constants in a new file

kariminf · Jul 23, 2019 · 06d8fb1 · 06d8fb1
1 parent 781e604
commit 06d8fb1
Show file tree

Hide file tree

Showing 4 changed files with 62 additions and 54 deletions.
diff --git a/aruudy/lists/change.py b/aruudy/lists/change.py
@@ -31,13 +31,16 @@
     u"ذلكم": u"ذَالِكُم",
     u"الله": u"أَللَاه",
     u"إله": u"إِلَاه",
-    u"لإله": u"لإِلَاه",
+    u"الإله": u"الإِلَاه",
 }
 
-
 def modify(word):
-    res = word
-    nodiac = re.sub(ur'[%s]' % const.DIAC, "", word)
-    if nodiac in CHANGE_LST:
-        res = CHANGE_LST[nodiac]
-    return res
+    m = re.match(u"((?:%s)?)(.*)([%s]?)" % (const.SPREP, const.DIAC), word)
+    begining = m.group(1)
+    nodiac = re.sub(u"[%s]" % const.DIAC , "", m.group(2))
+    ending = m.group(3)
+    res = CHANGE_LST.get(nodiac, "")
+    if res:
+        return begining + res + ending
+
+    return word
diff --git a/aruudy/lists/const.py b/aruudy/lists/const.py
@@ -18,7 +18,32 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-TASHKIIL = [u'ِ', u'ُ', u'َ', u'ْ']
-TANWIIN = [u'ٍ', u'ٌ', u'ً']
+TASHKIIL = [u"ِ", u"ُ", u"َ", u"ْ"]
+TANWIIN = [u"ٍ", u"ٌ", u"ً"]
+SHADDA = u"ّ"
+# unknown diacretic (haraka)
+UHARAKA = u"\u0653"
 
-DIAC = u'ّ'.join(TASHKIIL).join(TANWIIN)
+# Add shdda and madda
+# Madda, in our case, is used to indicate an unknown haraka
+DIAC = UHARAKA.join(SHADDA).join(TASHKIIL).join(TANWIIN)
+
+# sun letters in arabic
+SUN = u"[تثدذرزسشصضطظلن]"
+
+# Sticky prepositions (bi-, li-)kasra? or (ka-, fa-, wa-)fatha?
+# kasra and fatha can be madda in case there is no tashkiil
+SPREP = "[\u0644\u0628][\u0650%s]?|[\u0643\u0641\u0648][\u064E%s]?" % (UHARAKA, UHARAKA)
+
+# alif in the middle of sentence
+# DORJ = spaces or (bi-, li-)kasra? or (ka-, fa-, wa-)fatha?
+DORJ = u"[^\\s]\\s+|%s" % SPREP
+
+# ahruf al3illa: alif, alif maqsura, waw, yaa
+ILLA = u"[اىوي]"
+
+TATWEEL = {
+    u"\u064E": u"\u064E\u0627",
+    u"\u064F": u"\u064F\u0648",
+    u"\u0650": u"\u0650\u064A",
+}
diff --git a/aruudy/poetry/prosody.py b/aruudy/poetry/prosody.py
@@ -30,17 +30,7 @@
 
 import re
 from aruudy.poetry import meter, foot
-from aruudy.lists import change
-
-# sun letters in arabic
-SUN = u"([تثدذرزسشصضطظلن])"
-# alif in the middle of sentence
-# DORJ = spaces or (bi-, li-)kasra? or (ka-, fa-, wa-)fatha?
-DORJ = u"([^\\s]\\s+|[\u0644\u0628][\u0650]?|[\u0643\u0641\u0648][\u064E]?)"
-
-# ahruf al3illa: alif, alif maqsura, waw, yaa
-ILLA = u"[اىوي]"
-
+from aruudy.lists import change, const
 
 def normalize(text):
     """Normalize a given text.
@@ -102,10 +92,10 @@ def normalize(text):
         res = re.sub(u"([^\u064B-\u0652\\s])\u064A([^\u064B-\u0652]|$)", u"\\1\u0650\u064A\\2", res)
 
     # add Shadda to shamsi characters after al-
-    res = re.sub(u"(^|\\s)\u0627\u0644" + SUN + u"([^\u0651])", u"\\1\u0627\u0644\\2\u0651\\3", res)
+    res = re.sub(u"(^|\\s)\u0627(\u064E?)\u0644(%s)([^\u0651])" % const.SUN, u"\\1\u0627\\2\u0644\\3\u0651\\4", res)
 
     # add madda to other characters after al-
-    res = re.sub(u"((?:^|\\s)\u0627\u0644[^\u0651])([^\u064E-\u0651])", u"\\1\u0653\\2", res)
+    res = re.sub(u"((?:^|\\s)\u0627\u0644[^\u0651])([^\u064E-\u0651])", u"\\1%s\\2" % const.UHARAKA, res)
 
     # add kasra to li
     res = re.sub(u"(^|\\s)\u0644([^\u064E-\u0652])", u"\\1\u0644\u0650\\2", res)
@@ -120,43 +110,43 @@ def normalize(text):
     res = re.sub(u"(^|\\s)\u0648([^\u064E-\u0652])", u"\\1\u0648\u064E\\2", res)
 
     # madda over alif with no vocalization
-    res = re.sub(u"\u0623([^\u064B-\u0653\\s])", u"\u0623\u0653\\1", res)
+    res = re.sub(u"\u0623([^\u064B-\u0653\\s])", u"\u0623%s\\1" % const.UHARAKA, res)
 
     # hamza under alif with no kasra
     res = re.sub(u"\u0625([^\u0650])", u"\u0625\u0650\\1", res)
 
     #shadda not followed by a diacritic: add a madda above
-    res = res = re.sub(u"\u0651([^\u064B-\u0650])", u"\u0651\u0653\\1", res)
+    res = res = re.sub(u"\u0651([^\u064B-\u0650])", u"\u0651%s\\1" % const.UHARAKA, res)
 
     #add madda to any leading letter except alif
-    res = res = re.sub(u"(^|\\s)([^\u0627])([^\u064E-\u0653])", u"\\1\\2\u0653\\3", res)
+    res = res = re.sub(u"(^|\\s)([^\u0627])([^\u064E-\u0653])", u"\\1\\2%s\\3" % const.UHARAKA, res)
 
     #after sukuun must be a haraka
-    res = res = re.sub(u"\u0652([^\\s])([^\u064B-\u0650\\s])", u"\u0652\\1\u0653\\2", res)
+    res = res = re.sub(u"\u0652([^\\s])([^\u064B-\u0650\\s])", u"\u0652\\1%s\\2" % const.UHARAKA, res)
 
     return res
 
 # https://ar.wikipedia.org/wiki/عروض
 
 def _prosody_del(text):
     res = text
-
     # Replace al- with sun character (it can be preceded by prepositions bi- li-)
     # والصِّدق، والشَّمس ---> وصصِدق، وَششَمس
-    res = re.sub(DORJ + u"\u0627\u0644" + SUN , u"\\1\\2", res)
+    res = re.sub(u"(%s)\u0627\u0644(%s)" % (const.DORJ, const.SUN) , u"\\1\\2", res)
+    res = re.sub(u"\u0627\u064E\u0644(%s)" % const.SUN , u"\u0627\u064E\\1", res)
 
     # Replace al- with l otherwise
     # # والكتاب، فالعلم ---> وَلكتاب، فَلعِلم
-    res = re.sub(DORJ + u"\u0627(\u0644[^\u064E-\u0650])", u"\\1\\2", res)
+    res = re.sub(u"(%s)\u0627(\u0644[^\u064E-\u0650])" % const.DORJ, u"\\1\\2", res)
 
 
     # delete first alif of a word in the middle of sentence
     # فاستمعَ، وافهم، واستماعٌ، وابنٌ، واثنان ---> فَستَمَعَ، وَفهَم، وَستِماعُن، وَبنُن، وَثنانِ
-    res = re.sub(DORJ + u"\u0627([^\\s][^\u064B-\u0651\u0653])" , u"\\1\\2", res)
+    res = re.sub(u"(%s)\u0627([^\\s][^\u064B-\u0651\u0653])" % const.DORJ , u"\\1\\2", res)
 
     # delete ending alif, waw and yaa preceeding a sakin
     # أتى المظلوم إلى القاضي فأنصفه قاضي العدل ---> أتَ لمظلوم إلَ لقاضي فأنصفه قاضِ لعدل.
-    res = re.sub(ILLA + u"\\s+(.[^\u064B-\u0651\u0653])", u" \\1", res)
+    res = re.sub(const.ILLA + u"\\s+(.[^\u064B-\u0651\u0653])", u" \\1", res)
 
     # delete alif of plural masculin conjugation
     # رجعوا ---> رجعو
@@ -171,12 +161,6 @@ def _prosody_del(text):
 
     return res
 
-tatweel = {
-    u"\u064E": u"\u064E\u0627",
-    u"\u064F": u"\u064F\u0648",
-    u"\u0650": u"\u0650\u064A",
-}
-
 def _prosody_add(text):
     res = text
 
@@ -201,13 +185,13 @@ def _prosody_add(text):
     # hamza mamduuda --> alif-hamza + fatha + alif + sukuun
     res = re.sub(u"\u0622", u"\u0623\u064E\u0627\u0652", res)
 
-    res = re.sub(u"([\u064E-\u0650])$", lambda m: tatweel[m.group(1)], res)
+    res = re.sub(u"([\u064E-\u0650])$", lambda m: const.TATWEEL[m.group(1)], res)
 
     return res
 
 def _prosody_change(text):
     res = text
-    res = re.sub(u"([^\s]+)", lambda m: change.modify(m.group(1)) , res)
+    res = re.sub(u"([^\s]+)", lambda m: change.modify(m.group(1)), res)
     return res
 
 #TODO trait these
@@ -247,9 +231,9 @@ def prosody_form(text):
 
     """
     res = text
-    res = _prosody_del(text)
-    res = _prosody_add(res)
     res = _prosody_change(res)
+    res = _prosody_del(res)
+    res = _prosody_add(res)
     return res
 
 

diff --git a/tests/meter_u.py b/tests/meter_u.py
@@ -26,7 +26,7 @@
 sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '..')))
 
 from aruudy.poetry import meter
-from aruudy.poetry.meter import Bahr, BahrError
+from aruudy.poetry.meter import Bahr
 
 
 def test_name_type():
@@ -41,21 +41,17 @@ def test_get_bahr():
     assert meter.get_bahr("aaa") == None
 
 def test_get_names():
-    assert meter.arabic_names()[0] == u"طويل"
-    assert meter.english_names()[0] == "long"
-    assert meter.trans_names()[0] == u"ṭawīl"
+    assert meter.get_names("arabic")[0] == u"طويل"
+    assert meter.get_names("english")[0] == "long"
+    assert meter.get_names("trans")[0] == u"ṭawīl"
     assert meter.get_names()[0]["trans"] == u"ṭawīl"
 
 def test_bahr():
     b = meter.get_bahr("overtaking", dic=False)
 
-    assert b.test_property("name", u"mutadārik", "trans")
-    assert not b.test_property("name", u"kamil", "trans")
-    with pytest.raises(BahrError):
-        b.test_property("transliterate", u"kamil")
-
-    assert b.get_value("name", "trans") == u"mutadārik"
-    with pytest.raises(BahrError):
-        b.get_value("transliterate")
+    assert b.test_name("trans", u"mutadārik")
+    assert not b.test_name("trans", u"kamil")
+    #with pytest.raises(BahrError):
+    #    b.test_property("transliterate", u"kamil")
 
     assert b.to_dict() == meter.mutadaarik.to_dict()