Allow crawling courses or folders with sync_url

Video folders do not work, if they are passed directly. Their containing
folder must be specified instead.
This commit is contained in:
I-Al-Istannen 2020-09-28 20:00:01 +02:00
parent 74ea039458
commit 51a713fa04
3 changed files with 83 additions and 11 deletions

View File

@ -116,6 +116,16 @@ class IliasCrawler:
return urlunsplit((scheme, netloc, path, new_query_string, fragment)) return urlunsplit((scheme, netloc, path, new_query_string, fragment))
def recursive_crawl_url(self, url: str) -> List[IliasDownloadInfo]:
"""
Crawls a given url *and all reachable elements in it*.
Args:
url {str} -- the *full* url to crawl
"""
start_entries: List[IliasCrawlerEntry] = self._crawl_folder(Path(""), url)
return self._iterate_entries_to_download_infos(start_entries)
def crawl_course(self, course_id: str) -> List[IliasDownloadInfo]: def crawl_course(self, course_id: str) -> List[IliasDownloadInfo]:
""" """
Starts the crawl process for a course, yielding a list of elements to (potentially) Starts the crawl process for a course, yielding a list of elements to (potentially)

View File

@ -230,6 +230,70 @@ class Pferd(Location):
return organizer return organizer
@swallow_and_print_errors
def ilias_kit_folder(
self,
target: PathLike,
full_url: str,
dir_filter: IliasDirectoryFilter = lambda x, y: True,
transform: Transform = lambda x: x,
cookies: Optional[PathLike] = None,
username: Optional[str] = None,
password: Optional[str] = None,
download_strategy: IliasDownloadStrategy = download_modified_or_new,
clean: bool = True,
timeout: int = 5,
) -> Organizer:
"""
Synchronizes a folder with a given folder on the ILIAS instance of the KIT.
Arguments:
target {Path} -- the target path to write the data to
full_url {str} -- the full url of the folder/videos/course to crawl
Keyword Arguments:
dir_filter {IliasDirectoryFilter} -- A filter for directories. Will be applied on the
crawler level, these directories and all of their content is skipped.
(default: {lambdax:True})
transform {Transform} -- A transformation function for the output paths. Return None
to ignore a file. (default: {lambdax:x})
cookies {Optional[Path]} -- The path to store and load cookies from.
(default: {None})
username {Optional[str]} -- The SCC username. If none is given, it will prompt
the user. (default: {None})
password {Optional[str]} -- The SCC password. If none is given, it will prompt
the user. (default: {None})
download_strategy {DownloadStrategy} -- A function to determine which files need to
be downloaded. Can save bandwidth and reduce the number of requests.
(default: {download_modified_or_new})
clean {bool} -- Whether to clean up when the method finishes.
timeout {int} -- The download timeout for opencast videos. Sadly needed due to a
requests bug.
"""
# This authenticator only works with the KIT ilias instance.
authenticator = KitShibbolethAuthenticator(username=username, password=password)
PRETTY.starting_synchronizer(target, "ILIAS", "An ILIAS element by url")
if not full_url.startswith("https://ilias.studium.kit.edu"):
raise FatalException("Not a valid KIT ILIAS URL")
organizer = self._ilias(
target=target,
base_url="https://ilias.studium.kit.edu/",
crawl_function=lambda crawler: crawler.recursive_crawl_url(full_url),
authenticator=authenticator,
cookies=cookies,
dir_filter=dir_filter,
transform=transform,
download_strategy=download_strategy,
clean=clean,
timeout=timeout
)
self._download_summary.merge(organizer.download_summary)
return organizer
@swallow_and_print_errors @swallow_and_print_errors
def diva_kit( def diva_kit(
self, self,

View File

@ -23,10 +23,15 @@ def main() -> None:
parser.add_argument('folder', nargs='?', default=None, help="Folder to put stuff into") parser.add_argument('folder', nargs='?', default=None, help="Folder to put stuff into")
args = parser.parse_args() args = parser.parse_args()
# parse provided course URL
url = urlparse(args.url) url = urlparse(args.url)
query = parse_qs(url.query)
course_id = query['ref_id'][0] cookie_jar = CookieJar(to_path(args.cookies) if args.cookies else None)
session = cookie_jar.create_session()
authenticator = KitShibbolethAuthenticator()
crawler = IliasCrawler(url.scheme + '://' + url.netloc, session,
authenticator, lambda x, y: True)
cookie_jar.load_cookies()
if args.folder is not None: if args.folder is not None:
folder = args.folder folder = args.folder
@ -36,13 +41,6 @@ def main() -> None:
pferd = Pferd(Path(Path(__file__).parent, folder).parent, test_run=args.test_run) pferd = Pferd(Path(Path(__file__).parent, folder).parent, test_run=args.test_run)
else: else:
# fetch course name from ilias # fetch course name from ilias
cookie_jar = CookieJar(to_path(args.cookies) if args.cookies else None)
session = cookie_jar.create_session()
authenticator = KitShibbolethAuthenticator()
crawler = IliasCrawler(url.scheme + '://' + url.netloc, session,
authenticator, lambda x, y: True)
cookie_jar.load_cookies()
folder = crawler.find_element_name(args.url) folder = crawler.find_element_name(args.url)
cookie_jar.save_cookies() cookie_jar.save_cookies()
@ -51,7 +49,7 @@ def main() -> None:
pferd.enable_logging() pferd.enable_logging()
# fetch # fetch
pferd.ilias_kit(target=folder, course_id=course_id, cookies=args.cookies) pferd.ilias_kit_folder(target=folder, full_url=args.url, cookies=args.cookies)
if __name__ == "__main__": if __name__ == "__main__":