def count_unique_translations(self, src, lang):
"""Count the translations for a given key in the given language.
Args:
src (str): The text that is translated
lang (str): The language in which it is translated
Returns:
An integer of how many translations are found for the given source
text
"""
# Set the default return value to 0
count_unique = 0
# Clean source text
cleaned_src = utils.td_clean_string(src)
# If the text key is in translations
if cleaned_src in self.data:
# Get all translations for a given key
all_src_data = self.data[cleaned_src]
# If the language is among the translations
if lang in all_src_data:
# Collect all the unique translation strings
translations = all_src_data[lang]
all_found = [other['translation'] for other in translations]
unique_all_found = set(all_found)
# Set the return value to the size of the set
count_unique = len(unique_all_found)
return count_unique
def get_unique_translations(self, src, lang):
"""Return the unique translations for a given text in a given language.
Args:
src (str): The text that is translated
lang (str): The language in which it is translated
Returns:
A list of sorted, unique translations
"""
# Set the default return value to an empty list
unique = []
# Clean source text
cleaned_src = utils.td_clean_string(src)
# If the text key is in translations
if cleaned_src in self.data:
# Get all translations for a given key
all_src_data = self.data[cleaned_src]
# If the language is among the translations
if lang in all_src_data:
# Collect all the unique translation strings
translations = all_src_data[lang]
all_found = [other['translation'] for other in translations]
unique_all_found = set(all_found)
# Set the return value to the size of the set
unique = sorted(list(unique_all_found))
return unique
def add_translation(self, src, other, lang, correct=False):
"""Add a translation to the dictionary.
Many strings to be added come from questionnaires where a numbering
scheme is used. The question number is removed from the text if it is
discovered using the `number_prog` attribute.
The cleaned translation is stored in the `other` dictionary.
Args:
src (str): String in the base language
other (dict): A dictionary containing the CellData namedtuple and
other metadata.
lang (str): String name of other language
correct (bool): Whether or not the input file is treated as correct
"""
cleaned_src = utils.td_clean_string(src)
cleaned_other = utils.td_clean_string(str(other['cell']))
other['translation'] = cleaned_other
if not correct and cleaned_src in self.correct:
# Currently not a correct translation, but we have correct
return
if correct and cleaned_src not in self.correct and cleaned_src in \
self.data:
# Remove the old, non-correct translation
self.data.pop(cleaned_src, None)
if correct:
self.correct.add(cleaned_src)
try:
this_dict = self.data[cleaned_src]
if lang in this_dict:
this_dict[lang].append(other)
else:
this_dict[lang] = [other]
except KeyError:
self.data[cleaned_src] = {lang: [other]}
def get_numbered_translation(self, src, lang):
"""Return a translation for a source string, respecting numbering.
Since many strings come from questionnaires with a numbering scheme,
this method first removes the number, then translates the numberless
text, and finally adds the number back.
Args:
src (str): String in base language for this translator.
lang (str): String name of language for the translation.
Returns:
String in other language that is a translation of `src`. This
string also has the same numbering as `src`.
"""
number, _ = utils.td_split_text(src)
cleaned_src = utils.td_clean_string(src)
clean_translation = self.get_translation(cleaned_src, lang)
numbered_translation = number + clean_translation
return numbered_translation