Merge branch 'master' of gitlab.cl.uni-heidelberg.de:Messerschleifer/Allzweckmesser

allzweckmesser/model.py

@@ -62,7 +62,7 @@ class Syllable:
 
     def __init__(self, syllable: str, span: List[int], idx: int,
                  syllable_length: int, vowel_length: int,
-                 phenomena: dict = dict()):
+                 phenomena: dict = None):
         if len(syllable) != span[1] - span[0]:
             raise ValueError('Syllable length does not match syllable span.')
         else:
@@ -71,7 +71,7 @@ class Syllable:
             self.id = idx
             self.syllable_length = syllable_length
             self.vowel_length = vowel_length
-            self.phenomena = phenomena
+            self.phenomena = phenomena or dict()
 
     @classmethod
     def from_json(cls, json_file):
@@ -397,7 +397,7 @@ class Verse:
         self.text = verse
         self.source = source
         self.readings = readings or list()
-
+    
     @classmethod
     def from_plain_verse(cls, plain_verse):
         verse = cls(plain_verse)
@@ -437,6 +437,7 @@ class Verse:
 
         return json.dumps(self.to_dict())
 
+
     def __str__(self):
         s = 'Verse: {verse}\n{reading_num} Readings:\n{readings}'
         readings_str = '\n'.join(str(r) for r in self.readings)

allzweckmesser/scanner.py

@@ -3,6 +3,7 @@
 import copy
 import re
 from typing import Dict, List, Set, Tuple
+from itertools import product
 
 from .db import FormAnalysis
 from .model import Reading, Syllable, Token, Verse, Phenomenon
@@ -253,32 +254,82 @@ def muta_cum_liquida(verse):
                         if syllable.span[0]<= match.start() < syllable.span[1]:
 
                             syllable.phenomena['muta cum liquida'] = Phenomenon(chars=match.group(0))
+                            
+                            
+def positional_lengthening(verse):
+    
+    pl_regex = re.compile(r'[aeiouv](((([bcdfgjklmnprstvwxz]|(qu)),?\s?){2,})|[xz])')
+    if re.search(pl_regex, verse.text):
+        matches = re.finditer(pl_regex, verse.text)
 
+        for match in matches:
+            for reading in verse.readings:
+                for token in reading.tokens:
+                    for syllable in token.syllables:
 
-def positional_lengthening(verse):
+                        if syllable.span[0]<= match.start() < syllable.span[1]:
+                            syllable.phenomena['positional lengthening'] = Phenomenon(chars=match.group(1))
 
-    muta_cum_liquida(verse)
 
-    pl_regex = re.compile(r'[aeiouv]((([bcdfgjklmnprstvwxz]|(qu),?\s?){2,})|[xz])')
-    match_starts = [match.start() for match in re.finditer(pl_regex, verse.text)]
+def parse_verse(verse):
+    """Annotates syllable lengths based on positional_lengthening and muta cum liquida 
+    
+    """
+    
+    positional_lengthening(verse)
+    
+    muta_cum_liquida(verse)
 
+    new_readings = list()
+    
     for reading in verse.readings:
         syllables = [syllable for token in reading.tokens 
                      for syllable in token.syllables]
-        lengths = str()
+        abstract = str()
+        mcl_count = 0
         for syllable in syllables:
+            
             if 'muta cum liquida' in syllable.phenomena:
-                lengths += 'M'
-            elif syllable.syllable_length == 2:
-                lengths += 'L'
-            elif any([syllable.span[0]<= start < syllable.span[1] 
-                      for start in match_starts]):
-                lengths += 'L'
+                abstract += '{}'
+                mcl_count += 1
+            elif 'positional lengthening' in syllable.phenomena:
+                abstract += '2'
+            elif syllable.syllable_length == 0:
+                abstract += '0'
             elif syllable.syllable_length == 1:
-                lengths += 'S'
-        
-        print(lengths)
+                abstract += '1'
+            elif syllable.syllable_length == 2:
+                abstract += '2'
+
+        if mcl_count > 0:
+            new_abstracts = list()
+            combinations = list(product(['1','2'],repeat=mcl_count))
+            for combi in combinations:
+                new_abstracts.append(abstract.format(*combi))
+            
+            reading_copies = multiply_readings([reading], (mcl_count)*2)
+            for i in range(len(new_abstracts)):
+                blueprint = new_abstracts[i]
+                new_reading = reading_copies[i]
+                
+                syll_id = 0
+                for token in new_reading.tokens:
+                    for s in token.syllables:
+                        if blueprint[syll_id] == "1":
+                            s.syllable_length = 1
+                            if 'positional lengthening' in s.phenomena and 'muta cum liquida' in s.phenomena:
+                                s.phenomena['positional lengthening'].overruled_by = 'muta cum liquida'
+                        elif blueprint[syll_id] == "2":
+                            s.syllable_length = 2
+                        syll_id += 1
                 
+                new_readings.append(copy.deepcopy(new_reading))
+                
+        else:
+            new_readings.append(copy.deepcopy(reading))
+             
+        return new_readings
+            
 
         #for reading in verse.readings:
             #for token in reading.tokens:

tests/test_model.py

@@ -21,10 +21,12 @@ def test_import_json():
 
     assert verse_list_from_model == verse_list_from_json
 
-    azm.scanner.muta_cum_liquida(verse_models[0])
+    for verse in verse_models:
+        parsed_verses = azm.scanner.parse_verse(verse)
     
-    for verse_model in verse_models[:1]:
-        print(json.dumps(json.loads(verse_model.to_json()), indent=4, sort_keys=True))
+    for verse_model in parsed_verses:
+        #print(json.dumps(json.loads(verse_model.to_json()), indent=4, sort_keys=True))
+        pass
 
 if __name__ == "__main__":
     test_import_json()