MusicXML: Fix small issues detected with Recordare sample MusicXML files

-) Open the .ly output files using utf-8 encoding so that special characters
   like Umlaute, ©, » «, etc. can also be written out. Otherwise, python
	 will simply exit the script.
-) Fix line break issues with sys.stderr.write (line break explicitly needed
   in contrast to print)
-) Fix extraction of creator from the XML. In particular, don't look only at
   the very first creator tag, but loop through all of them. Now, a score
	 can have a composer, a lyricist, etc.
-) Extract lyrics syllables only for notes, but not for rests. In lilypond
   a rest can have lyrics attached, in lilypond, the lyrics are assigned to
	 notes only. All lyrics assigned to rests in MusicXML will be lost!

Signed-off-by: Reinhold Kainhofer <reinhold@kainhofer.com>

diff --git a/python/musicxml.py b/python/musicxml.py
index 5b335baafbef1873263d234a3e19d19fa0ddabbb..3ae5954e33fcbb08217fb42afe438527e3797709 100644 (file)
--- a/python/musicxml.py
+++ b/python/musicxml.py
@@ -6,7 +6,7 @@ import sys
 
 def escape_ly_output_string (input_string):
     return_string = input_string
-    needs_quotes = not re.match ("^[a-zA-ZäöüÜÄÖßñ]*$", return_string);
+    needs_quotes = not re.match (u"^[a-zA-ZäöüÜÄÖßñ]*$", return_string);
     if needs_quotes:
         return_string = "\"" + string.replace (return_string, "\"", "\\\"") + "\""
     return return_string
@@ -42,11 +42,11 @@ class Xml_node:
        return ''.join ([c.get_text () for c in self._children])
 
     def message (self, msg):
-        sys.stderr.write (msg)
+        sys.stderr.write (msg+'\n')
 
         p = self
         while p:
-            sys.stderr.write ('  In: <%s %s>' % (p._name, ' '.join (['%s=%s' % item for item in p._attribute_dict.items()])))
+            sys.stderr.write ('  In: <%s %s>\n' % (p._name, ' '.join (['%s=%s' % item for item in p._attribute_dict.items()])))
             p = p.get_parent ()
         
     def get_typed_children (self, klass):
@@ -82,7 +82,7 @@ class Xml_node:
     def get_unique_typed_child (self, klass):
        cn = self.get_typed_children(klass)
        if len (cn) <> 1:
-           sys.stderr.write (self.__dict__)
+           sys.stderr.write (self.__dict__ + '\n')
            raise 'Child is not unique for', (klass, 'found', cn)
 
        return cn[0]
@@ -122,8 +122,7 @@ class Identification (Xml_node):
         for i in creators:
             if hasattr (i, 'type') and i.type == type:
                 return i.get_text ()
-            else:
-                return ''
+        return None
 
     def get_composer (self):
         c = self.get_creator ('composer')
@@ -134,13 +133,17 @@ class Identification (Xml_node):
         for i in creators:
             if not hasattr (i, 'type'):
                 return i.get_text ()
-        return c
+        return None
     def get_arranger (self):
         return self.get_creator ('arranger')
     def get_editor (self):
         return self.get_creator ('editor')
     def get_poet (self):
-        return self.get_creator ('poet')
+        v = self.get_creator ('lyricist')
+        if v:
+            return v
+        v = self.get_creator ('poet')
+        return v
     
     def get_encoding_information (self, type):
         enc = self.get_named_children ('encoding')
@@ -149,7 +152,7 @@ class Identification (Xml_node):
             if children:
                 return children[0].get_text ()
         else:
-            return ''
+            return None
       
     def get_encoding_software (self):
         return self.get_encoding_information ('software')
@@ -234,7 +237,7 @@ class Attributes (Measure_element):
             else:
                 return (4, 4)
         except KeyError:
-            sys.stderr.write ('error: requested time signature, but time sig unknown')
+            sys.stderr.write ('error: requested time signature, but time sig unknown\n')
             return (4, 4)
 
     # returns clef information in the form ("cleftype", position, octave-shift)
@@ -323,7 +326,7 @@ class Part_list (Music_xml_node):
         if instrument_name:
             return instrument_name
         else:
-            sys.stderr.write ("Opps, couldn't find instrument for ID=%s" % id)
+            sys.stderr.write ("Opps, couldn't find instrument for ID=%s\n" % id)
             return "Grand Piano"
         
 class Measure (Music_xml_node):

diff --git a/scripts/musicxml2ly.py b/scripts/musicxml2ly.py
index 11527b90c11f4fb53809116c13e25314c44e1817..a75eff825b6366d59d12d385c268fac65093768e 100644 (file)
--- a/scripts/musicxml2ly.py
+++ b/scripts/musicxml2ly.py
@@ -5,6 +5,7 @@ import sys
 import re
 import os
 import string
+import codecs
 from gettext import gettext as _
 
 """
@@ -707,19 +708,20 @@ def musicxml_voice_to_lily_voice (voice):
                         ev_chord.append (ev)
 
         # Extract the lyrics
-        note_lyrics_processed = []
-        note_lyrics_elements = n.get_typed_children (musicxml.Lyric)
-        for l in note_lyrics_elements:
-            if l.get_number () < 0:
-                for k in lyrics.keys ():
-                    lyrics[k].append (l.lyric_to_text ())
-                    note_lyrics_processed.append (k)
-            else:
-                lyrics[l.number].append(l.lyric_to_text ())
-                note_lyrics_processed.append (l.number)
-        for lnr in lyrics.keys ():
-            if not lnr in note_lyrics_processed:
-                lyrics[lnr].append ("\skip4")
+        if not rest:
+            note_lyrics_processed = []
+            note_lyrics_elements = n.get_typed_children (musicxml.Lyric)
+            for l in note_lyrics_elements:
+                if l.get_number () < 0:
+                    for k in lyrics.keys ():
+                        lyrics[k].append (l.lyric_to_text ())
+                        note_lyrics_processed.append (k)
+                else:
+                    lyrics[l.number].append(l.lyric_to_text ())
+                    note_lyrics_processed.append (l.number)
+            for lnr in lyrics.keys ():
+                if not lnr in note_lyrics_processed:
+                    lyrics[lnr].append ("\skip4")
 
 
         mxl_beams = [b for b in n.get_named_children ('beam')
@@ -1016,7 +1018,7 @@ def convert (filename, options):
 
     printer = musicexp.Output_printer()
     progress ("Output to `%s'" % defs_ly_name)
-    printer.set_file (open (defs_ly_name, 'w'))
+    printer.set_file (codecs.open (defs_ly_name, 'wb', encoding='utf-8'))
 
     print_ly_preamble (printer, filename)
     score_information.print_ly (printer)
@@ -1027,7 +1029,7 @@ def convert (filename, options):
     
     progress ("Output to `%s'" % driver_ly_name)
     printer = musicexp.Output_printer()
-    printer.set_file (open (driver_ly_name, 'w'))
+    printer.set_file (codecs.open (driver_ly_name, 'wb', encoding='utf-8'))
     print_ly_preamble (printer, filename)
     printer.dump (r'\include "%s"' % os.path.basename (defs_ly_name))
     print_score_setup (printer, part_list, voices)