def test_creates_folder_with_contents(self):
exercises = sorted(self.khan_exercises)[:5] # use only first five for fast testing
exercise_path, retrieved_exercises = retrieve_html_exercises(exercises, lang="es")
assert retrieved_exercises # not empty
assert set(retrieved_exercises).issubset(self.khan_exercises)
assert os.path.exists(exercise_path)
def make_language_pack(lang, version, sublangargs, filename, no_assessment_items, no_subtitles, no_assessment_resources):
node_data, subtitle_data, interface_catalog, content_catalog = retrieve_language_resources(version, sublangargs, no_subtitles)
subtitles, subtitle_paths = subtitle_data.keys(), subtitle_data.values()
node_data = translate_nodes(node_data, content_catalog)
node_data = list(node_data)
node_data, dubbed_video_count = apply_dubbed_video_map(node_data, subtitles, sublangargs["video_lang"])
html_exercise_ids, assessment_exercise_ids, node_data = separate_exercise_types(node_data)
html_exercise_path, translated_html_exercise_ids = retrieve_html_exercises(html_exercise_ids, lang)
# now include only the assessment item resources that we need
all_assessment_data, all_assessment_files = retrieve_all_assessment_item_data(
no_item_data=no_assessment_items,
no_item_resources=no_assessment_resources
)
assessment_data = list(translate_assessment_item_text(all_assessment_data, content_catalog)) if lang != "en" else all_assessment_data
node_data = remove_untranslated_exercises(node_data, translated_html_exercise_ids, assessment_data) if lang != "en" else node_data
pack_metadata = generate_kalite_language_pack_metadata(lang, version, interface_catalog, content_catalog, subtitles,
dubbed_video_count)
bundle_language_pack(str(filename), node_data, interface_catalog, interface_catalog,
pack_metadata, assessment_data, all_assessment_files, subtitle_paths, html_exercise_path)
def make_language_pack(lang, version, sublangargs, filename, ka_domain, no_assessment_items, no_subtitles, no_assessment_resources, no_dubbed_videos):
node_data, subtitle_data, interface_catalog, content_catalog = retrieve_language_resources(version, sublangargs, ka_domain, no_subtitles, no_dubbed_videos)
subtitles, subtitle_paths = subtitle_data.keys(), subtitle_data.values()
node_data = translate_nodes(node_data, content_catalog)
node_data = list(node_data)
node_data, dubbed_video_count = apply_dubbed_video_map(node_data, subtitles, sublangargs["video_lang"])
html_exercise_ids, assessment_exercise_ids, node_data = separate_exercise_types(node_data)
html_exercise_path, translated_html_exercise_ids = retrieve_html_exercises(html_exercise_ids, lang)
# now include only the assessment item resources that we need
all_assessment_data, all_assessment_files = retrieve_all_assessment_item_data(
no_item_data=no_assessment_items,
no_item_resources=no_assessment_resources,
node_data=node_data,
lang=lang,
content_catalog=content_catalog,
)
all_assessment_data = list(remove_assessment_data_with_empty_widgets(all_assessment_data))
node_data = remove_nonexistent_assessment_items_from_exercises(node_data, all_assessment_data)
node_data = clean_node_data_items(node_data)
assessment_data = list(all_assessment_data)
node_data = remove_untranslated_exercises(node_data, translated_html_exercise_ids, assessment_data) if lang != "en" else node_data
pack_metadata = generate_kalite_language_pack_metadata(lang, version, sublangargs, interface_catalog, content_catalog, subtitles,
dubbed_video_count)
bundle_language_pack(str(filename), node_data, interface_catalog, interface_catalog,
pack_metadata, assessment_data, all_assessment_files, subtitle_paths, html_exercise_path)
def test_creates_folder_with_contents(self):
exercises = sorted(
self.khan_exercises)[:5] # use only first five for fast testing
exercise_path, retrieved_exercises = retrieve_html_exercises(exercises,
lang="es")
assert retrieved_exercises # not empty
assert set(retrieved_exercises).issubset(self.khan_exercises)
assert os.path.exists(exercise_path)
def test_returns_exercise_with_translation(self):
# espanol has almost complete
# translation. Assuming this specific
# exercise has one
lang = "es"
exercise = self.khan_exercises[0]
path, retrieved_exercises = retrieve_html_exercises([exercise], lang, force=True)
assert retrieved_exercises == [exercise]
def test_doesnt_return_exercises_without_translation(self):
# The expected behaviour from KA's API is that it would return
# the english version of an exercise if either a translated
# exercise for the given language doesn't exist, or the
# language is unsupported.
exercise = self.khan_exercises[0]
lang = "aaa" # there's no language with code aaa... I hope?
path, retrieved_exercises = retrieve_html_exercises([exercise], lang, force=True)
assert not retrieved_exercises
def test_returns_exercise_with_translation(self):
# espanol has almost complete
# translation. Assuming this specific
# exercise has one
lang = "es"
exercise = self.khan_exercises[0]
path, retrieved_exercises = retrieve_html_exercises([exercise],
lang,
force=True)
assert retrieved_exercises == [exercise]
def test_doesnt_return_exercises_without_translation(self):
# The expected behaviour from KA's API is that it would return
# the english version of an exercise if either a translated
# exercise for the given language doesn't exist, or the
# language is unsupported.
exercise = self.khan_exercises[0]
lang = "aaa" # there's no language with code aaa... I hope?
path, retrieved_exercises = retrieve_html_exercises([exercise],
lang,
force=True)
assert not retrieved_exercises
def make_language_pack(lang, version, sublangargs, filename, ka_domain,
no_assessment_items, no_subtitles,
no_assessment_resources, no_dubbed_videos):
node_data, subtitle_data, content_catalog = retrieve_language_resources(
version, sublangargs, ka_domain, no_subtitles, no_dubbed_videos)
subtitles, subtitle_paths = subtitle_data.keys(), subtitle_data.values()
node_data = translate_nodes(node_data, content_catalog)
node_data = list(node_data)
node_data, dubbed_video_count = apply_dubbed_video_map(
node_data, subtitles, sublangargs["video_lang"])
html_exercise_ids, assessment_exercise_ids, node_data = separate_exercise_types(
node_data)
html_exercise_path, translated_html_exercise_ids = retrieve_html_exercises(
html_exercise_ids, lang)
# now include only the assessment item resources that we need
all_assessment_data, all_assessment_files = retrieve_all_assessment_item_data(
no_item_data=no_assessment_items,
no_item_resources=no_assessment_resources,
node_data=node_data,
lang=lang,
)
all_assessment_data = list(
remove_assessment_data_with_empty_widgets(all_assessment_data))
node_data = remove_nonexistent_assessment_items_from_exercises(
node_data, all_assessment_data)
node_data = clean_node_data_items(node_data)
assessment_data = list(
translate_assessment_item_text(
all_assessment_data,
content_catalog)) if lang != "en" else all_assessment_data
node_data = remove_untranslated_exercises(
node_data, translated_html_exercise_ids,
assessment_data) if lang != "en" else node_data
node_data = list(node_data)
node_data = sorted(node_data, key=lambda x: x.get('sort_order'))
with open('node_data_{0}.pickle'.format(lang), 'wb') as handle:
pickle.dump(node_data, handle)
with open('assessment_data_{0}.pickle'.format(lang), 'wb') as handle:
pickle.dump(assessment_data, handle)