From 67b65dffa838bb657067f3a40ff6b7f8a600fe1b Mon Sep 17 00:00:00 2001
From: Matthew Olsson <matthewcolsson@gmail.com>
Date: Sun, 23 May 2021 21:27:17 -0700
Subject: [PATCH] LibPDF: Handle string encodings

Strings can be encoded in either UTF16-BE or UTF8. In either case,
there are a few initial bytes which specify the encoding that must
be checked and also removed from the final string.
---
 Userland/Libraries/LibPDF/CMakeLists.txt |  2 +-
 Userland/Libraries/LibPDF/Parser.cpp     | 25 +++++++++++++++++++++---
 2 files changed, 23 insertions(+), 4 deletions(-)

diff --git a/Userland/Libraries/LibPDF/CMakeLists.txt b/Userland/Libraries/LibPDF/CMakeLists.txt
index 6d52363eb6..aa8b86d850 100644
--- a/Userland/Libraries/LibPDF/CMakeLists.txt
+++ b/Userland/Libraries/LibPDF/CMakeLists.txt
@@ -8,4 +8,4 @@ set(SOURCES
     )
 
 serenity_lib(LibPDF pdf)
-target_link_libraries(LibPDF LibC LibCore LibIPC LibGfx)
+target_link_libraries(LibPDF LibC LibCore LibIPC LibGfx LibTextCodec)
diff --git a/Userland/Libraries/LibPDF/Parser.cpp b/Userland/Libraries/LibPDF/Parser.cpp
index f78f5211c0..7aa434c102 100644
--- a/Userland/Libraries/LibPDF/Parser.cpp
+++ b/Userland/Libraries/LibPDF/Parser.cpp
@@ -9,6 +9,7 @@
 #include <LibPDF/Document.h>
 #include <LibPDF/Filter.h>
 #include <LibPDF/Parser.h>
+#include <LibTextCodec/Decoder.h>
 #include <ctype.h>
 #include <math.h>
 
@@ -422,9 +423,27 @@ NonnullRefPtr<StringObject> Parser::parse_string()
 {
     ScopeGuard guard([&] { consume_whitespace(); });
 
-    if (m_reader.matches('('))
-        return make_object<StringObject>(parse_literal_string(), false);
-    return make_object<StringObject>(parse_hex_string(), true);
+    String string;
+    bool is_binary_string;
+
+    if (m_reader.matches('(')) {
+        string = parse_literal_string();
+        is_binary_string = false;
+    } else {
+        string = parse_hex_string();
+        is_binary_string = true;
+    }
+
+    if (string.bytes().starts_with(Array<u8, 2> { 0xfe, 0xff })) {
+        // The string is encoded in UTF16-BE
+        string = TextCodec::decoder_for("utf-16be")->to_utf8(string.substring(2));
+    } else if (string.bytes().starts_with(Array<u8, 3> { 239, 187, 191 })) {
+        // The string is encoded in UTF-8. This is the default anyways, but if these bytes
+        // are explicitly included, we have to trim them
+        string = string.substring(3);
+    }
+
+    return make_object<StringObject>(string, is_binary_string);
 }
 
 String Parser::parse_literal_string()