def vocalism(self, str1, str2):
"""Check correct word-endings. All words should end with a/i/u, but a handful of exceptions:
- me, to, so (possessive pronouns)
- po (verb "putiri")
- no
- jo (personal pronoun)
- se (yes)
- nne (in)
:param str1: the source string
:param str2: the target (translated) string
:return: True if there are no words with endings not in respect of vocalism (or if they appear in source string as well)
"""
exceptions = ["me", "to", "so", "po", "no", "jo", "se", "nne"]
stopwords = []
for word in self.config.lang.words(str2):
if word not in str1 and word.lower() not in exceptions:
if word.lower().endswith(("e", "o")) and word.lower() not in stopwords:
stopwords.append(word.lower())
if stopwords:
raise FilterFailure("Please respect vocalism: %s" % (", ".join(stopwords)))
return True
def italianisms(self, str1, str2):
"""Check if the translation contains common errors done by italophones.
Mainly inspired by musttranslatewords(), but with a different logic: return True if
the given word appears in the translation but not in the source (if it's in the source,
then presumably it's being kept untranslated).
:param str1: the source string
:param str2: the target (translated) string
:return: True if str2 doesn't contain an "italianism"
"""
str1 = self.removevariables(str1)
str2 = self.removevariables(str2)
errors = {
"io": "ju/jeu/iu/...",
"tantu": "assai",
"menu": "cchiù picca",
}
# The above is full of strange quotes and things in utf-8 encoding.
# single apostrophe perhaps problematic in words like "doesn't"
for separator in self.config.punctuation:
str1 = str1.replace(separator, u" ")
str2 = str2.replace(separator, u" ")
words1 = self.filteraccelerators(str1).split()
words2 = self.filteraccelerators(str2).split()
stopwords = ["%s (%s)" % (word, errors[word]) for word in words2 if word.lower() in errors.keys() and word not in words1]
if stopwords:
raise FilterFailure(u"Please translate: %s" % (u", ".join(stopwords)))
return True
def niciun_nicio(self, str1, str2):
"""
Checks for sequences containing 'nici un'/'nici o' which are obsolete
Romanian syntax. Correct is 'niciun'/'nicio'
"""
if contains_illegal(['nici un', 'nici o'], str2):
raise FilterFailure("String contains 'nici un' or 'nici o'")
return True
def cedillas(self, str1, str2):
"""Check if the translation contains an illegal cedilla character
Cedillas are obsoleted diacritics for Romanian:
- U+0162 Latin capital letter T with cedilla
- U+0163 Latin small letter T with cedilla
- U+015E Latin capital letter S with cedilla
- U+015F Latin small letter S with cedilla
Cedilla-letters are only valid for Turkish (S-cedilla) and Gagauz
languages (S-cedilla and T-comma). Fun fact: Gagauz is the only known
language to use T-cedilla.
:param str1: the source string
:param str2: the target (translated) string
:return: True if str2 contains a cedilla character
"""
if contains_illegal(['Ţ', 'Ş', 'ţ', 'ş'], str2):
raise FilterFailure("String contains illegal cedillas")
return True
def suffixes(self, str1, str2):
"""Check for common word suffixes to be written correctly.
:param str1: the source string
:param str2: the target (translated) string
:return: True if there are no common suffixes wrongly written
"""
suffixes = {
"zzioni": "zziuni",
}
stopwords = []
for word in self.config.lang.words(str2):
for suffix in suffixes.keys():
if word not in str1 and word.lower().endswith(suffix):
stopwords.append("%s (-%s)" % (word, suffixes[suffix]))
if stopwords:
raise FilterFailure(u"Please use the correct word endings: %s" % (u", ".join(stopwords)))
return True