From 8bd8cb23204a8ee665c2c568a34cf8ba74e68e52 Mon Sep 17 00:00:00 2001
From: Nicolas Sceaux <nicolas.sceaux@free.fr>
Date: Fri, 28 Aug 2015 18:47:12 +0200
Subject: [PATCH] =?utf8?q?syllabify=20:=20prise=20en=20compte=20langues=20?=
 =?utf8?q?sans=20espace=20ins=C3=A9cable=20avant=20les=20ponctuations=20do?=
 =?utf8?q?ubles?=
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

---
 scripts/livret.py    | 37 +++++++++++++++++++++++++++----------
 scripts/paroles.py   | 28 ++++++++++++++++++++++------
 scripts/syllabify.py |  4 +++-
 3 files changed, 52 insertions(+), 17 deletions(-)

diff --git a/scripts/livret.py b/scripts/livret.py
index ce84cfff..cda442cf 100644
--- a/scripts/livret.py
+++ b/scripts/livret.py
@@ -65,8 +65,9 @@ class LilyLine():
         return self._text
 
 class Lilybretto():
-    def __init__(self):
+    def __init__(self, language):
         self._lines = []
+        self.language = language
 
     def add_line(self, line):
         self._lines.append(line)
@@ -74,17 +75,20 @@ class Lilybretto():
     def get_lines(self):
         return self._lines
 
-    def syllabify(self,
-                  sign_tokenizer = SignTokenizer(),
-                  syllable_tokenizer = SyllableTokenizerWithWordSeparation()):
+    def syllabify(self):
+        sign_tokenizer = SignTokenizer(language = self.language)
+        syllable_tokenizer = SyllableTokenizerWithWordSeparation()
         for line in self._lines:
             line.syllabify(sign_tokenizer, syllable_tokenizer)
 
 
 class RawLibrettoReader():
-    def read(self, filename):
-        file = open(filename, 'r')
-        libretto = Lilybretto()
+    def __init__(self, language="fr"):
+        self.language = language
+
+    def read(self, file):
+        #file = open(filename, 'r')
+        libretto = Lilybretto(self.language)
         verse_parts = []
         for line in file:
             verse_match = re.match(r"^%#(\S*) (.*)$", line)
@@ -119,9 +123,22 @@ class RawLibrettoReader():
         return libretto
 
 if __name__ == '__main__':
-    for filename in sys.argv[1:]:
-        reader = RawLibrettoReader()
-        libretto = reader.read(filename)
+    parser = argparse.ArgumentParser(
+        description='LilPond libretto generation.',
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    parser.add_argument(
+        '--language',
+        default='fr',
+        help='verse language (fr, it)')
+    parser.add_argument(
+        'files', metavar='FILE',
+        type=argparse.FileType('r'),
+        nargs='+',
+        help='input files')
+    args = vars(parser.parse_args())
+    for file in args['files']:
+        reader = RawLibrettoReader(args['language'])
+        libretto = reader.read(file)
         libretto.syllabify()
         for line in libretto.get_lines():
             print(line.get_lily_text())
diff --git a/scripts/paroles.py b/scripts/paroles.py
index b916217e..b295bad8 100644
--- a/scripts/paroles.py
+++ b/scripts/paroles.py
@@ -1,9 +1,12 @@
 from syllabify import *
 
 class Lyricsifier():
-    def read_and_write(self, filename):
-        file = open(filename, 'r')
-        sign_tokenizer = SignTokenizer()
+    def __init__(self, language='fr'):
+        self.language = language
+
+    def read_and_write(self, file):
+        #file = open(filename, 'r')
+        sign_tokenizer = SignTokenizer(language=self.language)
         syllable_tokenizer = SyllableTokenizerWithWordSeparation()
         for line in file:
             verse_match = re.match(r"^%#(\S*) (.*)$", line)
@@ -17,7 +20,20 @@ class Lyricsifier():
                 print("")
 
 if __name__ == '__main__':
-    for filename in sys.argv[1:]:
-        transformer = Lyricsifier()
-        transformer.read_and_write(filename)
+    parser = argparse.ArgumentParser(
+        description='LilPond lyrics generation.',
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    parser.add_argument(
+        '--language',
+        default='fr',
+        help='verse language (fr, it)')
+    parser.add_argument(
+        'files', metavar='FILE',
+        type=argparse.FileType('r'),
+        nargs='+',
+        help='input files')
+    args = vars(parser.parse_args())
+    for file in args['files']:
+        transformer = Lyricsifier(args['language'])
+        transformer.read_and_write(file)
 
diff --git a/scripts/syllabify.py b/scripts/syllabify.py
index 74977374..28181d07 100644
--- a/scripts/syllabify.py
+++ b/scripts/syllabify.py
@@ -80,6 +80,7 @@ class SignTokenizer():
        Can be overriden with ignored_markers constructor keyword
     """
     def __init__(self,
+                 language = "fr",
                  word_separators = " -",
                  word_separator_markers = "°",
                  simple_punctuations = ".,",
@@ -96,6 +97,7 @@ class SignTokenizer():
                                             word_separator_markers))
         self.simple_punctuations = simple_punctuations
         self.double_punctuations = double_punctuations
+        self.space_before_double_punctuations = (language == "fr")
         self.apostrophes = apostrophes
         self.forced_syllable_end_marker = forced_syllable_end_marker
         self.mute_character_marker = mute_character_marker
@@ -160,7 +162,7 @@ class SignTokenizer():
             # punctuation
             elif punctuation_match:
                 punct = punctuation_match.group(1)
-                if punct in self.double_punctuations:
+                if self.space_before_double_punctuations and punct in self.double_punctuations:
                     self._add_text("\u00A0")
                 self._add_text(punct)
                 i += len(punctuation_match.group(0))
-- 
2.11.4.GIT