Handle odd PDFDoc codepoints in UTF-8 during transcoding (fixes #650)

There are codepoints in PDFDoc that are not valid UTF-8 but map to
valid UTF-8. We were handling those correctly with bidirectional
mapping.

However, if those same code points appeared in UTF-8, where they have
no meaning, they were left as fixed points when converting to PDFDoc,
where they do have meaning. This change recognizes them as errors.

2 files changed, 12 insertions, 4 deletions

diff --git a/libtests/qtest/qutil/qutil.out b/libtests/qtest/qutil/qutil.out
index aedf49e1..fa284237 100644
--- a/libtests/qtest/qutil/qutil.out
+++ b/libtests/qtest/qutil/qutil.out
@@ -88,7 +88,8 @@ alternatives
 2: 83a9e99e
 0: 717561636b
 done alternatives
-w˘wˇwˆw˙w˝w˛w˚w˜w�w�w
+w˘wˇwˆw˙w˝w˛w˚w˜w�w�w�w
+w?w?w?w?w?w?w?w?w?w?w-w
 done other characters
 ---- whoami
 quack1
diff --git a/libtests/qutil.cc b/libtests/qutil.cc
index 2142346e..2e4d9cdd 100644
--- a/libtests/qutil.cc
+++ b/libtests/qutil.cc
@@ -418,9 +418,16 @@ void transcoding_test()
     print_alternatives(utf8);
     print_alternatives("quack");
     std::cout << "done alternatives" << std::endl;
-    std::string other = QUtil::pdf_doc_to_utf8(
-        "w\030w\031w\032w\033w\034w\035w\036w\037w\177w\255w");
-    std::cout << other << std::endl;
+    // These are characters are either valid in PDFDoc and invalid in
+    // UTF-8 or the other way around.
+    std::string other("w\x18w\x19w\x1aw\x1bw\x1cw\x1dw\x1ew\x1fw\x7fw");
+    std::string other_doc = other + "\x9fw\xadw";
+    std::cout << QUtil::pdf_doc_to_utf8(other_doc) << std::endl;
+    std::string other_utf8 =
+        other + QUtil::toUTF8(0x9f) + "w" + QUtil::toUTF8(0xad) + "w";
+    std::string other_to_utf8;
+    assert(! QUtil::utf8_to_pdf_doc(other_utf8, other_to_utf8));
+    std::cout << other_to_utf8 << std::endl;
     std::cout << "done other characters" << std::endl;
 }