Exercise snippets only in tests (#982)
* Exercise snippets only in tests * Increase Jsoup version
This commit is contained in:
parent
a0de46efb9
commit
4772976bda
|
@ -25,6 +25,6 @@
|
||||||
<maven.compiler.source>1.8</maven.compiler.source>
|
<maven.compiler.source>1.8</maven.compiler.source>
|
||||||
<maven.compiler.target>1.8</maven.compiler.target>
|
<maven.compiler.target>1.8</maven.compiler.target>
|
||||||
|
|
||||||
<jsoup.version>1.10.1</jsoup.version>
|
<jsoup.version>1.10.2</jsoup.version>
|
||||||
</properties>
|
</properties>
|
||||||
</project>
|
</project>
|
||||||
|
|
|
@ -1,91 +0,0 @@
|
||||||
package com.baeldung.jsoup;
|
|
||||||
|
|
||||||
import java.io.IOException;
|
|
||||||
import org.jsoup.Jsoup;
|
|
||||||
import org.jsoup.nodes.Document;
|
|
||||||
import org.jsoup.nodes.Element;
|
|
||||||
import org.jsoup.parser.Tag;
|
|
||||||
import org.jsoup.select.Elements;
|
|
||||||
|
|
||||||
public class JsoupParser {
|
|
||||||
|
|
||||||
Document doc;
|
|
||||||
|
|
||||||
public void loadDocument(String blogUrl) throws IOException {
|
|
||||||
doc = Jsoup.connect(blogUrl).get();
|
|
||||||
}
|
|
||||||
|
|
||||||
void loadDocumentCustomized(String blogUrl) throws IOException {
|
|
||||||
doc = Jsoup.connect(blogUrl)
|
|
||||||
.userAgent("Mozilla")
|
|
||||||
.timeout(5000)
|
|
||||||
.cookie("cookiename", "val234")
|
|
||||||
.cookie("anothercookie", "ilovejsoup")
|
|
||||||
.referrer("http://google.com")
|
|
||||||
.header("headersecurity", "xyz123")
|
|
||||||
.get();
|
|
||||||
}
|
|
||||||
|
|
||||||
void examplesSelectors() {
|
|
||||||
Elements links = doc.select("a");
|
|
||||||
Elements logo = doc.select(".spring-logo--container");
|
|
||||||
Elements pagination = doc.select("#pagination_control");
|
|
||||||
Elements divsDescendant = doc.select("header div");
|
|
||||||
Elements divsDirect = doc.select("header > div");
|
|
||||||
|
|
||||||
Element pag = doc.getElementById("pagination_control");
|
|
||||||
Elements desktopOnly = doc.getElementsByClass("desktopOnly");
|
|
||||||
|
|
||||||
Elements sections = doc.select("section");
|
|
||||||
Element firstSection = sections.first();
|
|
||||||
Elements sectionParagraphs = firstSection.select(".paragraph");
|
|
||||||
}
|
|
||||||
|
|
||||||
void examplesTraversing() {
|
|
||||||
Elements sections = doc.select("section");
|
|
||||||
|
|
||||||
Element firstSection = sections.first();
|
|
||||||
Element lastSection = sections.last();
|
|
||||||
Element secondSection = sections.get(2);
|
|
||||||
Elements allParents = firstSection.parents();
|
|
||||||
Element parent = firstSection.parent();
|
|
||||||
Elements children = firstSection.children();
|
|
||||||
Elements siblings = firstSection.siblingElements();
|
|
||||||
|
|
||||||
sections.stream().forEach(el -> System.out.println("section: " + el));
|
|
||||||
}
|
|
||||||
|
|
||||||
void examplesExtracting() {
|
|
||||||
Element firstArticle = doc.select("article").first();
|
|
||||||
Element timeElement = firstArticle.select("time").first();
|
|
||||||
String dateTimeOfFirstArticle = timeElement.attr("datetime");
|
|
||||||
Element sectionDiv = firstArticle.select("section div").first();
|
|
||||||
String sectionDivText = sectionDiv.text();
|
|
||||||
String articleHtml = firstArticle.html();
|
|
||||||
String outerHtml = firstArticle.outerHtml();
|
|
||||||
}
|
|
||||||
|
|
||||||
void examplesModifying() {
|
|
||||||
Element firstArticle = doc.select("article").first();
|
|
||||||
Element timeElement = firstArticle.select("time").first();
|
|
||||||
Element sectionDiv = firstArticle.select("section div").first();
|
|
||||||
|
|
||||||
String dateTimeOfFirstArticle = timeElement.attr("datetime");
|
|
||||||
timeElement.attr("datetime", "2016-12-16 15:19:54.3");
|
|
||||||
sectionDiv.text("foo bar");
|
|
||||||
firstArticle.select("h2").html("<div><span></span></div>");
|
|
||||||
|
|
||||||
Element link = new Element(Tag.valueOf("a"), "")
|
|
||||||
.text("Checkout this amazing website!")
|
|
||||||
.attr("href", "http://baeldung.com")
|
|
||||||
.attr("target", "_blank");
|
|
||||||
firstArticle.appendChild(link);
|
|
||||||
|
|
||||||
doc.select("li.navbar-link").remove();
|
|
||||||
firstArticle.select("img").remove();
|
|
||||||
}
|
|
||||||
|
|
||||||
String getTidyHtml() {
|
|
||||||
return doc.html();
|
|
||||||
}
|
|
||||||
}
|
|
|
@ -2,6 +2,11 @@ package com.baeldung.jsoup;
|
||||||
|
|
||||||
import java.io.IOException;
|
import java.io.IOException;
|
||||||
import org.jsoup.HttpStatusException;
|
import org.jsoup.HttpStatusException;
|
||||||
|
import org.jsoup.Jsoup;
|
||||||
|
import org.jsoup.nodes.Document;
|
||||||
|
import org.jsoup.nodes.Element;
|
||||||
|
import org.jsoup.parser.Tag;
|
||||||
|
import org.jsoup.select.Elements;
|
||||||
import static org.junit.Assert.assertEquals;
|
import static org.junit.Assert.assertEquals;
|
||||||
import static org.junit.Assert.assertTrue;
|
import static org.junit.Assert.assertTrue;
|
||||||
import org.junit.Before;
|
import org.junit.Before;
|
||||||
|
@ -9,28 +14,96 @@ import org.junit.Test;
|
||||||
|
|
||||||
public class JsoupParserTest {
|
public class JsoupParserTest {
|
||||||
|
|
||||||
JsoupParser jsoupParser;
|
Document doc;
|
||||||
|
|
||||||
@Before
|
@Before
|
||||||
public void setUp() {
|
public void setUp() throws IOException {
|
||||||
jsoupParser = new JsoupParser();
|
doc = Jsoup.connect("https://spring.io/blog").get();
|
||||||
}
|
}
|
||||||
|
|
||||||
@Test
|
@Test
|
||||||
public void test404() throws IOException {
|
public void loadDocument404() throws IOException {
|
||||||
try {
|
try {
|
||||||
jsoupParser.loadDocument("https://spring.io/will-not-be-found");
|
doc = Jsoup.connect("https://spring.io/will-not-be-found").get();
|
||||||
} catch (HttpStatusException ex) {
|
} catch (HttpStatusException ex) {
|
||||||
assertEquals(404, ex.getStatusCode());
|
assertEquals(404, ex.getStatusCode());
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
@Test
|
@Test
|
||||||
public void testChange() throws IOException {
|
public void loadDocumentCustomized() throws IOException {
|
||||||
jsoupParser.loadDocument("http://spring.io/blog");
|
doc = Jsoup.connect("https://spring.io/blog")
|
||||||
|
.userAgent("Mozilla")
|
||||||
|
.timeout(5000)
|
||||||
|
.cookie("cookiename", "val234")
|
||||||
|
.cookie("anothercookie", "ilovejsoup")
|
||||||
|
.referrer("http://google.com")
|
||||||
|
.header("headersecurity", "xyz123")
|
||||||
|
.get();
|
||||||
|
}
|
||||||
|
|
||||||
jsoupParser.examplesModifying();
|
@Test
|
||||||
|
public void examplesSelectors() {
|
||||||
|
Elements links = doc.select("a");
|
||||||
|
Elements logo = doc.select(".spring-logo--container");
|
||||||
|
Elements pagination = doc.select("#pagination_control");
|
||||||
|
Elements divsDescendant = doc.select("header div");
|
||||||
|
Elements divsDirect = doc.select("header > div");
|
||||||
|
|
||||||
assertTrue(jsoupParser.getTidyHtml().contains("http://baeldung.com"));
|
Element pag = doc.getElementById("pagination_control");
|
||||||
|
Elements desktopOnly = doc.getElementsByClass("desktopOnly");
|
||||||
|
|
||||||
|
Elements sections = doc.select("section");
|
||||||
|
Element firstSection = sections.first();
|
||||||
|
Elements sectionParagraphs = firstSection.select(".paragraph");
|
||||||
|
}
|
||||||
|
|
||||||
|
@Test
|
||||||
|
public void examplesTraversing() {
|
||||||
|
Elements sections = doc.select("section");
|
||||||
|
|
||||||
|
Element firstSection = sections.first();
|
||||||
|
Element lastSection = sections.last();
|
||||||
|
Element secondSection = sections.get(2);
|
||||||
|
Elements allParents = firstSection.parents();
|
||||||
|
Element parent = firstSection.parent();
|
||||||
|
Elements children = firstSection.children();
|
||||||
|
Elements siblings = firstSection.siblingElements();
|
||||||
|
|
||||||
|
sections.stream().forEach(el -> System.out.println("section: " + el));
|
||||||
|
}
|
||||||
|
|
||||||
|
@Test
|
||||||
|
public void examplesExtracting() {
|
||||||
|
Element firstArticle = doc.select("article").first();
|
||||||
|
Element timeElement = firstArticle.select("time").first();
|
||||||
|
String dateTimeOfFirstArticle = timeElement.attr("datetime");
|
||||||
|
Element sectionDiv = firstArticle.select("section div").first();
|
||||||
|
String sectionDivText = sectionDiv.text();
|
||||||
|
String articleHtml = firstArticle.html();
|
||||||
|
String outerHtml = firstArticle.outerHtml();
|
||||||
|
}
|
||||||
|
|
||||||
|
@Test
|
||||||
|
public void examplesModifying() {
|
||||||
|
Element firstArticle = doc.select("article").first();
|
||||||
|
Element timeElement = firstArticle.select("time").first();
|
||||||
|
Element sectionDiv = firstArticle.select("section div").first();
|
||||||
|
|
||||||
|
String dateTimeOfFirstArticle = timeElement.attr("datetime");
|
||||||
|
timeElement.attr("datetime", "2016-12-16 15:19:54.3");
|
||||||
|
sectionDiv.text("foo bar");
|
||||||
|
firstArticle.select("h2").html("<div><span></span></div>");
|
||||||
|
|
||||||
|
Element link = new Element(Tag.valueOf("a"), "")
|
||||||
|
.text("Checkout this amazing website!")
|
||||||
|
.attr("href", "http://baeldung.com")
|
||||||
|
.attr("target", "_blank");
|
||||||
|
firstArticle.appendChild(link);
|
||||||
|
|
||||||
|
doc.select("li.navbar-link").remove();
|
||||||
|
firstArticle.select("img").remove();
|
||||||
|
|
||||||
|
assertTrue(doc.html().contains("http://baeldung.com"));
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
Loading…
Reference in New Issue