#
#
# patch "commands.cc"
#  from [83c0ca592e23b9cf6a61fb553f2abf10bd80abc4]
#    to [03612a96ebcb165144d602f09a96344771522963]
#
============================================================
--- commands.cc	83c0ca592e23b9cf6a61fb553f2abf10bd80abc4
+++ commands.cc	03612a96ebcb165144d602f09a96344771522963
@@ -310,46 +310,48 @@ namespace commands
 
     const size_t maxcol = terminal_width();
     vector<string>::const_iterator i = words.begin();
-    while (i != words.end()) {
+    while (i != words.end())
+      {
         string const & word = *i;
 
-        if (col + word.length() + 1 >= maxcol) {
-          out << std::endl;
-          col = 0;
+        if (col + word.length() + 1 >= maxcol)
+          {
+            out << std::endl;
+            col = 0;
 
-          // Skip empty words at the beginning of the line so that they do
-          // not mess with indentation.  These "words" appear because we
-          // put two spaces between sentences, and one of these is
-          // transformed into a word.
-          //
-          // Another approach could be to simply omit these words (by
-          // modifying split_into_words) and then add this formatting (two
-          // spaces between sentences) from this algorithm.  But then,
-          // other kinds of formatting could not be allowed in the original
-          // strings... (i.e., they'd disappear after we mangled them).
-          if (word == "")
-            {
-              do
-                i++;
-              while (i != words.end() && (*i) == "");
-              if (i == words.end())
-                break;
-              else
-                {
-                  while (col++ < colabstract - 1)
-                    out << ' ';
-                  continue;
-                }
-            }
+            // Skip empty words at the beginning of the line so that they do
+            // not mess with indentation.  These "words" appear because we
+            // put two spaces between sentences, and one of these is
+            // transformed into a word.
+            //
+            // Another approach could be to simply omit these words (by
+            // modifying split_into_words) and then add this formatting (two
+            // spaces between sentences) from this algorithm.  But then,
+            // other kinds of formatting could not be allowed in the original
+            // strings... (i.e., they'd disappear after we mangled them).
+            if (word == "")
+              {
+                do
+                  i++;
+                while (i != words.end() && (*i) == "");
+                if (i == words.end())
+                  break;
+                else
+                  {
+                    while (col++ < colabstract - 1)
+                      out << ' ';
+                    continue;
+                  }
+              }
 
-          while (col++ < colabstract - 1)
-            out << ' ';
-        }
+            while (col++ < colabstract - 1)
+              out << ' ';
+          }
 
         out << ' ' << word;
         col += word.length() + 1;
         i++;
-    }
+      }
     out << std::endl;
   }