Add some more tests to catch faulty language definitions

evennia/contrib/rplanguage.py

@@ -110,7 +110,7 @@ _PHONEMES = "ea oh ae aa eh ah ao aw ai er ey ow ia ih iy oy ua uh uw a e i u y
             "s z sh zh ch jh k ng g m n l r w"
 _VOWELS = "eaoiuy"
 # these must be able to be constructed from phonemes (so for example,
-# if you have v here, there must exixt at least one single-character
+# if you have v here, there must exist at least one single-character
 # vowel phoneme defined above)
 _GRAMMAR = "v cv vc cvv vcc vcv cvcc vccv cvccv cvcvcc cvccvcv vccvccvc cvcvccvv cvcvcvcvv"
 
@@ -120,12 +120,12 @@ _RE_WORD = re.compile(r'\w+', _RE_FLAGS)
 _RE_EXTRA_CHARS = re.compile(r'\s+(?=\W)|[,.?;](?=[,.?;]|\s+[,.?;])', _RE_FLAGS)
 
 
-class LanguageExistsError(Exception):
+class LanguageError(RuntimeError):
-    message = "Language is already created. Re-adding it will re-build" \
+    pass
-              " its dictionary map. Use 'force=True' keyword if you are sure."
 
-    def __str__(self):
+
-        return self.message
+class LanguageExistsError(LanguageError):
+    pass
 
 
 class LanguageHandler(DefaultScript):
@@ -229,21 +229,28 @@ class LanguageHandler(DefaultScript):
 
         """
         if key in self.db.language_storage and not force:
-            raise LanguageExistsError
+            raise LanguageExistsError(
-
+                "Language is already created. Re-adding it will re-build"
-        # allowed grammar are grouped by length
+                " its dictionary map. Use 'force=True' keyword if you are sure.")
-        gramdict = defaultdict(list)
-        for gram in grammar.split():
-            gramdict[len(gram)].append(gram)
-        grammar = dict(gramdict)
 
         # create grammar_component->phoneme mapping
         # {"vv": ["ea", "oh", ...], ...}
         grammar2phonemes = defaultdict(list)
         for phoneme in phonemes.split():
+            if re.search("\W", phoneme):
+                raise LanguageError("The phoneme '%s' contains an invalid character" % phoneme)
             gram = "".join(["v" if char in vowels else "c" for char in phoneme])
             grammar2phonemes[gram].append(phoneme)
 
+        # allowed grammar are grouped by length
+        gramdict = defaultdict(list)
+        for gram in grammar.split():
+            if re.search("\W|(!=[cv])", gram):
+                raise LanguageError("The grammar '%s' is invalid (only 'c' and 'v' are allowed)" % gram)
+            gramdict[len(gram)].append(gram)
+        grammar = dict(gramdict)
+
+
         # create automatic translation
         translation = {}
 

evennia/contrib/tests.py

@@ -34,9 +34,8 @@ class TestLanguage(EvenniaTest):
                                 auto_translations=atrans,
                                 force=True)
         rplanguage.add_language(key="binary",
-                                phonemes="oo ii ck w b d t",
+                                phonemes="oo ii a ck w b d t",
                                 grammar="cvvv cvv cvvcv cvvcvv cvvvc cvvvcvv cvvc",
-                                vowels="oei",
                                 noun_prefix='beep-',
                                 word_length_variance=4)
 
@@ -50,13 +49,26 @@ class TestLanguage(EvenniaTest):
         self.assertEqual(result0, text)
         result1 = rplanguage.obfuscate_language(text, level=1.0, language="testlang")
         result2 = rplanguage.obfuscate_language(text, level=1.0, language="testlang")
+        result3 = rplanguage.obfuscate_language(text, level=1.0, language='binary')
+
         self.assertNotEqual(result1, text)
+        self.assertNotEqual(result3, text)
         result1, result2 = result1.split(), result2.split()
         self.assertEqual(result1[:4], result2[:4])
         self.assertEqual(result1[1], "1")
         self.assertEqual(result1[2], "2")
         self.assertEqual(result2[-1], result2[-1])
-        print(rplanguage.obfuscate_language(text, level=1.0, language='binary'))
+
+    def test_faulty_language(self):
+        self.assertRaises(
+            rplanguage.LanguageError,
+            rplanguage.add_language,
+            key='binary2',
+            phonemes="w b d t oe ee, oo e o a wh dw bw",  # erroneous comma
+            grammar="cvvv cvv cvvcv cvvcvvo cvvvc cvvvcvv cvvc c v cc vv ccvvc ccvvccvv ",
+            vowels="oea",
+            word_length_variance=4)
+
 
     def test_available_languages(self):
         self.assertEqual(rplanguage.available_languages(), ["testlang", "binary"])