def generate_download_page(url, zipper):
""" Create a page for files that are meant to be downloaded (e.g. worksheets)
Args:
url (str): url to file that is meant to be downloaded
zipper (html_writer): where to write download page to
Returns path to page in zipfile (str)
"""
# Get template soup
soup = BeautifulSoup("", "html.parser")
with open('download.html', 'rb') as templatecode:
newpage = BeautifulSoup(templatecode.read(), 'html5lib')
# Determine if link is one of the recognized file types
download_url = url.split("?")[0]
filename = download_url.split("/")[-1]
if download_url.endswith('pdf'):
render_tag = soup.new_tag('embed')
elif next(
(e for e in IMAGE_EXTENSIONS if download_url.lower().endswith(e)),
None):
render_tag = soup.new_tag('img')
else:
LOGGER.error("Unknown file type found at {}".format(download_url))
return ""
# Add tag to new page and write page to zip
render_tag['src'] = zipper.write_url(format_url(download_url), filename)
newpage.body.append(render_tag)
return zipper.write_contents(
filename.split('.')[0] + ".html", newpage.prettify())
def get_parsed_html_from_url(url, *args, **kwargs):
response = sess.get(url, *args, **kwargs)
if response.status_code != 200:
LOGGER.error("STATUS: {}, URL: {}", response.status_code, url)
elif not response.from_cache:
LOGGER.debug("NOT CACHED:", url)
return BeautifulSoup(response.content, "html.parser")
def website_game_webresouce_to_ricecooker_node(lang, web_resource):
"""
Create Ricecooker Json structure for game from web resource dict `web_resource`.
"""
game_node = dict(
kind=content_kinds.HTML5,
source_id=web_resource['source_id'],
language=lang,
title=web_resource['title'],
description='source_url=' + web_resource['url'] if DEBUG_MODE else '',
license=PRADIGI_LICENSE,
thumbnail=web_resource.get('thumbnail_url'),
files=[],
)
zip_tmp_path = get_zip_file(web_resource['url'], web_resource['main_file'])
if zip_tmp_path:
zip_file = dict(
file_type=file_types.HTML5,
path=zip_tmp_path,
language=lang,
)
game_node['files'].append(zip_file)
LOGGER.debug('Created HTML5AppNode for game ' + web_resource['title'])
return game_node
else:
LOGGER.error('Failed to create zip for game at url=' +
web_resource['url'])
return None
def scrape_resource(url, topic):
resource = BeautifulSoup(downloader.read(url), 'html5lib')
LOGGER.info(' {}'.format(resource.find('h2').text))
filepath = download_resource(resource.find('div', {'class': 'decargas'}).find('a')['href'])
license = None
author = ''
for data_section in resource.find('div', {'class': 'datos_generales'}).find_all('h4'):
if 'Licencia' in data_section.text:
try:
license = LICENSE_MAP[data_section.find_next_sibling('p').text](copyright_holder="Ceibal")
except KeyError as e:
LOGGER.error(str(e))
license = licenses.CC_BYLicense
elif 'Autor' in data_section.text:
author = data_section.find_next_sibling('p').text
if filepath:
thumbnail = resource.find('div', {'class': 'img-recurso'}).find('img')['src']
if thumbnail.endswith('.gif'):
thumbnail = os.path.sep.join([DOWNLOAD_DIRECTORY, thumbnail.split('/')[-1].replace('.gif', '.png')])
with open(thumbnail, 'wb') as fobj:
fobj.write(downloader.read(resource.find('div', {'class': 'img-recurso'}).find('img')['src']))
topic.add_child(nodes.HTML5AppNode(
title=resource.find('h2').text,
source_id=url,
license=license,
author=author,
description=resource.find('form').find_all('p')[1].text,
thumbnail=thumbnail,
tags = [tag.text[:30] for tag in resource.find_all('a', {'class': 'tags'})],
files=[files.HTMLZipFile(path=filepath)],
))
def make_request(self, url, timeout=60, *args, method='GET', **kwargs):
"""
Failure-resistant HTTP GET/HEAD request helper method.
"""
retry_count = 0
max_retries = 5
while True:
try:
response = self.SESSION.request(method,
url,
*args,
timeout=timeout,
**kwargs)
break
except (requests.exceptions.ConnectionError,
requests.exceptions.ReadTimeout) as e:
retry_count += 1
LOGGER.warning(
"Connection error ('{msg}'); about to perform retry {count} of {trymax}."
.format(msg=str(e), count=retry_count, trymax=max_retries))
time.sleep(retry_count * 1)
if retry_count >= max_retries:
LOGGER.error("FAILED TO RETRIEVE:" + str(url))
return None
if response.status_code != 200:
LOGGER.error("ERROR " + str(response.status_code) +
' when getting url=' + url)
return None
return response
def to_tag(self, filename=None):
try:
img = self.create_tag('img')
img['src'] = self.to_zip(filename=filename)
return img
except BROKEN_EXCEPTIONS as e:
LOGGER.error(str(e))
return self.create_broken_link_message(self.url)
def download_resource(endpoint):
try:
url = '{}{}'.format(BASE_URL, endpoint.lstrip('/'))
filename, ext = os.path.splitext(endpoint)
filename = '{}.zip'.format(filename.lstrip('/').replace('/', '-'))
write_to_path = CeibalPageScraper(url, locale='es').to_file(filename=filename, directory=DOWNLOAD_DIRECTORY)
return write_to_path
except Exception as e:
LOGGER.error(str(e))
def _download_file(self, write_to_path):
with html_writer.HTMLWriter(write_to_path) as zipper:
try:
self.zipper = zipper
self.to_zip(filename='index.html')
except Exception as e:
# Any errors here will just say index.html file does not exist, so
# print out error for more descriptive debugging
LOGGER.error(str(e))
def to_tag(self, filename=None):
try:
embed = self.create_tag('embed')
embed['src'] = self.to_zip(filename=filename)
embed['width'] = '100%'
embed['style'] = 'height: 500px;max-height: 100vh;'
return embed
except BROKEN_EXCEPTIONS as e:
LOGGER.error(str(e))
return self.create_broken_link_message(self.url)
def transform_video_vertical(vertical, parent_title=None):
if 'children' not in vertical:
return None, []
# 1. LOOK FOR AN OPTIONAL html PREFIX TO USE AS DESCRIPTION
description = ''
# Extract an optional description from the first html node
first_child = vertical['children'][0]
if first_child['kind'] == 'html':
description = extract_text_from_html_item(first_child,
translate_from='ar')
if parent_title:
video_title = parent_title + ' ' + vertical['display_name']
else:
video_title = vertical['display_name']
# 2. GET THE VIDEO
videos = [ch for ch in vertical['children'] if ch['kind'] == 'video']
assert len(videos) == 1, 'multiple videos found'
video = videos[0]
video_dict = dict(kind=content_kinds.VIDEO,
source_id=video.get('youtube_id') or video.get('path'),
title=video_title,
author='Edraak',
description=description,
language=getlang('ar').code,
license=EDRAAK_LICENSE,
files=[])
if 'youtube_id' in video:
file_dict = dict(
file_type=content_kinds.VIDEO,
youtube_id=video['youtube_id'],
language=getlang('ar').code,
high_resolution=False,
)
elif 'path' in video:
file_dict = dict(
file_type=content_kinds.VIDEO,
path=video['path'],
language=getlang('ar').code,
ffmpeg_settings={"crf": 24},
)
else:
LOGGER.error('Video does not have youtube_id or path ' + str(video))
video_dict['files'].append(file_dict)
# 3. LOOK FOR AN OPTIONAL RESOURCES html
downloadable_resources = []
htmls = [ch for ch in vertical['children'] if ch['kind'] == 'html']
for html in htmls:
if 'downloadable_resources' in html:
downloadable_resources.extend(html['downloadable_resources'])
return video_dict, downloadable_resources
def to_tag(self, filename=None):
try:
audio = self.create_tag('audio')
audio['controls'] = 'controls'
audio['style'] = 'width: 100%;'
source = self.create_tag('source')
source['src'] = self.to_zip(filename=filename)
audio.append(source)
return audio
except BROKEN_EXCEPTIONS as e:
LOGGER.error(str(e))
return self.create_broken_link_message(self.url)
def to_tag(self, filename=None):
try:
video = self.create_tag('video')
video['controls'] = 'controls'
video['style'] = 'width: 100%;'
video['preload'] = 'auto'
source = self.create_tag('source')
source['src'] = self.to_zip(filename=filename)
video.append(source)
return video
except BROKEN_EXCEPTIONS as e:
LOGGER.error(str(e))
return self.create_broken_link_message(self.url)
def on_story_resource_page(self, url, page, context):
LOGGER.debug(' in on_story_resource_page' + url)
html = str(page)
story_resource_url = get_respath_url_from_html(html)
if story_resource_url:
page_dict = dict(
url=story_resource_url,
children=[],
)
page_dict.update(context)
context['parent']['children'].append(page_dict)
else:
LOGGER.error('Failed to find story_resource_url on page %s' % url)
def save_book(book_detail, channel):
book_id = book_detail["id"]
book_source_id = get_book_source_id(book_id)
book_title = book_detail["name"]
level_id = book_detail["readingLevel"]
language = book_detail["language"]
language_id = language["id"]
tags = book_detail["tags"]
epub_url = book_detail["epubUrl"]
pdf_urls = book_detail["pdfUrl"]
pdf_portrait_url = pdf_urls.get("portraitUrl", "") if pdf_urls else ""
pdf_landscape_url = pdf_urls.get("landscapeUrl", "") if pdf_urls else ""
pdf_booklet_url = pdf_urls.get("bookletUrl", "") if pdf_urls else ""
pdf_url = pdf_portrait_url or pdf_landscape_url or pdf_booklet_url
if not pdf_url and not epub_url:
LOGGER.error("No file found for \n {}".format(book_source_id))
raise NoFileAvailableError()
book_files = []
if pdf_url:
pdf_file = files.DocumentFile(path=pdf_url)
book_files.append(pdf_file)
if epub_url:
epub_file = files.EPubFile(path=epub_url)
book_files.append(epub_file)
book = nodes.DocumentNode(
source_id=book_source_id,
title=book_title,
license=licenses.
PUBLIC_DOMAIN, # TODO: get a real license and copyright holder
files=book_files)
language_topic = get_or_create_language_topic(language, channel)
level_topic = get_or_create_level_topic(level_id, language_id,
language_topic)
if not tags:
level_topic.add_child(book)
return
for tag in tags:
tag_topic = get_or_create_tag_topic(tag, language_id, level_id,
level_topic)
tag_topic.add_child(book)
def on_story_page(self, url, page, context):
LOGGER.debug(' in on_story_page' + url)
page_dict = dict(
kind='story_page',
url=url,
children=[],
)
page_dict.update(context)
context['parent']['children'].append(page_dict)
try:
body_row = page.find('div', {'id': 'body-row'})
contents_row = body_row.find('div', {'class': 'row'})
except Exception as e:
LOGGER.error('ERROR on_story_page: %s : %s' % (e, url))
return
contents = contents_row.find_all('div', {'class': 'col-md-3'})
for content in contents:
try:
title = get_text(content.find('div', {'class': 'txtline'}))
# TODO: description
thumbnail = content.find('a').find('img')['src']
thumbnail = get_absolute_path(thumbnail)
# get_fun_content_link
link = content.find('a')
source_id = link['href'][1:]
story_resource_url = get_absolute_path(link['href'])
if self.should_ignore_url(story_resource_url):
print('ignoring story content', title, story_resource_url)
continue
LOGGER.debug(' story_resource_page: %s: %s' % (source_id, title))
context = dict(
parent = page_dict,
kind='story_resource_page',
title=title,
source_id=source_id,
thumbnail_url=thumbnail,
)
self.enqueue_url_and_context(story_resource_url, context)
except Exception as e:
LOGGER.error('on_story_page: %s : %s' % (e, content))
def scrape_video_collection(url, topic):
""" Scrape videos under video collection and add to the topic node
Args:
url (str): url to video page (e.g. https://www.exploratorium.edu/video/inflatable-jimmy-kuehnle)
topic (TopicNode): topic to add video nodes to
"""
try:
collection_contents = BeautifulSoup(read(url), 'html5lib')
for result in collection_contents.find_all('div',
{'class': 'search-result'}):
header = result.find('div',
{'class': 'views-field-field-html-title'})
LOGGER.info(" {}".format(header.text.strip()))
# Get video from given url
description = result.find('div', {'class': 'search-description'})
video_contents = BeautifulSoup(read(header.find('a')['href']),
'html.parser')
for k, v in get_brightcove_mapping(video_contents).items():
video_node = nodes.VideoNode(
source_id=k,
title=header.text.strip().replace("’", "'"),
description=description.text.strip()
if description else "",
license=LICENSE,
copyright_holder=COPYRIGHT_HOLDER,
author=v.get('author') or "",
files=[
files.WebVideoFile(v['url'], high_resolution=False)
],
thumbnail=get_thumbnail_url(result.find('img')['src']),
)
# If video doesn't already exist here, add to topic
if not next((c for c in topic.children
if c.source_id == video_node.source_id), None):
topic.add_child(video_node)
# Scrape next page (if any)
next_page_url = get_next_page_url(collection_contents)
if next_page_url:
scrape_video_collection(next_page_url, topic)
except requests.exceptions.HTTPError:
LOGGER.error("Could not read collection at {}".format(url))
def get_phet_zip_file(zip_file_url, main_file_and_query):
"""
Phet simulations are provided in the zip file `phet.zip`, and the entry point
is passed as a GET parameter in `main_file_and_query`. To make these compatible
with Kolibri's default behaviour of loading index.html, we will:
- Rename index.html to phetindex.thml
- Add a custom index.html that uses javascrpt redirect to phetindex.thml?{sim_id}
"""
u = urlparse(main_file_and_query)
idk, sim_id = u.query.split('=')
assert idk == 'id', 'unknown query sting format found' + main_file_and_query
main_file = u.scheme + '://' + u.netloc + u.path # skip querystring
destpath = tempfile.mkdtemp()
LOGGER.info('saving phet zip file in dir ' + destpath)
try:
download_file(zip_file_url, destpath, request_fn=make_request)
zip_filename = zip_file_url.split('/')[-1]
zip_basename = zip_filename.rsplit('.', 1)[0]
zip_folder = os.path.join(destpath, zip_basename)
# Extract zip file contents.
local_zip_file = os.path.join(destpath, zip_filename)
with zipfile.ZipFile(local_zip_file) as zf:
zf.extractall(destpath)
# Rename main_file to phetindex.html
main_file = main_file.split('/')[-1]
src = os.path.join(zip_folder, main_file)
dest = os.path.join(zip_folder, 'phetindex.html')
os.rename(src, dest)
# Create the
index_html = PHET_INDEX_HTML_TEMPLATE.format(sim_id=sim_id)
with open(os.path.join(zip_folder, 'index.html'), 'w') as indexf:
indexf.write(index_html)
# Always be zipping!
return create_predictable_zip(zip_folder)
except Exception as e:
LOGGER.error("get_phet_zip_file: %s, %s, %s, %s" %
(zip_file_url, main_file_and_query, destpath, e))
return None
def parse_tsv_file(filepath):
"""
Load data from the TSV file located at `filepath` using csv.DictReader.
Returns: a dict {id --> datum} of all the rows.
"""
print('Loading TSV file', filepath)
data_by_id = {}
with open(filepath, encoding="utf-8-sig") as tsvfile:
reader = csv.DictReader(tsvfile, dialect='excel-tab')
for row in reader:
if not row['id']:
raise ValueError("Row with missing id " + str(row))
try:
clean_row = clean_tsv_row(row)
data_by_id[row['id']] = clean_row
except json.JSONDecodeError as e:
LOGGER.error('Failed to parse row=' + str(dict(row)))
return data_by_id
def get_subtopics(parent, path):
doc = get_page(path)
try:
menu_row = doc.find('div', {'id': 'body-row'})
menu_row = menu_row.find('div', {'class': 'col-md-2'})
except Exception as e:
LOGGER.error('get_subtopics: %s : %s' % (e, doc))
return
for subtopic in menu_row.find_all('a'):
try:
title = subtopic.get_text().strip()
source_id = get_source_id(subtopic['href'])
LOGGER.info(' subtopic: %s: %s' % (source_id, title))
node = TopicNode(title=title, source_id=source_id)
parent.add_child(node)
get_lessons(node, subtopic['href'])
except Exception as e:
LOGGER.error('get_subtopics: %s : %s' % (e, subtopic))
def on_subtopic_page(self, url, page, context):
LOGGER.debug(' in on_subtopic_page ' + url)
page_dict = dict(
kind='subtopic_page', # redundant...
url=url,
children=[],
)
page_dict.update(context)
context['parent']['children'].append(page_dict)
try:
menu_row = page.find('div', {'id': 'body-row'})
menu_row = menu_row.find('div', {'class': 'col-md-9'})
except Exception as e:
LOGGER.error('on_subtopic_page: %s : %s' % (e, page))
return
for lesson in menu_row.find_all('div', {'class': 'thumbnail'}):
try:
title = lesson.find('div', {'class': 'txtline'}).get_text().strip()
caption = lesson.find('div', class_='caption')
description = get_text(caption) if caption else ''
lesson_url = urljoin(url, lesson.find('a')['href'])
if self.should_ignore_url(lesson_url):
LOGGER.info('ignoring lesson' + lesson_url)
continue
thumbnail_src = lesson.find('a').find('img')['src']
thumbnail_url = urljoin(url, thumbnail_src)
source_id = get_source_id(lesson.find('a')['href'])
LOGGER.debug(' lesson: %s: %s' % (source_id, title))
context = dict(
parent=page_dict,
kind='lesson_page',
title=title,
description=description,
source_id=source_id,
thumbnail_url=thumbnail_url,
children=[],
)
self.enqueue_url_and_context(lesson_url, context)
# get_contents(node, link)
except Exception as e:
LOGGER.error('on_subtopic_page: %s : %s' % (e, lesson))
def get_subtitles_using_youtube_dl(youtube_id):
youtube_url = 'https://youtube.com/watch?v=' + youtube_id
yt_resource = YouTubeResource(youtube_url)
lang_codes = []
try:
result = yt_resource.get_resource_subtitles()
# TODO(ivan) Consider including auto-generated subtitles to increase
# coverage and handle edge cases of videos that are transalted
# but no metadata: https://www.youtube.com/watch?v=qlGjA9p1UAM
if result:
for lang_code, lang_subs in result['subtitles'].items():
for lang_sub in lang_subs:
if 'ext' in lang_sub and lang_sub[
'ext'] == 'vtt' and lang_code not in lang_codes:
lang_codes.append(lang_code)
except Exception as e:
LOGGER.error('get_subtitles_using_youtube_dl failed for ' +
youtube_url)
LOGGER.error(str(e))
return lang_codes
def on_special_subtopic_page(self, url, page, context):
LOGGER.debug(' in on_special_subtopic_page ' + url)
page_dict = dict(
kind='special_subtopic_page', # redundant... -- mismatc with original special_subtopic_page
url=url,
children=[],
)
page_dict.update(context)
context['parent']['children'].append(page_dict)
try:
menu_row = page.find('div', {'id': 'body-row'})
menu_row = menu_row.find('div', {'class': 'col-md-2'})
print(str(menu_row))
except Exception as e:
LOGGER.error('on_subtopic_page: %s : %s' % (e, page))
return
for link in menu_row.find_all('a', {'class': 'list-group-item'}):
try:
title = link.get_text().strip()
description = ''
lesson_url = urljoin(url, link['href'])
if self.should_ignore_url(lesson_url):
LOGGER.info('ignoring lesson' + lesson_url)
continue
source_id = get_source_id(link['href'])
LOGGER.debug(' special lesson: %s: %s' % (source_id, title))
context = dict(
parent=page_dict,
kind='fun_page',
title=title,
description=description,
source_id=source_id,
thumbnail_url=None,
children=[],
)
self.enqueue_url_and_context(lesson_url, context)
# get_contents(node, link)
except Exception as e:
LOGGER.error('on_special_subtopic_page: %s : %s' % (e, link))
def get_topics(parent, path):
doc = get_page(path)
try:
menu_row = doc.find('div', {'id': 'menu-row'})
except Exception as e:
LOGGER.error('get_topics: %s : %s' % (e, doc))
return
for topic in menu_row.find_all('a'):
try:
if topic['href'] == '#':
continue
title = topic.get_text().strip()
source_id = get_source_id(topic['href'])
LOGGER.info('topic: %s: %s' % (source_id, title))
node = TopicNode(title=title, source_id=source_id)
parent.add_child(node)
get_subtopics(node, topic['href'])
if DEBUG_MODE:
return
except Exception as e:
LOGGER.error('get_topics: %s : %s' % (e, topic))
def _get_youtube_info(self, use_proxy=True, use_cache=True, options=None):
youtube_info = None
# 1. Try to get from cache if allowed:
if os.path.exists(self.cache_path) and use_cache:
LOGGER.info("==> [%s] Retrieving cached information...", self.__str__())
youtube_info = json.load(open(self.cache_path))
# 2. Fetch info from youtube_dl
if not youtube_info:
LOGGER.info("==> [%s] Requesting info from youtube...", self.__str__())
os.makedirs(self.cache_dir, exist_ok=True)
try:
youtube_resource = YouTubeResource(self.url, useproxy=use_proxy)
except youtube_dl.utils.ExtractorError as e:
if "unavailable" in str(e):
LOGGER.error("==> [%s] Resource unavailable for URL: %s", self.__str__, self.url)
return None
if youtube_resource:
try:
# Save YouTube info to JSON cache file
youtube_info = youtube_resource.get_resource_info(options)
if youtube_info:
json.dump(youtube_info,
open(self.cache_path, 'w'),
indent=4,
ensure_ascii=False,
sort_keys=True)
else:
LOGGER.error("==> [%s] Failed to extract YouTube info", self.__str__())
except Exception as e:
LOGGER.error("==> [%s] Failed to get YouTube info: %s", self.__str__(), e)
return None
return youtube_info
def download(url, write_to_path, attempts=DOWNLOAD_ATTEMPTS):
""" Download the web video
Args:
url (str): url to video to download
write_to_path (str): where to write video to
attempts (int): how many times to reattempt a download
"""
try:
video_format = "bestvideo[height<=480][ext=mp4]+bestaudio[ext=m4a]/best[height<=480][ext=mp4]"
with youtube_dl.YoutubeDL({
"format": video_format,
"outtmpl": write_to_path
}) as ydl:
ydl.download([url])
except youtube_dl.utils.DownloadError as e:
# If there are more attempts, try again. Otherwise, return error
if attempts > 0:
download(url, write_to_path, attempts=attempts - 1)
else:
LOGGER.error("Could not download video {} ({})".format(
url, str(e)))
raise e
def on_topic_page(self, url, page, context):
LOGGER.debug('in on_topic_page ' + url)
page_dict = dict(
kind='topic_page',
url=url,
children=[],
)
page_dict.update(context)
context['parent']['children'].append(page_dict)
try:
body_row = page.find('div', {'id': 'body-row'})
menu_row = body_row.find('div', {'class': 'col-md-2'})
subtopics = menu_row.find_all('a')
except Exception as e:
LOGGER.error('ERROR get_subtopics: %s : %s' % (e, url))
return
for subtopic in subtopics:
try:
subtopic_url = urljoin(url, subtopic['href'])
if self.should_ignore_url(subtopic_url):
print('ignoring subtopic', subtopic_url)
continue
title = get_text(subtopic)
source_id = get_source_id(subtopic['href'])
LOGGER.debug(' found subtopic: %s: %s' % (source_id, title))
context = dict(
parent=page_dict,
kind='subtopic_page',
title=title,
source_id=source_id,
children=[],
)
self.enqueue_url_and_context(subtopic_url, context)
except Exception as e:
LOGGER.error('on_topic_page: %s : %s' % (e, subtopic))
def get_subtree_by_subject_en(lang, subject):
if lang not in PRADIGI_LANG_URL_MAP:
raise ValueError('Language `lang` must be in PRADIGI_LANG_URL_MAP')
wrt_filename = 'chefdata/trees/pradigi_{}_web_resource_tree.json'.format(
lang)
with open(wrt_filename) as jsonfile:
web_resource_tree = json.load(jsonfile)
subject_subtrees = web_resource_tree['children']
try:
for subject_subtree in subject_subtrees:
if 'subject_en' in subject_subtree and subject_subtree[
'subject_en'] == subject:
return subject_subtree
elif 'source_id' in subject_subtree and subject_subtree[
'source_id'] == subject:
return subject_subtree
else:
pass
# print('no subject_en in '+ subject_subtree['source_id'])
except Exception as e:
LOGGER.error("in get_subtree_by_subject_en: %s, %s, %s, %s" %
(lang, subject, subject_subtree, e))
return None
def get_lessons(parent, path):
doc = get_page(path)
try:
menu_row = doc.find('div', {'id': 'body-row'})
menu_row = menu_row.find('div', {'class': 'col-md-9'})
except Exception as e:
LOGGER.error('get_lessons: %s : %s' % (e, doc))
return
for lesson in menu_row.find_all('div', {'class': 'thumbnail'}):
try:
title = lesson.find('div', {'class': 'txtline'}).get_text().strip()
link = lesson.find('a')['href']
thumbnail = lesson.find('a').find('img')['src']
thumbnail = get_absolute_path(thumbnail)
source_id = get_source_id(link)
LOGGER.info(' lesson: %s: %s' % (source_id, title))
node = TopicNode(title=title,
source_id=source_id,
thumbnail=thumbnail)
parent.add_child(node)
get_contents(node, link)
except Exception as e:
LOGGER.error('get_lessons: %s : %s' % (e, lesson))
def get_zip_file(zip_file_url, main_file):
"""HTML games are provided as zip files, the entry point of the game is
main_file. main_file needs to be renamed to index.html to make it
compatible with Kolibri.
"""
destpath = tempfile.mkdtemp()
try:
download_file(zip_file_url, destpath, request_fn=make_request)
zip_filename = zip_file_url.split('/')[-1]
zip_basename = zip_filename.rsplit('.', 1)[0]
zip_folder = os.path.join(destpath, zip_basename)
# Extract zip file contents.
local_zip_file = os.path.join(destpath, zip_filename)
with zipfile.ZipFile(local_zip_file) as zf:
zf.extractall(destpath)
# In some cases, the files are under the www directory,
# let's move them up one level.
www_dir = os.path.join(zip_folder, 'www')
if os.path.isdir(www_dir):
files = os.listdir(www_dir)
for f in files:
shutil.move(os.path.join(www_dir, f), zip_folder)
# Rename main_file to index.html.
main_file = main_file.split('/')[-1]
src = os.path.join(zip_folder, main_file)
dest = os.path.join(zip_folder, 'index.html')
os.rename(src, dest)
return create_predictable_zip(zip_folder)
except Exception as e:
LOGGER.error("get_zip_file: %s, %s, %s, %s" %
(zip_file_url, main_file, destpath, e))
return None
def download_zip_file(url):
if not url:
return (False, None)
if get_suffix(url) != '.zip':
return (False, None)
response = sess.get(url)
if response.status_code != 200:
LOGGER.error("STATUS: {}, URL: {}", response.status_code, url)
return (False, None)
elif not response.from_cache:
LOGGER.debug("NOT CACHED:", url)
archive = zipfile.ZipFile(io.BytesIO(response.content))
archive_members = list(
filter(lambda f: f.filename.endswith('.pdf'), archive.infolist()))
archive_member_names = [None] * len(archive_members)
for i, pdf in enumerate(archive_members):
path = os.path.join(PDFS_DATA_DIR, pdf.filename)
archive_member_names[i] = path
if not os.path.exists(path):
archive.extract(pdf, PDFS_DATA_DIR)
return (True, archive_member_names)
