected abstract void waitFor() throws IOException;
+
+    @Override
+    public void close() {
+        driver.quit();
+    }
+}

selenium/src/main/java/module-info.java

@@ -0,0 +1,11 @@
+/**
+ * Selenium fetcher module for FRO
+ */
+module ch.gtache.fro.selenium {
+    requires transitive ch.gtache.fro.jsoup;
+    requires org.seleniumhq.selenium.chrome_driver;
+    requires org.jsoup;
+    requires org.seleniumhq.selenium.devtools_v137;
+    requires ch.gtache.fro.api;
+    exports ch.gtache.fro.selenium;
+}