Handle web links across lineschromium/3045

When a web link has a hyphen at the end of line, we consider it to
be continued to the next line. For example, "http://www.abc.com/my-\r\ntest"
should be extracted as "http://www.abc.com/my-test".

BUG=pdfium:650

Change-Id: I64a93d9c66faf2be0abdaf8cfe8ee496c435d0ca
Reviewed-on: https://pdfium-review.googlesource.com/3092
Commit-Queue: Wei Li <weili@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>

1 files changed, 68 insertions, 0 deletions

diff --git a/fpdfsdk/fpdftext_embeddertest.cpp b/fpdfsdk/fpdftext_embeddertest.cpp
index 198ef8a7f2..3d496bc06f 100644
--- a/fpdfsdk/fpdftext_embeddertest.cpp
+++ b/fpdfsdk/fpdftext_embeddertest.cpp
@@ -370,6 +370,74 @@ TEST_F(FPDFTextEmbeddertest, WebLinks) {
   UnloadPage(page);
 }
 
+TEST_F(FPDFTextEmbeddertest, WebLinksAcrossLines) {
+  EXPECT_TRUE(OpenDocument("weblinks_across_lines.pdf"));
+  FPDF_PAGE page = LoadPage(0);
+  EXPECT_TRUE(page);
+
+  FPDF_TEXTPAGE textpage = FPDFText_LoadPage(page);
+  EXPECT_TRUE(textpage);
+
+  FPDF_PAGELINK pagelink = FPDFLink_LoadWebLinks(textpage);
+  EXPECT_TRUE(pagelink);
+
+  static const char* const kExpectedUrls[] = {
+      "http://example.com?",          // from "http://www.example.com?\r\nfoo"
+      "http://example.com/",          // from "http://www.example.com/\r\nfoo"
+      "http://example.com/test-foo",  // from "http://example.com/test-\r\nfoo"
+      "http://abc.com/test-foo",      // from "http://abc.com/test-\r\n\r\nfoo"
+      // Next two links from "http://www.example.com/\r\nhttp://www.abc.com/"
+      "http://example.com/", "http://www.abc.com",
+  };
+  static const int kNumLinks = static_cast<int>(FX_ArraySize(kExpectedUrls));
+
+  EXPECT_EQ(kNumLinks, FPDFLink_CountWebLinks(pagelink));
+
+  unsigned short fixed_buffer[128];
+  for (int i = 0; i < kNumLinks; i++) {
+    const size_t expected_len = strlen(kExpectedUrls[i]) + 1;
+    memset(fixed_buffer, 0, FX_ArraySize(fixed_buffer));
+    EXPECT_EQ(static_cast<int>(expected_len),
+              FPDFLink_GetURL(pagelink, i, nullptr, 0));
+    EXPECT_EQ(
+        static_cast<int>(expected_len),
+        FPDFLink_GetURL(pagelink, i, fixed_buffer, FX_ArraySize(fixed_buffer)));
+    EXPECT_TRUE(
+        check_unsigned_shorts(kExpectedUrls[i], fixed_buffer, expected_len));
+  }
+
+  FPDFLink_CloseWebLinks(pagelink);
+  FPDFText_ClosePage(textpage);
+  UnloadPage(page);
+}
+
+TEST_F(FPDFTextEmbeddertest, WebLinksAcrossLinesBug) {
+  EXPECT_TRUE(OpenDocument("bug_650.pdf"));
+  FPDF_PAGE page = LoadPage(0);
+  EXPECT_TRUE(page);
+
+  FPDF_TEXTPAGE textpage = FPDFText_LoadPage(page);
+  EXPECT_TRUE(textpage);
+
+  FPDF_PAGELINK pagelink = FPDFLink_LoadWebLinks(textpage);
+  EXPECT_TRUE(pagelink);
+
+  EXPECT_EQ(2, FPDFLink_CountWebLinks(pagelink));
+  unsigned short fixed_buffer[128] = {0};
+  static const char kExpectedUrl[] =
+      "http://tutorial45.com/learn-autocad-basics-day-166/";
+  static const int kUrlSize = static_cast<int>(sizeof(kExpectedUrl));
+
+  EXPECT_EQ(kUrlSize, FPDFLink_GetURL(pagelink, 1, nullptr, 0));
+  EXPECT_EQ(kUrlSize, FPDFLink_GetURL(pagelink, 1, fixed_buffer,
+                                      FX_ArraySize(fixed_buffer)));
+  EXPECT_TRUE(check_unsigned_shorts(kExpectedUrl, fixed_buffer, kUrlSize));
+
+  FPDFLink_CloseWebLinks(pagelink);
+  FPDFText_ClosePage(textpage);
+  UnloadPage(page);
+}
+
 TEST_F(FPDFTextEmbeddertest, GetFontSize) {
   EXPECT_TRUE(OpenDocument("hello_world.pdf"));
   FPDF_PAGE page = LoadPage(0);