managed rfc2047 and rfc2184 for attache name

manage name= / name*= / name*[0-9]= / name*[0-9]*=

src/cpp/Attachment.cpp

@@ -65,8 +65,8 @@ const string Attachment::ALTERNATIVE			("alternative");
 const string Attachment::KAZ_ATTACH_NAME		("vos-pieces-jointes-kaz-ici.html");
 const string Attachment::MULTIPART			("multipart/");
 
-const regex Attachment::nameCharsetRegEx		( ".*name\\*=\\s*([; \t]*)");
-const regex Attachment::nameRegEx			(    ".*name=\\s*((\"(\\\\.|[^\\\\])*\")|[^; \t]*).*");
+const string Attachment::nameLeftToken			(".*name");
+const string Attachment::nameRightToken			("=\\s*((\"(\\\\.|[^\\\\])*\")|[^; \t]*).*");
 const regex Attachment::boundaryRegEx			(".*boundary=\\s*((\"(\\\\.|[^\\\\])*\")|[^; \t]*).*");
 const regex Attachment::cidDefRegEx			(".*<([^>]*)>.*");
 const regex Attachment::textRegEx			(".*text/("+PLAIN+"|"+HTML+").*");
@@ -179,47 +179,31 @@ Attachment::getAttachName () const {
   static string tokens [] = {contentTypeToken, contentDispositionToken};
   DEF_LOG ("Attachment::getAttachName", "");
   for (string token : tokens) {
-    // name=
-    string result = getProp (token, nameRegEx);
-    removeQuote (result);
-    if (result.length ()) {
-      LOG ("name=: " << result);
-      encodedWordDecode (result);
-      return result;
-    }
-    // name*x=
-    for (int id = 0; ; ++id) {
-      string item = getProp (token, regex (".*name\\*"+to_string (id)+"=\\s*((\"(\\\\.|[^\\\\])*\")|[; \t]*).*"));
-      if (item.empty ())
-	break;
-      result += item;
-    }
-    removeQuote (result);
-    if (result.length ()) {
-      LOG ("name*x=: " << result);
-      encodedWordDecode (result);
-      return result;
-    }
-    // name*=
-    result = getProp (token, nameCharsetRegEx);
-    removeQuote (result);
-    if (result.length ()) {
-      LOG ("name*=: " << result);
-      charsetValueDecode (result);
-      return result;
-    }
-    // name*x*=
-    for (int id = 0; ; ++id) {
-      string item = getProp (token, regex (".*name\\*"+to_string (id)+"\\*=\\s*([^; ]*)"));
-      if (item.empty ())
-	break;
-      result += item;
-    }
-    removeQuote (result);
-    if (result.length ()) {
-      LOG ("name*x*=: " << result);
-      encodedWordDecode (result);
-      return result;
+    for (string star : {"", "\\*"}) {
+
+      // name= | name*=
+      regex nameAloneRegEx (nameLeftToken+star+nameRightToken);
+      string result = getProp (token, nameAloneRegEx);
+      removeQuote (result);
+      if (result.length ()) {
+	LOG (("name"+star+=": ") << result);
+	charsetDecode (result);
+	return result;
+      }
+
+      // name*[0-9]= | name*[0-9]*=
+      for (int id = 0; ; ++id) {
+	string item = getProp (token, regex (nameLeftToken+"\\*"+to_string (id)+star+nameRightToken));
+	if (item.empty ())
+	  break;
+	result += item;
+      }
+      removeQuote (result);
+      if (result.length ()) {
+	LOG (("name*x"+star+"=: ") << result);
+	charsetDecode (result);
+	return result;
+      }
     }
   }
   return getUnknown (getContentType ());

src/cpp/kazMisc.cpp

@@ -439,16 +439,17 @@ kaz::encodedWordDecode (string &content) {
 }
 
 // ================================================================================
-void
+bool
 kaz::charsetValueDecode (string &content) {
   // rfc2184
   DEF_LOG ("kazMisc::charsetValueDecode", "content: " << content.substr (0, 100) << "...");
   string::size_type langPos = content.find ("'");
-
-  LOG_BUG (langPos == string::npos, return, "kazMisc::charsetValueDecode bug: no '. (content: " << content.substr (0, 100) << "...)");
+  if (langPos == string::npos)
+    return false;
   string::size_type contentPos = content.find ("'", langPos+1);
+  if (contentPos == string::npos)
+    return false;
 
-  LOG_BUG (contentPos == string::npos, return, "kazMisc::charsetValueDecode bug: no double '. (content: " << content.substr (0, 100) << "...)");
   string tmp (content.substr (contentPos+1));
   quotedDecode<'%'> (tmp);
   LOG ("tmp: " << tmp.substr (0, 100) << "...");
@@ -458,6 +459,17 @@ kaz::charsetValueDecode (string &content) {
     iso2utf (tmp);
   content = tmp;
   LOG ("content: " << content.substr (0, 100) << "...");
+  return true;
+}
+
+// ================================================================================
+void
+kaz::charsetDecode (string &content) {
+  // rfc2047 | rfc2184
+  DEF_LOG ("kazMisc::charsetDecode", "content: " << content.substr (0, 100) << "...");
+  if (charsetValueDecode (content))
+    return;
+  encodedWordDecode (content);
 }
 
 // ================================================================================

src/include/Attachment.hpp

@@ -55,8 +55,9 @@ namespace kaz {
     static vector<string> stringsToUpdate;
     /*! mime tokens */
     static const string contentTypeToken, contentDispositionToken, contentTransferEncodingToken, base64Token, quotedPrintableToken, contentIDToken, PLAIN, HTML, MULTIPART, RELATED, ALTERNATIVE, SIGNED, KAZ_ATTACH_NAME;
+    static const string nameLeftToken, nameRightToken;
     /*! pattern to extract mime values */
-    static const regex nameRegEx, nameCharsetRegEx, boundaryRegEx, cidDefRegEx, textRegEx, multiRegEx;
+    static const regex boundaryRegEx, cidDefRegEx, textRegEx, multiRegEx;
 
     /*! get uniq filename */
     static string getUnknown (const string &ext = "");

src/include/kazMisc.hpp

@@ -36,7 +36,7 @@
 #define _kaz_misc_hpp
 
 #include <string>
-#include <ctype.h>
+#include <inttypes.h>
 #include <map>
 #include <regex>
 
@@ -91,7 +91,9 @@ namespace kaz {
   /*! side effect to get the encoded word according rfc2047 rfc5987 rfc2978 */
   void encodedWordDecode (string &content);
   /*! side effect to get the charsetValue according rfc2184 */
-  void charsetValueDecode (string &content);
+  bool charsetValueDecode (string &content);
+  /*! side effect to remove quote */
+  void charsetDecode (string &content);
   /*! side effect to remove quote */
   void removeQuote (string &content);