use square brackets with output format '--brackets' (for cleaner escaping and consistency with decoder tree output)

moses/Parameter.cpp

@@ -57,7 +57,7 @@ Parameter::Parameter()
   AddParam("max-trans-opt-per-coverage", "maximum number of translation options per input span (after applying mapping steps)");
   AddParam("max-phrase-length", "maximum phrase length (default 20)");
   AddParam("n-best-list", "file and size of n-best-list to be generated; specify - as the file in order to write to STDOUT");
-  AddParam("n-best-trees", "Write n-best target-side trees (additional to normal n-best-list) to file {n-best-list-file}.trees");
+  AddParam("n-best-trees", "Write n-best target-side trees to n-best-list");
   AddParam("lattice-samples", "generate samples from lattice, in same format as nbest list. Uses the file and size arguments, as in n-best-list");
   AddParam("n-best-factor", "factor to compute the maximum number of contenders (=factor*nbest-size). value 0 means infinity, i.e. no threshold. default is 0");
   AddParam("print-all-derivations", "to print all derivations in search graph");

scripts/training/wrappers/conll2mosesxml.py

@@ -55,6 +55,8 @@ def escape_special_chars(line):
 
     line = line.replace('\'',''') # xml
     line = line.replace('"','"') # xml
+    line = line.replace('[','[') # syntax non-terminal
+    line = line.replace(']',']') # syntax non-terminal
 
     return line
 
@@ -91,11 +93,11 @@ def write(sentence, output_format='xml'):
         out = create_brackets(0,sentence)
 
     out = out.replace('|','|') # factor separator
-    out = out.replace('[','[') # syntax non-terminal
-    out = out.replace(']',']') # syntax non-terminal
 
     out = out.replace(''',''') # lxml is buggy if input is escaped
     out = out.replace('"','"') # lxml is buggy if input is escaped
+    out = out.replace('[','[') # lxml is buggy if input is escaped
+    out = out.replace(']',']') # lxml is buggy if input is escaped
 
     print(out)
 
@@ -138,9 +140,9 @@ def create_subtree(position, sentence):
 def create_brackets(position, sentence):
 
     if position:
-        element = "( " + sentence[position].proj_func + ' '
+        element = "[ " + sentence[position].proj_func + ' '
     else:
-        element = "( sent "
+        element = "[ sent "
 
     for i in range(1,position):
         if sentence[i].proj_head == position:
@@ -148,26 +150,19 @@ def create_brackets(position, sentence):
 
     if position:
         word = sentence[position].word
-        if word == ')':
-            word = 'RBR'
-        elif word == '(':
-            word = 'LBR'
-
         tag = sentence[position].tag
-        if tag == '$(':
-            tag = '$BR'
 
         if preterminals:
-            element += '( ' + tag + ' ' + word + ' ) '
+            element += '[ ' + tag + ' ' + word + ' ] '
         else:
-            element += word + ' ) '
+            element += word + ' ] '
 
     for i in range(position, len(sentence)):
         if i and sentence[i].proj_head == position:
             element += create_brackets(i, sentence)
 
     if preterminals or not position:
-        element += ') '
+        element += '] '
 
     return element