debugged the CFF driver. The "advance width" problem

was strictly unrelated to our previous worries, it was a
really simple bug in t2gload.c !!

CHANGES

@@ -1,7 +1,7 @@
 LATEST CHANGES
 
-  - added support for CID-keyed fonts to the CFF driver. There are still
-    some unexplained bugs though... ???
+  - added support for CID-keyed fonts to the CFF driver.
+    maybe support for pure CFF + CEF fonts should come in ??
 
 
   - cleaned up source code in order to avoid two functions with the

src/cff/t2gload.c

@@ -809,14 +809,14 @@
         if ( req_args & T2_COUNT_CHECK_WIDTH )
         {
           args = stack;
-          if ( decoder->read_width )
+          if ( num_args & 1 && decoder->read_width )
           {
             decoder->glyph_width = decoder->nominal_width +
                                      ( stack[0] >> 16 );
-            decoder->read_width  = 0;
             num_args--;
             args++;
           }
+          decoder->read_width  = 0;
           req_args = 0;
         }