Add basic functionality tests migrated from jsoup

test/migrate_unittest/test-group1.cpp

@@ -0,0 +1,125 @@
+#include <iostream>
+#include <gtest/gtest.h>
+#include <string>
+#include "html.hpp"
+
+using namespace std;
+using namespace html;
+
+
+//test1
+TEST(test, dropsUnterminatedTag){
+    string h1 = "<p";
+    parser parse;
+    node_ptr doc = parse.parse(h1);
+    ASSERT_EQ(0, doc->get_attr("p").size());
+    ASSERT_EQ("", doc->to_text()); 
+
+    string h2="<div id=1<p id='2'";
+    doc=parse.parse(h2);
+    ASSERT_EQ("", doc->to_text()); 
+}
+
+//test2
+TEST(test, testByAttributeRegexCombined){
+    parser parse;
+    node_ptr doc = parse.parse("<div><table class=x><td>Hello</td></table></div>");
+    node_ptr els = doc->select("div table[class~=x|y]");
+
+    ASSERT_EQ(1, els->size());
+    ASSERT_EQ("Hello", els->to_text());
+}
+
+//test3
+TEST(test, testAllWithClass){
+    string h = "<p class=first></p>One<p class=first>Two<p>Three";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    node_ptr ps = doc->select("[class*=first]");
+    // The last text node, although it does not have a class attribute itself,
+    // it is selected because its parent node <p> has a class attribute value that includes "first".
+    ASSERT_EQ(3, ps->size());
+}
+
+//test4
+// select is case sensitive.
+TEST(test, caseInsensitive){
+    string h = "<div tItle=bAr></div>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    // ASSERT_EQ(1, doc->select("DiV")->size());
+    // ASSERT_EQ(1, doc->select("dIv")->size());
+}
+
+//test5
+TEST(test, notAdjacent){
+    parser parse;
+    string h = "<ol><li id=1>One<li id=2>Two<li id=3>Three</ol>";
+    node_ptr doc = parse.parse(h);
+    node_ptr sibs = doc->select("li#1 + li#3");
+    ASSERT_EQ(0, sibs->size());
+}
+
+//test6
+TEST(test, selectSameElements){
+    parser parse;
+    string html = "<div>one</div><div>one</div>";
+    node_ptr doc = parse.parse(html);
+    node_ptr els = doc->select("div");
+    ASSERT_EQ(2, els->size());
+
+    // not support :contains
+    // node_ptr subSelect = els->select(":contains(one)");
+    // ASSERT_EQ(2, subSelect->size());
+}
+
+//test7
+TEST(test, matchTextAttributes){
+    parser parse;
+    node_ptr doc = parse.parse("<div><p class=one>One<br>Two<p class=two>Three<br>Four");
+
+    //":matchText" selector is not a standard selector in HTML.
+    node_ptr els = doc->select("p.two:matchText:last-child");
+
+    ASSERT_EQ(0, els->size());
+    // ASSERT_EQ("Four", els->to_text());
+}
+
+//test8
+TEST(test, startsWithBeginsWithSpace){
+    parser parse;
+    node_ptr doc = parse.parse("<small><a href=\" mailto:abc@def.net\">(abc@def.net)</a></small>");
+
+    node_ptr els = doc->select("a[href^=' mailto']");
+
+    ASSERT_EQ(1, els->size());
+}
+
+//test9
+TEST(test, endsWithEndsWithSpaces){
+    parser parse;
+    node_ptr doc = parse.parse("<small><a href=\" mailto:abc@def.net \">(abc@def.net)</a></small>");
+
+    node_ptr els = doc->select("a[href$='.net ']");
+
+    ASSERT_EQ(1, els->size());
+}
+
+//test10
+TEST(test, parsesQuiteRoughAttributes){
+    string html = "<p =a>One<a <p>Something</p>Else";
+    parser parse;
+    node_ptr doc = parse.parse(html);
+
+    // <p =a=\"\">One<a <p=\"\">Something</a></p>\nElse
+    ASSERT_EQ("<p =a=\"\">One<a <p=\"\">Something</a></p>\nElse", doc->to_html());
+
+    doc = parse.parse("<p .....>");
+    ASSERT_EQ("<p .....=\"\"></p>", doc->to_html());
+}
+
+GTEST_API_ int main(int argc, char ** argv) {
+    testing::InitGoogleTest(&argc, argv);
+    return RUN_ALL_TESTS();
+}
+

test/migrate_unittest/test-group2.cpp

@@ -0,0 +1,99 @@
+#include <iostream>
+#include <gtest/gtest.h>
+#include <string>
+#include "html.hpp"
+
+using namespace std;
+using namespace html;
+
+//test11
+TEST(test, dropsUnterminatedAttribute){
+    string h1 = "<p id=\"foo";
+    parser parse;
+    node_ptr doc = parse.parse(h1);
+
+    ASSERT_EQ("", doc->to_text());
+}
+
+//test12
+TEST(test, testSpaceAfterTag){
+    parser parse;
+    node_ptr doc = parse.parse("<div > <a name=\"top\"></a ><p id=1 >Hello</p></div>");
+
+    ASSERT_EQ("<div><a name=\"top\"></a>\n\t<p id=\"1\">Hello</p>\n</div>", doc->to_html());
+}
+
+//test13
+TEST(test, createsStructureFromBodySnippet){
+    string html = "foo <b>bar</b> baz";
+    parser parse;
+    node_ptr doc = parse.parse(html);
+
+    ASSERT_EQ("foo bar baz", doc->to_text());
+}
+
+//test14
+TEST(test, handlesTextAfterData){
+    string h = "<html><body>pre <script>inner</script> aft</body></html>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    //<html><head></head><body>pre <script>inner</script> aft</body></html>
+    // no head
+    ASSERT_EQ("<html>\n\t<body>pre \n\t\t<script>inner</script>\n\t\t aft\n\t</body>\n</html>", doc->to_html());
+}
+
+//test15
+TEST(test, handlesTextArea){
+    parser parse;
+    node_ptr doc = parse.parse("<textarea>Hello</textarea>");
+    node_ptr els = doc->select("textarea");
+    ASSERT_EQ("Hello", els->to_text());
+}
+
+//test16
+TEST(test, discardsNakedTds){
+    string h = "<td>Hello<td><p>There<p>now";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ("<td>Hello\n\t<td>\n\t\t<p>There\n\t\t\t<p>now</p>\n\t\t</p>\n\t</td>\n</td>", doc->to_html());
+}
+
+//test17
+TEST(test, handlesNestedImplicitTable){
+    parser parse;
+    node_ptr doc = parse.parse("<table><td>1</td></tr> <td>2</td></tr> <td> <table><td>3</td> <td>4</td></table> <tr><td>5</table>");
+    //<table><tbody><tr><td>1</td></tr><tr><td>2</td></tr><tr><td><table><tbody><tr><td>3</td><td>4</td></tr></tbody></table></td></tr><tr><td>5</td></tr></tbody></table>
+    // no tbody
+    ASSERT_EQ("<table>\n\t<td>1</td>\n\t<td>2</td>\n\t<td>\n\t\t<table>\n\t\t\t<td>3</td>\n\t\t\t<td>4</td>\n\t\t</table>\n\t\t<tr>\n\t\t\t<td>5</td>\n\t\t</tr>\n\t</td>\n</table>", doc->to_html());
+}
+
+//test18
+TEST(test, handlesImplicitCaptionClose){
+    parser parse;
+    node_ptr doc = parse.parse("<table><caption>A caption<td>One<td>Two");
+    // <table><caption>A caption</caption><tbody><tr><td>One</td><td>Two</td></tr></tbody></table>
+    ASSERT_EQ("<table>\n\t<caption>A caption\n\t\t<td>One\n\t\t\t<td>Two</td>\n\t\t</td>\n\t</caption>\n</table>", doc->to_html());
+}
+
+//test19
+TEST(test, handlesUnclosedCdataAtEOF){
+    string h = "<![CDATA[]]";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ(1, doc->get_children().size());
+}
+
+//test20
+TEST(test, handlesInvalidStartTags){
+    string h = "<div>Hello < There <&amp;></div>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    // cannot remove amp;
+    ASSERT_EQ("Hello < There <&amp;>", doc->select("div")->get_children()[0]->to_text());
+}
+
+
+GTEST_API_ int main(int argc, char ** argv) {
+    testing::InitGoogleTest(&argc, argv);
+    return RUN_ALL_TESTS();
+}

test/migrate_unittest/test-group3.cpp

@@ -0,0 +1,109 @@
+#include <iostream>
+#include <gtest/gtest.h>
+#include <string>
+#include "html.hpp"
+
+using namespace std;
+using namespace html;
+
+//test21
+TEST(test, handlesUnknownNamespaceTags){
+    string h = "<foo:bar id='1' /><abc:def id=2>Foo<p>Hello</p></abc:def><foo:bar>There</foo:bar>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ("<foo:bar id=\"1\" />\n<abc:def id=\"2\">Foo\n\t<p>Hello</p>\n</abc:def>\n<foo:bar>There</foo:bar>", doc->to_html());
+}
+
+//test22
+TEST(test, handlesKnownEmptyBlocks){
+    string h = "<div id='1' /><script src='/foo' /><div id=2><img /><img></div><a id=3 /><i /><foo /><foo>One</foo> <hr /> hr text <hr> hr text two";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ("<div id=\"1\" />\n<script src=\"/foo\" />\n<div id=\"2\"><img /><img /></div>\n<a id=\"3\" /><i />\n<foo />\n<foo>One</foo>\n<hr />\n hr text \n<hr />\n hr text two", doc->to_html());
+}
+
+//test23
+TEST(test, handlesKnownEmptyNoFrames){
+    string h = "<html><head><noframes /><meta name=foo></head><body>One</body></html>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+// <html>
+//    <head>
+//     <noframes />
+//     <meta name=\"foo\" />
+//    </head>
+//    <body>One</body>
+// </html>
+    ASSERT_EQ("<html>\n\t<head>\n\t\t<noframes />\n\t\t<meta name=\"foo\" />\n\t</head>\n\t<body>One</body>\n</html>", doc->to_html());
+}
+
+//test24
+TEST(test, handlesKnownEmptyStyle){
+    string h = "<html><head><style /><meta name=foo></head><body>One</body></html>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ("<html>\n\t<head>\n\t\t<style />\n\t\t<meta name=\"foo\" />\n\t</head>\n\t<body>One</body>\n</html>", doc->to_html());
+}
+
+//test25
+//issue title is not being converted to HTML format
+TEST(test, handlesKnownEmptyTitle){
+    string h = "<html><head><title /><meta name=foo></head><body>One</body></html>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+
+    //<html><head><title></title><meta name=\"foo\"></head><body>One</body></html>
+    ASSERT_EQ("<html>\n\t<head>\n\t\t<title />\n\t\t<meta name=\"foo\" />\n\t</head>\n\t<body>One</body>\n</html>", doc->to_html());
+}
+
+//test26
+//issue iframe is not being converted to HTML format
+TEST(test, handlesKnownEmptyIframe){
+    string h = "<p>One</p><iframe id=1 /><p>Two";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+
+    ASSERT_EQ("<p>One</p>\n<iframe id=\"1\" />\n<p>Two</p>", doc->to_html());
+}
+
+//test27
+TEST(test, handlesSolidusAtAttributeEnd){
+    // this test makes sure [<a href=/>link</a>] is parsed as [<a href="/">link</a>], not [<a href="" /><a>link</a>]
+    string h = "<a href=/>link</a>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+
+    ASSERT_EQ("<a href=\"/\">link</a>", doc->to_html());
+}
+
+//test28
+TEST(test, ignoresContentAfterFrameset){
+    string h = "<html><head><title>One</title></head><frameset><frame /><frame /></frameset><table></table></html>";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    // the HTML parser implementation provided only supports parsing a subset of HTML elements and attributes, 
+    // and does not support parsing elements such as the frame tag.
+    // can not remove table
+    ASSERT_EQ("<html>\n\t<head>\n\t\t<title>One</title>\n\t</head>\n\t<frameset>\n\t\t<frame />\n\t\t<frame />\n\t</frameset>\n\t<table></table>\n</html>", doc->to_html());
+}
+
+//test29
+TEST(test, normalisesDocument){
+    string h = "<!doctype html>One<html>Two<head>Three<link></head>Four<body>Five </body>Six </html>Seven ";
+    parser parse;
+    node_ptr doc = parse.parse(h);
+    ASSERT_EQ("<!--doctype html-->One\n<html>Two\n\t<head>Three\n\t\t<link />\n\t</head>\n\tFour\n\t<body>Five </body>\n\tSix \n</html>\nSeven ", doc->to_html());
+}
+
+//test30
+TEST(test, normalisesEmptyDocument){
+    parser parse;
+    node_ptr doc = parse.parse("");
+    ASSERT_EQ("", doc->to_html());
+}
+
+
+GTEST_API_ int main(int argc, char ** argv) {
+    testing::InitGoogleTest(&argc, argv);
+    return RUN_ALL_TESTS();
+}