openlp-core team mailing list archive

[Karan <Benny.Wasty@xxxxxxxxxxxxxx>]

Karan has proposed merging lp:~karan85/openlp/ewimport-fix into lp:openlp.

Requested reviews:
  Tim Bentley (trb143)
  Karan (karan85)

For more details, see:
https://code.launchpad.net/~karan85/openlp/ewimport-fix/+merge/66885

- added conversion of EW Song Tags (Verse, Chorus, Bridge etc)
- improved slide break detection
- fixed bug in rtf stripping code that caused lyrics to be stripped as well

Tested with my 700 song EW database.

Don't be too hard on me - I haven't used Bazaar before and didn't write any Python for 5 years :)
-- 
https://code.launchpad.net/~karan85/openlp/ewimport-fix/+merge/66885
Your team OpenLP Core is subscribed to branch lp:openlp.

=== modified file 'openlp/plugins/songs/lib/ewimport.py'
--- openlp/plugins/songs/lib/ewimport.py	2011-06-12 16:02:52 +0000
+++ openlp/plugins/songs/lib/ewimport.py	2011-07-05 11:16:33 +0000
@@ -31,6 +31,7 @@
 
 import os
 import struct
+import re
 
 from openlp.core.lib import translate
 from openlp.core.ui.wizard import WizardStrings
@@ -38,11 +39,26 @@
 from openlp.plugins.songs.lib import retrieve_windows_encoding
 from songimport import SongImport
 
+RTF_STRIPPING_REGEX = re.compile(r'\{\\tx[^}]*\}')
+# regex: at least two newlines, can have spaces between them
+SLIDE_BREAK_REGEX = re.compile(r'\n *?\n[\n ]*')
+NUMBER_REGEX = re.compile(r'[0-9]+')
+NOTE_REGEX = re.compile(r'\(.*?\)')
+
 def strip_rtf(blob, encoding):
     depth = 0
     control = False
     clear_text = []
     control_word = []
+    
+    # workaround for \tx bug: remove one pair of curly braces 
+    # if \tx is encountered
+    match = RTF_STRIPPING_REGEX.search(blob)
+    if match:
+        # start and end indices of match are curly braces - filter them out
+        blob = ''.join([blob[i] for i in xrange(len(blob)) 
+            if i != match.start() and i !=match.end()])
+    
     for c in blob:
         if control:
             # for delimiters, set control to False
@@ -259,9 +275,45 @@
                 if words:
                     # Format the lyrics
                     words = strip_rtf(words, self.encoding)
-                    for verse in words.split(u'\n\n'):
+                    verse_type = VerseType.Tags[VerseType.Verse]
+                    for verse in SLIDE_BREAK_REGEX.split(words):
+                        verse = verse.strip()
+                        if len(verse) == 0:
+                            continue
+                        verse_split = verse.split(u'\n', 1)
+                        first_line_is_tag = False
+                        # EW tags: verse, chorus, pre-chorus, bridge, tag, 
+                        # intro, ending, slide
+                        for type in VerseType.Names+['tag',  'slide']: 
+                            type = type.lower()
+                            ew_tag = verse_split[0].strip().lower()
+                            if ew_tag.startswith(type):
+                                verse_type = type[0]
+                                if type == 'tag' or type == 'slide':
+                                    verse_type = VerseType.Tags[VerseType.Other]
+                                first_line_is_tag = True
+                                number_found = False
+                                # check if tag is followed by number and/or note
+                                if len(ew_tag) > len(type): 
+                                    match = NUMBER_REGEX.search(ew_tag)
+                                    if match:
+                                        number = match.group()
+                                        verse_type +=number
+                                        number_found = True
+                                    match = NOTE_REGEX.search(ew_tag)
+                                    if match:
+                                        self.comments += ew_tag + u'\n'
+                                if not number_found:
+                                    verse_type += u'1'
+                                break
                         self.add_verse(
-                            verse.strip(), VerseType.Tags[VerseType.Verse])
+                            verse_split[-1].strip() if first_line_is_tag else verse, 
+                            verse_type)
+                if len(self.comments) > 5:
+                    self.comments += unicode(
+                        translate('SongsPlugin.EasyWorshipSongImport',
+                        '\n[above are Song Tags with notes imported from \
+                        EasyWorship]'))
                 if self.stop_import_flag:
                     break
                 if not self.finish():