From c569fc531ddabad8984fff4a3c5fae04b7968b61 Mon Sep 17 00:00:00 2001
From:  <Pabeulo@MSI.localdomain>
Date: Sun, 2 Jun 2019 21:09:12 +0900
Subject: [PATCH] Initial push with maven

---
 CAu_NLP_2019/.idea/compiler.xml           |  13 +
 CAu_NLP_2019/.idea/misc.xml               |  14 +
 CAu_NLP_2019/.idea/workspace.xml          | 393 ++++++++++++++++++++++
 CAu_NLP_2019/CAu_NLP_2019.iml             |   2 +
 CAu_NLP_2019/pom.xml                      |  32 ++
 CAu_NLP_2019/src/main/java/App.java       | 155 +++++++++
 CAu_NLP_2019/src/main/java/Article.java   |  51 +++
 CAu_NLP_2019/target/classes/App.class     | Bin 0 -> 5651 bytes
 CAu_NLP_2019/target/classes/Article.class | Bin 0 -> 1732 bytes
 9 files changed, 660 insertions(+)
 create mode 100644 CAu_NLP_2019/.idea/compiler.xml
 create mode 100644 CAu_NLP_2019/.idea/misc.xml
 create mode 100644 CAu_NLP_2019/.idea/workspace.xml
 create mode 100644 CAu_NLP_2019/CAu_NLP_2019.iml
 create mode 100644 CAu_NLP_2019/pom.xml
 create mode 100644 CAu_NLP_2019/src/main/java/App.java
 create mode 100644 CAu_NLP_2019/src/main/java/Article.java
 create mode 100644 CAu_NLP_2019/target/classes/App.class
 create mode 100644 CAu_NLP_2019/target/classes/Article.class

diff --git a/CAu_NLP_2019/.idea/compiler.xml b/CAu_NLP_2019/.idea/compiler.xml
new file mode 100644
index 0000000..8a9aa33
--- /dev/null
+++ b/CAu_NLP_2019/.idea/compiler.xml
@@ -0,0 +1,13 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="CompilerConfiguration">
+    <annotationProcessing>
+      <profile name="Maven default annotation processors profile" enabled="true">
+        <sourceOutputDir name="target/generated-sources/annotations" />
+        <sourceTestOutputDir name="target/generated-test-sources/test-annotations" />
+        <outputRelativeToContentRoot value="true" />
+        <module name="CAu_NLP_2019" />
+      </profile>
+    </annotationProcessing>
+  </component>
+</project>
\ No newline at end of file
diff --git a/CAu_NLP_2019/.idea/misc.xml b/CAu_NLP_2019/.idea/misc.xml
new file mode 100644
index 0000000..2c4e090
--- /dev/null
+++ b/CAu_NLP_2019/.idea/misc.xml
@@ -0,0 +1,14 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ExternalStorageConfigurationManager" enabled="true" />
+  <component name="MavenProjectsManager">
+    <option name="originalFiles">
+      <list>
+        <option value="$PROJECT_DIR$/pom.xml" />
+      </list>
+    </option>
+  </component>
+  <component name="ProjectRootManager" version="2" languageLevel="JDK_11" project-jdk-name="11" project-jdk-type="JavaSDK">
+    <output url="file://$PROJECT_DIR$/out" />
+  </component>
+</project>
\ No newline at end of file
diff --git a/CAu_NLP_2019/.idea/workspace.xml b/CAu_NLP_2019/.idea/workspace.xml
new file mode 100644
index 0000000..897575a
--- /dev/null
+++ b/CAu_NLP_2019/.idea/workspace.xml
@@ -0,0 +1,393 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ChangeListManager">
+    <list default="true" id="878e5c11-e08f-43ee-a4df-9d0edd20ed32" name="Default Changelist" comment="" />
+    <ignored path="$PROJECT_DIR$/out/" />
+    <ignored path="$PROJECT_DIR$/target/" />
+    <option name="EXCLUDED_CONVERTED_TO_IGNORED" value="true" />
+    <option name="SHOW_DIALOG" value="false" />
+    <option name="HIGHLIGHT_CONFLICTS" value="true" />
+    <option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
+    <option name="LAST_RESOLUTION" value="IGNORE" />
+  </component>
+  <component name="DefaultGradleProjectSettings">
+    <option name="testRunner" value="GRADLE" />
+    <option name="delegatedBuild" value="true" />
+  </component>
+  <component name="FileEditorManager">
+    <leaf>
+      <file pinned="false" current-in-tab="true">
+        <entry file="file://$PROJECT_DIR$/pom.xml">
+          <provider selected="true" editor-type-id="text-editor">
+            <state relative-caret-position="238">
+              <caret line="14" column="16" lean-forward="true" selection-start-line="14" selection-start-column="16" selection-end-line="14" selection-end-column="16" />
+            </state>
+          </provider>
+        </entry>
+      </file>
+      <file pinned="false" current-in-tab="false">
+        <entry file="file://$PROJECT_DIR$/src/main/java/App.java">
+          <provider selected="true" editor-type-id="text-editor">
+            <state relative-caret-position="199">
+              <caret line="100" column="15" selection-start-line="100" selection-start-column="15" selection-end-line="100" selection-end-column="15" />
+              <folding>
+                <element signature="imports" expanded="true" />
+              </folding>
+            </state>
+          </provider>
+        </entry>
+      </file>
+      <file pinned="false" current-in-tab="false">
+        <entry file="file://$PROJECT_DIR$/CAu_NLP_2019.iml">
+          <provider selected="true" editor-type-id="text-editor">
+            <state relative-caret-position="17">
+              <caret line="1" column="41" selection-start-line="1" selection-start-column="41" selection-end-line="1" selection-end-column="41" />
+            </state>
+          </provider>
+        </entry>
+      </file>
+      <file pinned="false" current-in-tab="false">
+        <entry file="file://$PROJECT_DIR$/src/main/java/Article.java">
+          <provider selected="true" editor-type-id="text-editor">
+            <state>
+              <folding>
+                <element signature="e#443#444#0" expanded="true" />
+                <element signature="e#472#473#0" expanded="true" />
+                <element signature="e#516#517#0" expanded="true" />
+                <element signature="e#550#551#0" expanded="true" />
+                <element signature="e#588#589#0" expanded="true" />
+                <element signature="e#616#617#0" expanded="true" />
+                <element signature="e#656#657#0" expanded="true" />
+                <element signature="e#686#687#0" expanded="true" />
+                <element signature="e#721#722#0" expanded="true" />
+                <element signature="e#759#760#0" expanded="true" />
+                <element signature="e#794#795#0" expanded="true" />
+                <element signature="e#830#831#0" expanded="true" />
+                <element signature="e#872#873#0" expanded="true" />
+                <element signature="e#912#913#0" expanded="true" />
+                <element signature="e#949#950#0" expanded="true" />
+                <element signature="e#984#985#0" expanded="true" />
+                <element signature="e#1021#1022#0" expanded="true" />
+                <element signature="e#1057#1058#0" expanded="true" />
+              </folding>
+            </state>
+          </provider>
+        </entry>
+      </file>
+    </leaf>
+  </component>
+  <component name="FileTemplateManagerImpl">
+    <option name="RECENT_TEMPLATES">
+      <list>
+        <option value="Class" />
+      </list>
+    </option>
+  </component>
+  <component name="IdeDocumentHistory">
+    <option name="CHANGED_PATHS">
+      <list>
+        <option value="$PROJECT_DIR$/src/main/java/App.java" />
+        <option value="$PROJECT_DIR$/src/main/java/Article.java" />
+        <option value="$PROJECT_DIR$/CAu_NLP_2019.iml" />
+        <option value="$PROJECT_DIR$/pom.xml" />
+      </list>
+    </option>
+  </component>
+  <component name="ProjectFrameBounds" extendedState="6">
+    <option name="x" value="260" />
+    <option name="y" value="40" />
+    <option name="width" value="1400" />
+    <option name="height" value="1000" />
+  </component>
+  <component name="ProjectView">
+    <navigator proportions="" version="1">
+      <foldersAlwaysOnTop value="true" />
+    </navigator>
+    <panes>
+      <pane id="ProjectPane">
+        <subPane>
+          <expand>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="CAu_NLP_2019" type="462c0819:PsiDirectoryNode" />
+            </path>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="CAu_NLP_2019" type="462c0819:PsiDirectoryNode" />
+              <item name="src" type="462c0819:PsiDirectoryNode" />
+            </path>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="CAu_NLP_2019" type="462c0819:PsiDirectoryNode" />
+              <item name="src" type="462c0819:PsiDirectoryNode" />
+              <item name="main" type="462c0819:PsiDirectoryNode" />
+            </path>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="CAu_NLP_2019" type="462c0819:PsiDirectoryNode" />
+              <item name="src" type="462c0819:PsiDirectoryNode" />
+              <item name="main" type="462c0819:PsiDirectoryNode" />
+              <item name="java" type="462c0819:PsiDirectoryNode" />
+            </path>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="CAu_NLP_2019" type="462c0819:PsiDirectoryNode" />
+              <item name="src" type="462c0819:PsiDirectoryNode" />
+              <item name="test" type="462c0819:PsiDirectoryNode" />
+            </path>
+            <path>
+              <item name="CAu_NLP_2019" type="b2602c69:ProjectViewProjectNode" />
+              <item name="External Libraries" type="cb654da1:ExternalLibrariesNode" />
+            </path>
+          </expand>
+          <select />
+        </subPane>
+      </pane>
+      <pane id="PackagesPane" />
+      <pane id="Scope" />
+    </panes>
+  </component>
+  <component name="PropertiesComponent">
+    <property name="project.structure.last.edited" value="SDKs" />
+    <property name="project.structure.proportion" value="0.0" />
+    <property name="project.structure.side.proportion" value="0.2" />
+    <property name="settings.editor.selected.configurable" value="preferences.lookFeel" />
+  </component>
+  <component name="RunDashboard">
+    <option name="ruleStates">
+      <list>
+        <RuleState>
+          <option name="name" value="ConfigurationTypeDashboardGroupingRule" />
+        </RuleState>
+        <RuleState>
+          <option name="name" value="StatusDashboardGroupingRule" />
+        </RuleState>
+      </list>
+    </option>
+  </component>
+  <component name="RunManager">
+    <configuration name="App" type="Application" factoryName="Application" temporary="true" nameIsGenerated="true">
+      <option name="MAIN_CLASS_NAME" value="App" />
+      <module name="CAu_NLP_2019" />
+      <method v="2">
+        <option name="Make" enabled="true" />
+      </method>
+    </configuration>
+    <recent_temporary>
+      <list>
+        <item itemvalue="Application.App" />
+      </list>
+    </recent_temporary>
+  </component>
+  <component name="SvnConfiguration">
+    <configuration />
+  </component>
+  <component name="TaskManager">
+    <task active="true" id="Default" summary="Default task">
+      <changelist id="878e5c11-e08f-43ee-a4df-9d0edd20ed32" name="Default Changelist" comment="" />
+      <created>1559476475559</created>
+      <option name="number" value="Default" />
+      <option name="presentableId" value="Default" />
+      <updated>1559476475559</updated>
+    </task>
+    <servers />
+  </component>
+  <component name="ToolWindowManager">
+    <frame x="-8" y="-8" width="1936" height="1096" extended-state="6" />
+    <layout>
+      <window_info id="Image Layers" />
+      <window_info id="Designer" />
+      <window_info id="UI Designer" />
+      <window_info id="Capture Tool" />
+      <window_info id="Favorites" side_tool="true" />
+      <window_info content_ui="combo" id="Project" order="0" visible="true" weight="0.17857143" />
+      <window_info id="Structure" order="1" side_tool="true" weight="0.25" />
+      <window_info anchor="bottom" id="Version Control" />
+      <window_info anchor="bottom" id="Terminal" sideWeight="0.49946696" weight="0.32917964" />
+      <window_info anchor="bottom" id="Event Log" sideWeight="0.50053304" side_tool="true" visible="true" weight="0.32917964" />
+      <window_info anchor="bottom" id="Messages" weight="0.32917964" />
+      <window_info anchor="bottom" id="Message" order="0" />
+      <window_info anchor="bottom" id="Find" order="1" />
+      <window_info anchor="bottom" id="Run" order="2" sideWeight="0.49946696" weight="0.32917964" />
+      <window_info anchor="bottom" id="Debug" order="3" weight="0.4" />
+      <window_info anchor="bottom" id="Cvs" order="4" weight="0.25" />
+      <window_info anchor="bottom" id="Inspection" order="5" weight="0.4" />
+      <window_info anchor="bottom" id="TODO" order="6" />
+      <window_info anchor="right" id="Palette" />
+      <window_info anchor="right" id="Theme Preview" />
+      <window_info anchor="right" id="Maven" />
+      <window_info anchor="right" id="Capture Analysis" />
+      <window_info anchor="right" id="Palette&#9;" />
+      <window_info anchor="right" id="Commander" internal_type="SLIDING" order="0" type="SLIDING" weight="0.4" />
+      <window_info anchor="right" id="Ant Build" order="1" weight="0.25" />
+      <window_info anchor="right" content_ui="combo" id="Hierarchy" order="2" weight="0.25" />
+    </layout>
+    <layout-to-restore>
+      <window_info id="Image Layers" order="0" />
+      <window_info id="Designer" order="1" />
+      <window_info id="UI Designer" order="2" />
+      <window_info id="Capture Tool" order="3" />
+      <window_info id="Favorites" order="4" side_tool="true" />
+      <window_info content_ui="combo" id="Project" order="5" visible="true" weight="0.17857143" />
+      <window_info id="Structure" order="6" side_tool="true" weight="0.25" />
+      <window_info anchor="bottom" id="Version Control" order="0" />
+      <window_info anchor="bottom" id="Terminal" order="1" />
+      <window_info anchor="bottom" id="Event Log" order="2" side_tool="true" />
+      <window_info active="true" anchor="bottom" id="Messages" order="3" visible="true" weight="0.32917964" />
+      <window_info anchor="bottom" id="Message" order="4" />
+      <window_info anchor="bottom" id="Find" order="5" />
+      <window_info anchor="bottom" id="Run" order="6" />
+      <window_info anchor="bottom" id="Debug" order="7" weight="0.4" />
+      <window_info anchor="bottom" id="Cvs" order="8" weight="0.25" />
+      <window_info anchor="bottom" id="Inspection" order="9" weight="0.4" />
+      <window_info anchor="bottom" id="TODO" order="10" />
+      <window_info anchor="right" id="Palette" order="0" />
+      <window_info anchor="right" id="Capture Analysis" order="1" />
+      <window_info anchor="right" id="Theme Preview" order="2" />
+      <window_info anchor="right" id="Palette&#9;" order="3" />
+      <window_info anchor="right" id="Maven" order="4" />
+      <window_info anchor="right" id="Commander" internal_type="SLIDING" order="5" type="SLIDING" weight="0.4" />
+      <window_info anchor="right" id="Ant Build" order="6" weight="0.25" />
+      <window_info anchor="right" content_ui="combo" id="Hierarchy" order="7" weight="0.25" />
+    </layout-to-restore>
+  </component>
+  <component name="editorHistoryManager">
+    <entry file="file://$PROJECT_DIR$/CAu_NLP_2019.iml">
+      <provider selected="true" editor-type-id="text-editor">
+        <state relative-caret-position="17">
+          <caret line="1" column="41" selection-start-line="1" selection-start-column="41" selection-end-line="1" selection-end-column="41" />
+        </state>
+      </provider>
+    </entry>
+    <entry file="file://$PROJECT_DIR$/src/main/java/App.java">
+      <provider selected="true" editor-type-id="text-editor">
+        <state relative-caret-position="199">
+          <caret line="100" column="15" selection-start-line="100" selection-start-column="15" selection-end-line="100" selection-end-column="15" />
+          <folding>
+            <element signature="imports" expanded="true" />
+          </folding>
+        </state>
+      </provider>
+    </entry>
+    <entry file="file://$PROJECT_DIR$/src/main/java/Article.java">
+      <provider selected="true" editor-type-id="text-editor">
+        <state>
+          <folding>
+            <element signature="e#443#444#0" expanded="true" />
+            <element signature="e#472#473#0" expanded="true" />
+            <element signature="e#516#517#0" expanded="true" />
+            <element signature="e#550#551#0" expanded="true" />
+            <element signature="e#588#589#0" expanded="true" />
+            <element signature="e#616#617#0" expanded="true" />
+            <element signature="e#656#657#0" expanded="true" />
+            <element signature="e#686#687#0" expanded="true" />
+            <element signature="e#721#722#0" expanded="true" />
+            <element signature="e#759#760#0" expanded="true" />
+            <element signature="e#794#795#0" expanded="true" />
+            <element signature="e#830#831#0" expanded="true" />
+            <element signature="e#872#873#0" expanded="true" />
+            <element signature="e#912#913#0" expanded="true" />
+            <element signature="e#949#950#0" expanded="true" />
+            <element signature="e#984#985#0" expanded="true" />
+            <element signature="e#1021#1022#0" expanded="true" />
+            <element signature="e#1057#1058#0" expanded="true" />
+          </folding>
+        </state>
+      </provider>
+    </entry>
+    <entry file="file://$PROJECT_DIR$/pom.xml">
+      <provider selected="true" editor-type-id="text-editor">
+        <state relative-caret-position="238">
+          <caret line="14" column="16" lean-forward="true" selection-start-line="14" selection-start-column="16" selection-end-line="14" selection-end-column="16" />
+        </state>
+      </provider>
+    </entry>
+  </component>
+  <component name="masterDetails">
+    <states>
+      <state key="ArtifactsStructureConfigurable.UI">
+        <settings>
+          <artifact-editor />
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="FacetStructureConfigurable.UI">
+        <settings>
+          <last-edited>No facets are configured</last-edited>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="GlobalLibrariesConfigurable.UI">
+        <settings>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="JdkListConfigurable.UI">
+        <settings>
+          <last-edited>11</last-edited>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="ModuleStructureConfigurable.UI">
+        <settings>
+          <last-edited>CAu_NLP_2019</last-edited>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+                <option value="0.6" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="ProjectJDKs.UI">
+        <settings>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+      <state key="ProjectLibrariesConfigurable.UI">
+        <settings>
+          <last-edited>Maven: org.jsoup:jsoup:1.11.3</last-edited>
+          <splitter-proportions>
+            <option name="proportions">
+              <list>
+                <option value="0.2" />
+              </list>
+            </option>
+          </splitter-proportions>
+        </settings>
+      </state>
+    </states>
+  </component>
+</project>
\ No newline at end of file
diff --git a/CAu_NLP_2019/CAu_NLP_2019.iml b/CAu_NLP_2019/CAu_NLP_2019.iml
new file mode 100644
index 0000000..4098198
--- /dev/null
+++ b/CAu_NLP_2019/CAu_NLP_2019.iml
@@ -0,0 +1,2 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="JAVA_MODULE" version="4" />
\ No newline at end of file
diff --git a/CAu_NLP_2019/pom.xml b/CAu_NLP_2019/pom.xml
new file mode 100644
index 0000000..f836009
--- /dev/null
+++ b/CAu_NLP_2019/pom.xml
@@ -0,0 +1,32 @@
+<?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>
+    <dependencies>
+    <dependency>
+        <groupId>org.jsoup</groupId>
+        <artifactId>jsoup</artifactId>
+        <version>1.11.3</version>
+    </dependency>
+    </dependencies>
+    <build>
+    <plugins>
+        <plugin>
+            <groupId>org.apache.maven.plugins</groupId>
+            <artifactId>maven-compiler-plugin</artifactId>
+            <version>3.8.1</version>
+            <configuration>
+                <source>1.8</source>
+                <target>1.8</target>
+            </configuration>
+        </plugin>
+    </plugins>
+    </build>
+    <groupId>PQMAN</groupId>
+    <artifactId>CAu_NLP_2019</artifactId>
+    <version>1.0-SNAPSHOT</version>
+    <!-- https://mvnrepository.com/artifact/org.jsoup/jsoup -->
+
+</project>
+
diff --git a/CAu_NLP_2019/src/main/java/App.java b/CAu_NLP_2019/src/main/java/App.java
new file mode 100644
index 0000000..0339684
--- /dev/null
+++ b/CAu_NLP_2019/src/main/java/App.java
@@ -0,0 +1,155 @@
+
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Scanner;
+import java.util.StringTokenizer;
+
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
+import org.jsoup.nodes.Element;
+import org.jsoup.select.Elements;
+
+public class App {
+    public static void main(String[] args) throws Exception {
+        String url[] = {"https://www.thesun.co.uk/?s=", "https://www.bbc.co.uk/search?q=", "https://www.skysports.com/search?q="};
+        //ArrayList<String> key = new ArrayList<String>();
+        Article data = new Article();
+        //key.add("salah");
+        //key.add("Liverpool");
+        //key.add("Champions league");
+        Document doc = null;
+        Elements element = null;
+
+        Scanner scanner = new Scanner(System.in);
+        System.out.print("Please type keywords : ");
+        String key[] = scanner.nextLine().split(",");
+        scanner.close();
+        for (int j = 0; j < url.length; j++) {
+            for (int i = 0; i < key.length; i++) {
+                String urlTmp = url[j] + key[i];
+                doc = Jsoup.connect(urlTmp).execute().parse();
+            /* } catch (IOException e) {
+                   e.printStackTrace();
+             }*/
+                if (url[j].equals("https://www.thesun.co.uk/?s=")) {
+                    element = doc.select("div.search-results-wrap");
+                    for (Element el : element.select(".teaser-item")) {
+                        if (el.select("p").text().toLowerCase().contains(key[i].toLowerCase())) {
+                            data.setHeadline(el.select("p").text());
+                            Elements elUrl = el.select(".teaser__copy-container a");
+                            data.setUrl(elUrl.first().absUrl("href"));
+                            String temp = el.select(".search-date").text();
+                            data.setDate(changeDate(temp));
+                            data.setSite("The Sun");
+                        }
+                    }
+                }
+                if (url[j].equals("https://www.bbc.co.uk/search?q=")) {
+                    element = doc.select("section.search-content");
+                    for (Element el : element.select("li[data-result-number]")) {
+                        data.setHeadline(el.select("h1").select("a").text());
+                        Elements elUrl = el.select("a[href]");
+                        data.setUrl(elUrl.first().absUrl("href"));
+                        String temp = el.select(".display-date").text();
+                        data.setDate(changeDate(temp));
+                        data.setSite("BBC");
+                    }
+                }
+                if (url[j].equals("https://www.skysports.com/search?q=")) {
+                    element = doc.select("div.news-list");
+                    for (Element el : element.select("div.news-list__item")) {
+                        data.setHeadline(el.select("h4").select("a").text());
+                        Elements elUrl = el.select("a[href]");
+                        data.setUrl(elUrl.first().absUrl("href"));
+                        String temp = el.select(".label__timestamp").text();
+                        data.setDate(changeDate2(temp));
+                        data.setSite("SKYSPORTS");
+                    }
+                }
+            }
+        }
+
+        for(int i = 0; i < data.getHowManyData(); i++) {
+            System.out.println(data.getDate(i));
+            System.out.println(data.getHeadline(i));
+            System.out.println(data.getUrl(i));
+            System.out.println(data.getSite(i));
+
+        }
+
+    }
+
+    public static int changeDate2(String date) {
+        date = date.substring(0, 2) + date.substring(2 + 1);
+        date = date.substring(0, 4) + date.substring(4 + 1);
+        String year = date.substring(4, 8);
+        String month = date.substring(2,4);
+        String day = date.substring(0, 2);
+        String fdate = year + month + day;
+
+        int mydate = Integer.parseInt(fdate);
+        return mydate;
+    }
+
+    public static int changeDate(String date) {
+        int formdate = 0;
+        String sp[] = date.split(" ");
+        formdate += Integer.parseInt(sp[2]) * 10000;
+        formdate += Integer.parseInt(sp[0]);
+
+        switch (sp[1]) {
+            case "January":
+            case "Jan":
+                formdate += 100;
+                break;
+            case "February":
+            case "Feb":
+                formdate += 200;
+                break;
+            case "March":
+            case "Mar":
+                formdate += 300;
+                break;
+            case "April":
+            case "Apr":
+                formdate += 400;
+                break;
+            case "May":
+                formdate += 500;
+                break;
+            case "June":
+            case "Jun":
+                formdate += 600;
+                break;
+            case "July":
+            case "Jul":
+                formdate += 700;
+                break;
+            case "August":
+            case "Aug":
+                formdate += 800;
+                break;
+            case "September":
+            case "Sep":
+                formdate += 900;
+                break;
+            case "October":
+            case "Oct":
+                formdate += 1000;
+                break;
+            case "November":
+            case "Nov":
+                formdate += 1100;
+                break;
+            case "December":
+            case "Dec":
+                formdate += 1200;
+                break;
+
+        }
+
+        return formdate;
+
+    }
+}
\ No newline at end of file
diff --git a/CAu_NLP_2019/src/main/java/Article.java b/CAu_NLP_2019/src/main/java/Article.java
new file mode 100644
index 0000000..a2d0d21
--- /dev/null
+++ b/CAu_NLP_2019/src/main/java/Article.java
@@ -0,0 +1,51 @@
+import java.util.ArrayList;
+
+public class Article {
+
+    public Article() {
+        date = new ArrayList<Integer>();
+        Headline = new ArrayList<String>();
+        Url = new ArrayList<String>();
+        Site = new ArrayList<String>();
+    }
+
+    private ArrayList<Integer> date;
+    private ArrayList<String> Headline;
+    private ArrayList<String> Url;
+    private ArrayList<String> Site;
+
+    public void setDate(int num) {
+        date.add(num);
+    }
+
+    public void setHeadline(String head) {
+        Headline.add(head);
+    }
+
+    public void setUrl(String url) {
+        Url.add(url);
+    }
+
+    public void setSite(String site) {
+        Site.add(site);
+    }
+
+    public int getHowManyData() {
+        return Headline.size();
+    }
+
+    public int getDate(int num) {
+        return date.get(num);
+    }
+
+    public String getHeadline(int num) {
+        return Headline.get(num);
+    }
+
+    public String getUrl(int num) {
+        return Url.get(num);
+    }
+    public String getSite(int num) {
+        return Site.get(num);
+    }
+}
\ No newline at end of file
diff --git a/CAu_NLP_2019/target/classes/App.class b/CAu_NLP_2019/target/classes/App.class
new file mode 100644
index 0000000000000000000000000000000000000000..c9ee7d7aaf599cf19be5890b484233a649f54999
GIT binary patch
literal 5651
zcma)93w%`7ng4(H&dj};+`w>20wF=dI|;c7B18<~89;*}0znYb!p-E83`}N*c@SdV
zty@vMm9DO?t-5`<Zrk)>w^p}<RO_<Vw$|F`uC?uUTeaKOb=}t8x>b?F{?DBm5+J{J
z^UFEs`@ZwN&iNka&a<b#@+5$zvL=iuK5XD4Chj$HpNabo{B;;1Mf+$OK8BBnVd4`C
zdcefrgz+H$R#84_;!|O0__RVs4SXgH8y{BFXVvt%Fg}lm%JBvKT{*sphZXCWOgs|C
z5j<+*@58tSkE-d*<+v4jH61hXn2F<I%r5MFMaev_rYDp?Up4Wh68HxbClt>=D&kWn
zz7|Fwo>s`$O?*SG-!$=ziEo*B*2K3>JZIuNCcbOpdnUeb;s+*vs6_tB#6K(U=hgHh
zb*>i_^kW76L_t4Q)4wSDKU34chVgU!!o-UTdCA1fCVpw+-%R|<#J`((#l)*7{=>v;
zCjQgJ>n8ro#IH^K#>5*Y{@cX=nE0)UH%<JniQk!c%f#CT-Z5}e5bB5}V%b#!y`gc3
zAkdlY_XLx>V+pS(H`M2)x4V6DQta+z)Q#_O(=mk?iGl23EF%bZuT7;^P<Y6VB?Q$C
zySw+hhuqe<n;2;A&8B0CffY)5(2HN4jtgqLlj(uh{h4Gg)td3*UNqafA?^)%iEKt`
z$$CR6W$*$W0?ng%%{cJ{-OZGwDIChB<J*<E#5lS*_EV#q9-y03Cvc1BrUY5~-K;Ap
z>t37A#-eeLs`UP3R50TLW9yUAobr_{2Ax+Rp-j|GB)l}O6#UL*WAWDB0#PYD;Ef0>
zdb4iyz-Bk)-=BfQ2HrJr%E0dh;SGnQUMd?)CNc(23(BK|ygqNeoAs6msv0i5jz%6&
z`OWAAMm(3cgG0$gc2K$Q(pqp|VKX!|QUJnZDh1}gWO}GT*D}1YaBnh~j(QtoD%vIk
z+M@2?LO+N{&X0wlZZMloW!hU?hlht-vV&eGmuQJ5TXF|lS7%lVW{uzO>x&lmGW0V#
zxcVSD=8osc92m)@lId)Qd_zo%vs^|oAc85r599CEz!{4X7CI@KCAtIz)3(GtH{;c3
zM^aurFMT+f?$6Y>*BcVFBqW9<rj${r*^;o7Ta@MTrk#Jhb-7r)-%DFkA(JebEM<m7
zESVy<AyX}>lxc=cx1>g9SW+u>RL~zg)Kc);N%Jb=*^D!sc2kzjl#2vbOIA6Rc4EwJ
zOJ+&EC9`FYC39sSg~zJloOGs_R4`x4ELotYh0<V2qZq7!qP{(QqRG^V6J^v{d0x8S
zwWLWFF&_rg-abp3#j&JCNmwQCRKzX(B|d0Lt1M>XZXfjOdvgg(E>?Wi8CD~6u*9xZ
z`w3nhkL{+cTRfhV@SDT+EXl!(=_qT*aCa-KBvrKZ$1<t7JK_}bOJBFHlT>v_32!*#
z#A6wj%+z!6J$uyMD0NF2>Xx|M=f(Hz$;O7fOx7JrS@?5&koDPn<*wc>S8dzgYsnJ1
z#FC}5%#t=)Y{{jvTtN)*WeRFn(+cTO({e*rTCz$hP&dC?E}^%Z+(gbzvrKeC%=nF7
zU%E&j$m-dwa**3xiLFhgW9$<;SD!|~z@}V+$)j`ek+`ocN3PJ?+(1s9Q753hW?_4(
znUX@7)~||YlLed*t@k7kl^6(;u-=Q72n3a;6y?U%;Z=S687jf_^RuVapO_|NTj#Y|
zw%4(0>dJa)H=Cq$Q^)GyY&SZ`Xzxg-5GB>;s|qbCmTc`xq;gp{de0qVubW=%Rp)J}
z7eG(u=<=B07QPK*f(7#w8%O5V^X65Hov;N`r<w2$XVrTpsBCDwuuW0}lQBjO3nyx#
z6Fv&TXgryr!jPLvd5M03Gm+>8wu)_ocVK3dg@=!C=~xstsV9%oXH_Nq;5ALCc1)nN
zz{rB6&dQ1C)aD)bqB->nnA<!y(S&l`>Qb*!XWY$XJU)H1c`*}7&nrn#{mehepyJ@B
zmFl{4bQL;QDd(sr<Q>epaoRj5UWII3(ReKx3b`m~{Ue7zFK5-QvPTqCwYn5)|3N6F
zFnNqYH`Bw6V6rtRa{-ke=ch)|9qwlR-STX*J2}h?W0$5G)%Dm=d5GHaK6Vbe>0Tbs
zOGG{YDCHS1dxhur$NlUJ?u(^W(brCJLB;PIbw6%j#{UXhFkW@(E9ruiQWCV{OBe*!
z7vB4ZE>%v7(91g!Od6mk$>GgzVuT#*#ymVf4-AsWBox6frZ}s&TuGeuy^0Dav`wyp
zZ2|>TWtm)G#{Z~OpL$&r%M~7L{{yg#FUWvueWss4L<urTgX&Y0AJcvO5upA85cRRa
zc^|P>uEI6gbPVz(0vZN5hkRlM1_@gQ2E?$RSQ!o=PA&=uQdBmHB3|gL(01vz-n%Pc
z2YPn}?O<>B;|T3K24l1INPFOMn7bZ9*)fFMgO8xxw(_W`3Fa}$o}5RdJ*40%dYe&Y
zEPC=md0uZbtIV1Z!KWXGO*K;&RT)e2sH_Q9nQG0Ws=chH>@idy$27rlOc(Sh9$U|&
z=04o+*fa8|Z4cB0?7BQ==5bMbxF&oYvjq2HNI~_26PVpzUQ=$*$zyIF^YWNq)IUFu
z1^1z|dHmi&A`N70x0-9{exu!#$D%wGk5iCuuV}8RcnmFhv^Lv|^SF5Q%pJZ@0pF*b
zW7GA5PnTTa)6VfeDWyxP)L!QE(KBVeZDQ+}PGEhxZ~argbzUuRgw`)BS#SU2)>r(0
zt#_PbeWic)RR!xuOc_1he#GG1<j1x-kJS(2GA2NCm66AqJl2-rb$N7_;PvC+4TP13
zGf%7J)cxfAA(j%Jj7m&F9jb5<s+r%@;h+Y~Fav8*i;djbf|=OO-9wng-=1@E7v|x9
z%*PX0fETb3zd{3kjYgb8lbBc}Q_w8a&>}OiSQg-7X~7cNf=eXI-_UU^lN-<`x8hQ{
z1Iy(eTqgHog?thn@(@<acd<%d!fJU9YvgUL)dE<jRiIOwiuGDGW@>e))8?Q{Yr!V1
z9am}_(XDO6X6-$s^<%4FXwULQ%AX-5!0%_CFo_f@=<jYEL>lGvqO^N!X_p>uDDJ*i
z+NJfPEd46(;*dOm4E2?vSH6WTa{N)>B|qWq<X~v?vD&8@s+=*5!@S{2Z7W6y@h#H2
za0A}UvcE{X95>=m7>U{1di*Ip4ZzWs<9)aZL7saa{)`eKyr1%0QE>)cFb&-7|A!15
zF>s6FQ`mG)p-g^{Z3Z+0x4w(iNoc1qi||QYUjW|0^(PVJ|0!s4602|<)v5;e0Um>2
zs)SS%dzG{h*Hp1tH0jLwjcgN5p*$`xw2Ldx?lJTHJ#%cY$%qi!*xeLVo0|*rJ*Cz|
z1<c$^R=MAI*i#<E6>Y();6t=o=$CVNnZR}?-PKr*9aw{%xE$A@2fOIuZie(aKg6pF
ze&Tip0_0kZqlARKjrb5DjS_S47ld@mg>eU=0Ht-@$uI?ZYW}1soH{U_)F5l3LGNg?
zyVyoI*ER>2(6uQGBZGT8YMXV1?Cod{Ea}nO0(5`V=as{aoTj}w>(QsUzWBSHZ*SuI
z)ahI$3yi5On$ESE>)l*G3Mqe?GIfxt>iPH$WcFKJe*l^1a{orIq|0>>XBF29t{b>^
zK$d)&>nX@GvdhmP>(%}fm^-+fYn*Ez*KX~uPoLoWoc_%HjHJ{B_P;!r0;lKi+)F*0
z<$dUuZKS`RJGz5(IqKL~s&*!>-WC{@-HzSuquT-s?RM<VKH3>n=t{@#@zJFrg|;~M
zRX#f3Q0Oej-r}Rxra~i*z12sJh|dkj-llN86;bSX!?AmP{FR8}$BT}=-N#>ulqvlC
zj(xR{e=`zR_zB0};p4|5<qCh;v3L6TXv9+Z1CD)-kAEamq3}B$dzX*j7MY~*n;iRE
zA0LiPR(Q&>cl-E2B%<)Wj(wev?~F`Q_!h^$-p4PG*a~0c*zfW2OCwVizSyz%`1ry|
zrNZkSd#{gAi&QB*;@GZ_8<Fa^Kvm##jDm`D;jfy@f#NvGRgD{p<7TdE9ObyDI6lr*
zjnU$Gn5!Dci$mp}8sApqi1d#l`t%Xz^)sWmRsqsHimg6y=P0i5fj397S^?TMqiFMi
zCr05YK(8OgJRdkRidr9dbriM&0^Or1^MQwJRkao#kVAYJfW53Xmlf5==OfA|!qZM*
zQ2QEU+BXo_p2Lv#J*2efaZvj)vf9s)(_X@`_6iPbui?Gg8@N&XE#9ZSgPYj3Zq_v%
z(Sx{6FUJS;Nw{6F#8G`3?$BrAPQ4y?>kDwN-vR3B18cU}0W{L@VrBEkt3C^Nvvq2I
z>+Hb%Gk6_A_D1o0r^bFc$^V;t-k<d@;z9H5-#hFH8vcss({PWdxP__xi}-?6h06X9
DKpa$V

literal 0
HcmV?d00001

diff --git a/CAu_NLP_2019/target/classes/Article.class b/CAu_NLP_2019/target/classes/Article.class
new file mode 100644
index 0000000000000000000000000000000000000000..a8a494f1c792d15535d260cc9e7d3a8c8dfeae1f
GIT binary patch
literal 1732
zcmaizTT|0e6opsX(zGE^KnqG0ykc8~0wP|D$X%*M9;!I{7TQ?ClyoLdaqu_!pfg$=
zopF5jM>(#O#HMA!@Q^)cpUbz`+9&7F-ygpK%wt=@02X9CP!PgG90@!W;gJZBMaU~q
zu_#ubh_EEWQyI%*dPT;nj5QhS5+W7N)+Hp12ilRAYuZLFw_;h^anWelc?t27Q8hKY
zX>m_BXwRZMS<}pFu3*}FRk!j>JQ&;3wMxw}c`kF_xsq)eX4N?ser?q#s$>XJ(M7{F
z>?H}|bY_>cYxN3e6Gc*AHV^l7>y@@w<IGU8Ue;>6nq`RGn~d1|MuP^5E0%4PYdST_
z4c%UM>_*ZBK@6MCLkUtrLNAwm#mRJsD^H!HV4tQ$J*QcYXK$5aNpQr6o5F&q<+_M8
z95ek@R;Yj2*34sS&>T}NEOXk|B$SAK?cOCRG_V?XIn~5@w~kV(H?6Y1X^57np6h9$
zRK;bCO1Kbk#VQ6dq#`N8hzOTN*ubWWXV_9vz?6#Tcp+iX_w3!XiXx^Y4EPIfGib`I
z-nF>1cc7PTE*@#MroQuD!f>~_|Gy7gqgB}Eu3K<V=QHp4A2f`Qv>2vyv8k&;-ZuV!
zhH3ALLb%S?Fu+F>;R~Qyuo28cxX6#b+xakawxBC#H#%~KV|J1;MM526j)2+h38d^#
zgx;PY{0$L**y9hQ{!sqnl)u6_iYQ|br9~-HW*)<B#4!T}SGl6P1u%wjW)iN^j9Ywz
zVndwk%YH>9^&PPm;%|Jgc<rEem0kr%c;jXdBt|!IlPC$dSkOhTdPs%Hp2;BIFr|%j
z@RT-B!36F&ymTNh8ps<ZFLeel<MFbAJUNgzM&9@tyh)FD7gL_!IY%!L6MkDzKCwd~
zh--;*1?-T21+n%OOe5!=U-SBvaIT<*-rrDL=sS&>=n#}>H!RV<fEmnsz-14p3PC9c
n*#G&IsB8yKZPU~?Z7z@&J%cuV4((nbO+JG*dk$^h@pAtkOJ5b5

literal 0
HcmV?d00001