Added a configuration parameter to set specific unaccenting/lowercasing for some characters to be handled differently than would result from using the Unicode database. Exemple: "a with ring above" could be set to be preserved by a Swedish locutor

src/common/rclinit.cpp

@@ -107,9 +107,15 @@ RclConfig *recollinit(RclInitFlags flags,
     // Make sure the locale charset is initialized (so that multiple
     // threads don't try to do it at once).
     config->getDefCharset();
+
     // Init unac locking
     unac_init_mt();
 
+    // Init Unac translation exceptions
+    string unacex;
+    if (config->getConfParam("unac_except_trans", unacex) && !unacex.empty()) 
+	unac_set_except_translations(unacex.c_str());
+
     int flushmb;
     if (config->getConfParam("idxflushmb", &flushmb) && flushmb > 0) {
 	LOGDEB1(("rclinit: idxflushmb=%d, set XAPIAN_FLUSH_THRESHOLD to 10E6\n",

src/common/unacpp.cpp

@@ -88,6 +88,7 @@ using namespace std;
 
 #include "unacpp.h"
 #include "readfile.h"
+#include "rclinit.h"
 
 int main(int argc, char **argv)
 {
@@ -98,8 +99,13 @@ int main(int argc, char **argv)
     }
     const char *encoding = argv[1];
     string ifn = argv[2];
+    if (!ifn.compare("stdin"))
+	ifn.clear();
     const char *ofn = argv[3];
 
+    string reason;
+    (void)recollinit(RCLINIT_NONE, 0, 0, reason, 0);
+
     string odata;
     if (!file_to_string(ifn, odata)) {
 	cerr << "file_to_string: " << odata << endl;
@@ -111,7 +117,12 @@ int main(int argc, char **argv)
 	exit(1);
     }
     
-    int fd = open(ofn, O_CREAT|O_EXCL|O_WRONLY, 0666);
+    int fd;
+    if (strcmp(ofn, "stdout")) {
+	fd = open(ofn, O_CREAT|O_EXCL|O_WRONLY, 0666);
+    } else {
+	fd = 1;
+    }
     if (fd < 0) {
 	cerr << "Open/Create " << ofn << " failed: " << strerror(errno) 
 	     << endl;