Add test for search helpers.

- search::cleanup_text()
- search::headline()
- search::page

tests/test_search_helpers.cpp

@@ -0,0 +1,158 @@
+#include "fs-compat.hpp"
+#include "search.hpp"
+
+#include <catch.hpp>
+
+#include <clocale>
+#include <exception>
+#include <string>
+
+SCENARIO("Searching helpers work as intended")
+{
+    GIVEN("Our test zip file")
+    {
+        fs::path zipfile{"test.zip"};
+        std::setlocale(LC_CTYPE, ""); // Needed for utf-8 support in libarchive.
+        bool exception{false};
+
+        REQUIRE(fs::exists(zipfile));
+
+        SECTION("cleanup_texts() does what it should do")
+        {
+            std::string text;
+
+            WHEN("Text is a single word")
+            {
+                text = "Moss";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The text is unchanged")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "Moss");
+                }
+            }
+
+            WHEN("Text has 2 \\r next to each other in it.")
+            {
+                text = "💖\r\r🦝";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The \\r are removed unchanged")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "💖🦝");
+                }
+            }
+
+            WHEN("Text has 6 \\n after another in it")
+            {
+                text = "Moss\n\n\n\n\n\nis good.";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The 6 \\n are condensed to one space")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "Moss is good.");
+                }
+            }
+        }
+
+        SECTION("headline() does what it should do")
+        {
+            std::string text;
+
+            WHEN("We have a text with a h3 headline")
+            {
+                text = "… <h3>Soup</h3> …";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                    text = epubgrep::search::headline(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The headline is correctly extracted")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "Soup");
+                }
+            }
+        }
+
+        SECTION("page() does what it should do")
+        {
+            std::string text;
+
+            WHEN("We have text with epub:type pagebreak number 69")
+            {
+                text = R"(… <span epub:type="pagebreak" … title="69"/> …)";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                    text = epubgrep::search::page(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The page number is correctly extracted")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "69");
+                }
+            }
+
+            WHEN("We have text with doc-pagebreak number 69")
+            {
+                text = R"(… <span role="doc-pagebreak" … aria-label="69"/> …)";
+                try
+                {
+                    epubgrep::search::cleanup_text(text);
+                    text = epubgrep::search::page(text);
+                }
+                catch (const std::exception &)
+                {
+                    exception = true;
+                }
+
+                THEN("No exception is thrown")
+                AND_THEN("The page number is correctly extracted")
+                {
+                    REQUIRE_FALSE(exception);
+                    REQUIRE(text == "69");
+                }
+            }
+        }
+    }
+}