From 31dc311f2a1b87c66eb0638d886c1d0d8a36bbfa Mon Sep 17 00:00:00 2001
From: kelson42 <kelson42@users.sourceforge.net>
Date: Fri, 16 Apr 2010 13:33:50 +0000
Subject: [PATCH] + improved version of removeAccents(9

---
 src/common/kiwix/indexer.cpp  |  6 +++---
 src/common/kiwix/searcher.cpp |  2 +-
 src/common/unaccent.cpp       | 21 +++++----------------
 src/common/unaccent.h         |  2 +-
 4 files changed, 10 insertions(+), 21 deletions(-)

diff --git a/src/common/kiwix/indexer.cpp b/src/common/kiwix/indexer.cpp
index 33b00f4f4..d2de96b23 100644
--- a/src/common/kiwix/indexer.cpp
+++ b/src/common/kiwix/indexer.cpp
@@ -123,19 +123,19 @@ namespace kiwix {
 	  
 	  /* Index the title */
 	  if (!this->htmlParser.title.empty()) {
-	    indexer.index_text_without_positions(removeAccents(this->htmlParser.title.c_str(), this->htmlParser.title.size()), 
+	    indexer.index_text_without_positions(removeAccents(this->htmlParser.title), 
 						 ((this->htmlParser.dump.size() / 100) + 1) / 
 						 countWords(this->htmlParser.title) );
 	  }
 	  
 	  /* Index the keywords */
 	  if (!this->htmlParser.keywords.empty()) {
-	    indexer.index_text_without_positions(removeAccents(this->htmlParser.keywords.c_str(), this->htmlParser.keywords.size()), 3);
+	    indexer.index_text_without_positions(removeAccents(this->htmlParser.keywords), 3);
 	  }
 	  
 	  /* Index the content */
 	  if (!this->htmlParser.dump.empty()) {
-	    indexer.index_text_without_positions(removeAccents(this->htmlParser.dump.c_str(), this->htmlParser.dump.size()));
+	    indexer.index_text_without_positions(removeAccents(this->htmlParser.dump));
 	  }
 	  
 	  /* add to the database */
diff --git a/src/common/kiwix/searcher.cpp b/src/common/kiwix/searcher.cpp
index 817940978..92a2579a1 100644
--- a/src/common/kiwix/searcher.cpp
+++ b/src/common/kiwix/searcher.cpp
@@ -36,7 +36,7 @@ namespace kiwix {
     
     /* Create the query term vector */
     /* I have the doublequote " because bug ID: 2939690 */
-    std::vector<std::string> queryTerms = split(removeAccents(search.c_str(), search.size()), " #@%$0/\\_-*()[]{},;:\"´`'");
+    std::vector<std::string> queryTerms = split(removeAccents(search), " #@%$0/\\_-*()[]{},;:\"´`'");
     
     /* Create query object */
     Xapian::Query query(Xapian::Query::OP_OR, queryTerms.begin(), queryTerms.end());
diff --git a/src/common/unaccent.cpp b/src/common/unaccent.cpp
index 0dd27057c..b025ef361 100644
--- a/src/common/unaccent.cpp
+++ b/src/common/unaccent.cpp
@@ -57,23 +57,12 @@ void UnaccentTransliterator::handleTransliterate(Replaceable& text,
 
 /* Remove accents from a String */
 UnaccentTransliterator unaccent;
-char *unaccentedString = NULL;
-unsigned unaccentedStringSize=0;
 UnicodeString unicodeAccentedString;
 
-const char* removeAccents(const char *accentedString, const unsigned size) {
-
-  /* Realloc memory if necessary */
-  if (size > unaccentedStringSize) {
-    unaccentedString = (char*)realloc(unaccentedString, size+1);
-    unaccentedStringSize = size+1;
-  }
-
-  /* Transcode the String */
-  unicodeAccentedString = UnicodeString(accentedString);
+std::string &removeAccents(std::string &text) {
+  unicodeAccentedString = UnicodeString(text.c_str());
   unaccent.transliterate(unicodeAccentedString);
-  
-  /* Extract and return the result */
-  unicodeAccentedString.extract(0, size, unaccentedString, size, "UTF-8");
-  return unaccentedString;
+  text.clear();
+  unicodeAccentedString.toUTF8String(text);
+  return text;
 }
diff --git a/src/common/unaccent.h b/src/common/unaccent.h
index c9de5309c..2d096717a 100644
--- a/src/common/unaccent.h
+++ b/src/common/unaccent.h
@@ -95,4 +95,4 @@ private:
     static const char fgClassID;
 };
 
-const char* removeAccents(const char *, const unsigned);
+std::string &removeAccents(std::string &text);