openlp/openlp/plugins/songs/lib/opensongimport.py

255 lines
10 KiB
Python
Raw Normal View History

2010-06-15 20:03:47 +00:00
# -*- coding: utf-8 -*-
# vim: autoindent shiftwidth=4 expandtab textwidth=80 tabstop=4 softtabstop=4
###############################################################################
# OpenLP - Open Source Lyrics Projection #
# --------------------------------------------------------------------------- #
2010-12-26 11:04:47 +00:00
# Copyright (c) 2008-2011 Raoul Snyman #
2011-05-26 16:25:54 +00:00
# Portions copyright (c) 2008-2011 Tim Bentley, Gerald Britton, Jonathan #
# Corwin, Michael Gorven, Scott Guerrieri, Matthias Hub, Meinert Jordan, #
2011-05-26 17:11:22 +00:00
# Armin Köhler, Joshua Miller, Stevan Pettit, Andreas Preikschat, Mattias #
2011-06-12 16:02:52 +00:00
# Põldaru, Christian Richter, Philip Ridout, Simon Scudder, Jeffrey Smith, #
2011-06-12 15:41:01 +00:00
# Maikel Stuivenberg, Martin Thompson, Jon Tibble, Frode Woldsund #
2010-06-15 20:03:47 +00:00
# --------------------------------------------------------------------------- #
# This program is free software; you can redistribute it and/or modify it #
# under the terms of the GNU General Public License as published by the Free #
# Software Foundation; version 2 of the License. #
# #
# This program is distributed in the hope that it will be useful, but WITHOUT #
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or #
# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for #
# more details. #
# #
# You should have received a copy of the GNU General Public License along #
# with this program; if not, write to the Free Software Foundation, Inc., 59 #
# Temple Place, Suite 330, Boston, MA 02111-1307 USA #
###############################################################################
2010-07-19 20:43:02 +00:00
import logging
2011-04-17 15:47:02 +00:00
import re
2010-06-15 21:25:50 +00:00
from lxml import objectify
from lxml.etree import Error, LxmlError
2010-06-15 20:03:47 +00:00
2011-04-25 22:15:38 +00:00
from openlp.plugins.songs.lib import VerseType
2010-07-18 19:27:27 +00:00
from openlp.plugins.songs.lib.songimport import SongImport
2011-04-18 16:46:22 +00:00
from openlp.plugins.songs.lib.ui import SongStrings
2010-06-30 20:05:43 +00:00
log = logging.getLogger(__name__)
2010-08-28 23:09:05 +00:00
class OpenSongImport(SongImport):
2010-06-15 20:03:47 +00:00
"""
2010-08-28 23:09:05 +00:00
Import songs exported from OpenSong
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
The format is described loosly on the `OpenSong File Format Specification
<http://www.opensong.org/d/manual/song_file_format_specification>`_ page on
the OpenSong web site. However, it doesn't describe the <lyrics> section,
so here's an attempt:
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
Verses can be expressed in one of 2 ways, either in complete verses, or by
line grouping, i.e. grouping all line 1's of a verse together, all line 2's
of a verse together, and so on.
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
An example of complete verses::
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
<lyrics>
[v1]
List of words
Another Line
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
[v2]
Some words for the 2nd verse
etc...
</lyrics>
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
The 'v' in the verse specifiers above can be left out, it is implied.
An example of line grouping::
<lyrics>
[V]
1List of words
2Some words for the 2nd Verse
1Another Line
2etc...
</lyrics>
Either or both forms can be used in one song. The number does not
necessarily appear at the start of the line. Additionally, the [v1] labels
can have either upper or lower case Vs.
2010-06-15 20:03:47 +00:00
Other labels can be used also:
2010-08-28 23:09:05 +00:00
C
Chorus
B
Bridge
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
All verses are imported and tagged appropriately.
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
Guitar chords can be provided "above" the lyrics (the line is preceeded by
a period "."), and one or more "_" can be used to signify long-drawn-out
words. Chords and "_" are removed by this importer. For example::
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
. A7 Bm
1 Some____ Words
2010-06-15 20:03:47 +00:00
2010-08-28 23:09:05 +00:00
The <presentation> tag is used to populate the OpenLP verse display order
field. The Author and Copyright tags are also imported to the appropriate
places.
2010-06-15 20:03:47 +00:00
"""
2010-08-28 23:09:05 +00:00
def __init__(self, manager, **kwargs):
2010-06-15 20:03:47 +00:00
"""
2010-08-28 23:09:05 +00:00
Initialise the class.
2010-06-15 20:03:47 +00:00
"""
2011-02-18 17:34:43 +00:00
SongImport.__init__(self, manager, **kwargs)
2011-09-05 12:51:16 +00:00
def doImport(self):
self.importWizard.progressBar.setMaximum(len(self.importSource))
for filename in self.importSource:
if self.stopImportFlag:
return
song_file = open(filename)
2011-09-05 12:51:16 +00:00
self.doImportFile(song_file)
song_file.close()
2010-07-07 20:16:14 +00:00
2011-09-05 12:51:16 +00:00
def doImportFile(self, file):
2010-06-15 20:03:47 +00:00
"""
2011-04-17 15:47:02 +00:00
Process the OpenSong file - pass in a file-like object, not a file path.
2010-08-28 23:09:05 +00:00
"""
2011-09-05 12:51:16 +00:00
self.setDefaults()
try:
tree = objectify.parse(file)
2010-09-14 14:21:44 +00:00
except (Error, LxmlError):
2011-09-05 12:51:16 +00:00
self.logError(file.name, SongStrings.XMLSyntaxError)
log.exception(u'Error parsing XML')
2011-04-19 11:34:04 +00:00
return
2010-06-28 19:55:04 +00:00
root = tree.getroot()
fields = dir(root)
2010-08-28 23:09:05 +00:00
decode = {
u'copyright': self.add_copyright,
u'ccli': u'ccli_number',
u'author': self.parse_author,
u'title': u'title',
u'aka': u'alternate_title',
u'hymn_number': u'song_number'
}
for attr, fn_or_string in decode.items():
if attr in fields:
ustring = unicode(root.__getattr__(attr))
if isinstance(fn_or_string, basestring):
setattr(self, fn_or_string, ustring)
else:
fn_or_string(ustring)
2010-08-28 23:09:05 +00:00
if u'theme' in fields and unicode(root.theme) not in self.topics:
self.topics.append(unicode(root.theme))
if u'alttheme' in fields and unicode(root.alttheme) not in self.topics:
self.topics.append(unicode(root.alttheme))
2010-06-15 21:25:50 +00:00
# data storage while importing
2010-06-28 19:55:04 +00:00
verses = {}
# keep track of verses appearance order
2010-06-28 19:55:04 +00:00
our_verse_order = []
2011-02-18 07:53:40 +00:00
# default verse
2011-04-17 15:47:02 +00:00
verse_tag = VerseType.Tags[VerseType.Verse]
2011-02-18 07:53:40 +00:00
verse_num = u'1'
# for the case where song has several sections with same marker
inst = 1
2011-04-25 22:15:38 +00:00
if u'lyrics' in fields:
lyrics = unicode(root.lyrics)
else:
lyrics = u''
2011-02-18 07:53:40 +00:00
for this_line in lyrics.split(u'\n'):
2010-06-15 21:25:50 +00:00
# remove comments
2011-02-18 07:53:40 +00:00
semicolon = this_line.find(u';')
2010-06-15 21:25:50 +00:00
if semicolon >= 0:
2011-02-18 07:53:40 +00:00
this_line = this_line[:semicolon]
this_line = this_line.strip()
if not len(this_line):
2010-06-15 21:25:50 +00:00
continue
# skip guitar chords and page and column breaks
2011-02-18 07:53:40 +00:00
if this_line.startswith(u'.') or this_line.startswith(u'---') \
or this_line.startswith(u'-!!'):
2010-06-15 21:25:50 +00:00
continue
# verse/chorus/etc. marker
2011-02-18 07:53:40 +00:00
if this_line.startswith(u'['):
2010-09-02 20:21:31 +00:00
# drop the square brackets
2011-02-18 07:53:40 +00:00
right_bracket = this_line.find(u']')
content = this_line[1:right_bracket].lower()
2010-11-03 18:03:28 +00:00
# have we got any digits?
2011-02-18 07:53:40 +00:00
# If so, verse number is everything from the digits
2010-09-02 20:21:31 +00:00
# to the end (even if there are some alpha chars on the end)
2011-04-25 22:37:58 +00:00
match = re.match(u'(\D*)(\d+.*)', content)
2010-09-02 20:21:31 +00:00
if match is not None:
2011-02-18 07:53:40 +00:00
verse_tag = match.group(1)
verse_num = match.group(2)
2010-06-15 21:25:50 +00:00
else:
2010-11-03 18:03:28 +00:00
# otherwise we assume number 1 and take the whole prefix as
2011-02-18 07:53:40 +00:00
# the verse tag
verse_tag = content
verse_num = u'1'
2011-08-16 08:46:55 +00:00
if len(verse_tag) == 0:
verse_index = 0
else:
verse_index = VerseType.from_loose_input(verse_tag)
2011-04-25 22:15:38 +00:00
verse_tag = VerseType.Tags[verse_index]
inst = 1
2011-02-18 07:53:40 +00:00
if [verse_tag, verse_num, inst] in our_verse_order \
and verses.has_key(verse_tag) \
and verses[verse_tag].has_key(verse_num):
2011-04-17 15:47:02 +00:00
inst = len(verses[verse_tag][verse_num]) + 1
2010-06-15 21:25:50 +00:00
continue
# number at start of line.. it's verse number
2011-02-18 07:53:40 +00:00
if this_line[0].isdigit():
verse_num = this_line[0]
this_line = this_line[1:].strip()
our_verse_order.append([verse_tag, verse_num, inst])
if not verses.has_key(verse_tag):
verses[verse_tag] = {}
if not verses[verse_tag].has_key(verse_num):
verses[verse_tag][verse_num] = {}
if not verses[verse_tag][verse_num].has_key(inst):
verses[verse_tag][verse_num][inst] = []
2011-04-25 22:15:38 +00:00
our_verse_order.append([verse_tag, verse_num, inst])
2011-02-16 09:43:07 +00:00
# Tidy text and remove the ____s from extended words
2011-02-18 07:53:40 +00:00
this_line = self.tidy_text(this_line)
this_line = this_line.replace(u'_', u'')
this_line = this_line.replace(u'|', u'\n')
verses[verse_tag][verse_num][inst].append(this_line)
2010-06-15 21:25:50 +00:00
# done parsing
# add verses in original order
2011-02-18 07:53:40 +00:00
for (verse_tag, verse_num, inst) in our_verse_order:
verse_def = u'%s%s' % (verse_tag, verse_num)
lines = u'\n'.join(verses[verse_tag][verse_num][inst])
self.add_verse(lines, verse_def)
2011-04-25 22:15:38 +00:00
if not self.verses:
self.add_verse('')
# figure out the presentation order, if present
2011-04-18 18:03:41 +00:00
if u'presentation' in fields and root.presentation:
2010-06-28 19:55:04 +00:00
order = unicode(root.presentation)
2011-02-16 19:28:55 +00:00
# We make all the tags in the lyrics lower case, so match that here
2010-09-02 20:21:31 +00:00
# and then split into a list on the whitespace
2011-02-16 19:28:55 +00:00
order = order.lower().split()
2011-02-18 07:53:40 +00:00
for verse_def in order:
2011-04-25 22:37:58 +00:00
match = re.match(u'(\D*)(\d+.*)', verse_def)
if match is not None:
2011-02-18 07:53:40 +00:00
verse_tag = match.group(1)
verse_num = match.group(2)
if not len(verse_tag):
2011-04-18 18:03:41 +00:00
verse_tag = VerseType.Tags[VerseType.Verse]
else:
# Assume it's no.1 if there are no digits
2011-02-18 07:53:40 +00:00
verse_tag = verse_def
verse_num = u'1'
verse_def = u'%s%s' % (verse_tag, verse_num)
2011-04-18 16:46:22 +00:00
if verses.has_key(verse_tag) and \
verses[verse_tag].has_key(verse_num):
2011-02-18 07:53:40 +00:00
self.verse_order_list.append(verse_def)
else:
2011-02-18 07:53:40 +00:00
log.info(u'Got order %s but not in verse tags, dropping'
u'this item from presentation order', verse_def)
2011-04-18 16:46:22 +00:00
if not self.finish():
2011-09-05 12:51:16 +00:00
self.logError(file.name)