Check that filenames are valid utf8 strings

Signed-off-by: Lauris Kaplinski <[email protected]>

Author: Lauris Kaplinski

cdoc/CDoc1Writer.cpp

@@ -292,6 +292,7 @@ CDoc1Writer::addFile(const std::string& name, size_t size)
 {
     if(!d)
         return WORKFLOW_ERROR;
+    if (name.empty() || !libcdoc::isValidUtf8(name)) return libcdoc::DATA_FORMAT_ERROR;
 	d->files.push_back({name, size, {}});
 	return libcdoc::OK;
 }

cdoc/CDoc2Writer.cpp

@@ -519,6 +519,7 @@ CDoc2Writer::addFile(const std::string& name, size_t size)
         LOG_ERROR("{}", last_error);
         return libcdoc::WORKFLOW_ERROR;
     }
+    if (name.empty() || !libcdoc::isValidUtf8(name)) return libcdoc::DATA_FORMAT_ERROR;
     if(auto rv = tar->open(name, size); rv < 0) {
         setLastError(tar->getLastErrorStr(rv));
         LOG_ERROR("{}", last_error);

cdoc/Utils.cpp

@@ -82,6 +82,28 @@ timeToISO(double time)
 #endif
 }
 
+bool
+isValidUtf8 (std::string str)
+{
+    const uint8_t *s = (const uint8_t *) str.data();
+    const uint8_t *e = s + str.size();
+    while (s < e) {
+        size_t s_len = e - s;
+        if ((s[0] & 0x80) == 0x0) {
+            s += 1;
+        } else if (((s[0] & 0xe0) == 0xc0) && (s_len >= 2) && ((s[1] & 0xc0) == 0x80)) {
+            s += 2;
+        } else if (((*s & 0xf0) == 0xe0) && (s_len >= 3) && ((s[1] & 0xc0) == 0x80) && ((s[2] & 0xc0) == 0x80)) {
+            s += 3;
+        } else if (((*s & 0xf8) == 0xf0) && (s_len >= 4) && ((s[1] & 0xc0) == 0x80) && ((s[2] & 0xc0) == 0x80) && ((s[3] & 0xc0) == 0x80)) {
+            s += 4;
+        } else {
+            return false;
+        }
+    }
+	return true;
+}
+
 int
 parseURL(const std::string& url, std::string& host, int& port, std::string& path, bool end_with_slash)
 {

cdoc/Utils.h

@@ -86,6 +86,8 @@ double getTime();
 double timeFromISO(std::string_view iso);
 std::string timeToISO(double time);
 
+bool isValidUtf8 (std::string str);
+
 static std::vector<uint8_t>
 readAllBytes(std::istream& ifs)
 {