mergeAlignment sometime fails...

[ksudoh]

Hi,

I've just started to use SyMGIZA++, and found a problem in mergeAlignment.cpp.
It sometimes fails to retrieve a few sentences in the end of dataset but correctly generated in .A3.final.part.
Here's my patch to fix this problem; it seems to work fine in my environments.

Best,

--- src/mergeAlignment.cpp.org  2015-10-23 16:09:57.000000000 +0900
+++ src/mergeAlignment.cpp  2015-10-23 16:10:33.000000000 +0900
@@ -121,7 +121,9 @@
 int last[number];
 for(int i=0;i<number;i++) last[i]=0;

-while(added<all){
+bool found_sent = true;
+while(found_sent){
+   found_sent = false;
    for(int d=0;d<number;d++)
     for(int i=last[d]; i<sent[d].size();i++){
        if (sent[d][i] != "" ) if(strContains(sent[d][i], prefixLine)){
@@ -143,6 +145,7 @@
                sent[d][i] = "";
                sent[d][i+1] = "";
                sent[d][i+2] = "";
+           found_sent = true;
            }
            if(iii > (added+1)) {
                if (i>6) last[d] = i-6;

UPDATED: another problem was found in endSymetrize.cpp, which wrongly removed all words with surface "NULL".

--- src/endSymetrize.cpp.org    2015-10-26 13:10:13.000000000 +0900
+++ src/endSymetrize.cpp    2015-10-23 18:53:47.000000000 +0900
@@ -34,6 +34,7 @@
{
    tInput.replace( uPos, uFindLen, tReplace );
    uPos += uReplaceLen;
+       if (tFind == "NULL") break;
}
return tInput;

@@ -79,11 +80,16 @@
 //int a_size=0;
 //int b_size=0;

+unsigned int state = 0;
+unsigned int numline = 0;
+
 while (inp) {
 getline(inp,s1);
 getline(inpInv,s2);
+if (++state == 3) { state = 0; }

-if(s1.find("NULL")!= string::npos){
+//if(s1.find("NULL")!= string::npos){
+if (state == 0) {
 const char *str1 = s1.c_str();
 //a_size=0;
 a.clear();
@@ -248,6 +254,7 @@
            bsp++;
        }

+       numline++;
        if((a.size()+1) == asp && (b.size()+1) == bsp ){

        out << "1" << endl;
@@ -259,7 +266,7 @@
        for(int i=0; i<a.size();i++) out    << " " << a[i];
        out << endl;
        } else{
-           cout << "Error in sentence "<< endl;
+           cout << "Error in sentence " << numline << " source_length:" << a.size() << " target_length:" << b.size() << endl;
        }
 }
 }