Merge pull request #10 from Garmelon/sync-url

Add "Sync url" script from Christophe and release it automatically
This commit is contained in:
I-Al-Istannen 2020-10-07 09:29:48 +02:00 committed by GitHub
commit 43100f69d5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
6 changed files with 236 additions and 0 deletions

67
.github/workflows/package.yml vendored Normal file
View File

@ -0,0 +1,67 @@
name: Package Application with Pyinstaller
on:
push:
branches:
- "*"
tags:
- "v*"
jobs:
build:
runs-on: ${{ matrix.os }}
strategy:
matrix:
os: [ubuntu-latest, windows-latest]
steps:
- uses: actions/checkout@v2
- uses: actions/setup-python@v2
with:
python-version: '3.x'
- name: "Install dependencies"
run: "pip install setuptools pyinstaller rich requests beautifulsoup4 -f --upgrade"
- name: "Install sync_url.py"
run: "pyinstaller sync_url.py -F"
- uses: actions/upload-artifact@v2
with:
name: "Pferd Sync URL"
path: "dist/sync_url*"
release:
name: Release
needs: [build]
runs-on: ubuntu-latest
if: startsWith(github.ref, 'refs/tags/')
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
steps:
- name: "Checkout"
uses: actions/checkout@v2
- name: "Download artifacts"
uses: actions/download-artifact@v2
with:
name: "Pferd Sync URL"
- name: "look at folder structure"
run: "ls -lah"
- name: "Create release"
uses: softprops/action-gh-release@v1
- name: "Upload release artifacts"
uses: softprops/action-gh-release@v1
with:
body: "Download sync_url (or sync_url.exe on Windows) and run it in the terminal or CMD."
files: |
sync_url
sync_url.exe

4
.gitignore vendored
View File

@ -8,3 +8,7 @@ build/
.env
.vscode
ilias_cookies.txt
# PyInstaller
sync_url.spec
dist/

View File

@ -116,6 +116,16 @@ class IliasCrawler:
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]:
"""
Starts the crawl process for a course, yielding a list of elements to (potentially)
@ -235,6 +245,15 @@ class IliasCrawler:
"""
soup = self._get_page(url, {})
if soup.find(id="headerimage"):
element: bs4.Tag = soup.find(id="headerimage")
if "opencast" in element.attrs["src"].lower():
PRETTY.warning(f"Switched to crawling a video at {folder_path}")
if not self.dir_filter(folder_path, IliasElementType.VIDEO_FOLDER):
PRETTY.not_searching(folder_path, "user filter")
return []
return self._crawl_video_directory(folder_path, url)
result: List[IliasCrawlerEntry] = []
# Fetch all links and throw them to the general interpreter

View File

@ -230,6 +230,70 @@ class Pferd(Location):
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
def diva_kit(
self,

View File

@ -2,6 +2,7 @@
**P**rogramm zum **F**lotten, **E**infachen **R**unterladen von **D**ateien
- [Quickstart with `sync_url`](#quickstart-with-sync_url)
- [Installation](#installation)
- [Upgrading from 2.0.0 to 2.1.0+](#upgrading-from-200-to-210)
- [Example setup](#example-setup)
@ -12,6 +13,20 @@
- [Transform combinators](#transform-combinators)
- [A short, but commented example](#a-short-but-commented-example)
## Quickstart with `sync_url`
The `sync_url` program allows you to just synchronize a given ILIAS URL (of a
course, a folder, your personal desktop, etc.) without any extra configuration
or setting up. Download the program, open ILIAS, copy the URL from the address
bar and pass it to sync_url.
It bundles everything it needs in one executable and is easy to
use, but doesn't expose all the configuration options and tweaks a full install
does.
1. Download the `sync_url` binary from the [latest release](https://github.com/Garmelon/PFERD/releases/latest).
2. Run the binary in your terminal (`./sync_url` or `sync_url.exe` in the CMD) to see the help and use it
## Installation
Ensure that you have at least Python 3.8 installed.

67
sync_url.py Executable file
View File

@ -0,0 +1,67 @@
#!/usr/bin/env python
"""
A simple script to download a course by name from ILIAS.
"""
import argparse
from pathlib import Path
from urllib.parse import urlparse
from PFERD import Pferd
from PFERD.cookie_jar import CookieJar
from PFERD.ilias import (IliasCrawler, IliasElementType,
KitShibbolethAuthenticator)
from PFERD.utils import to_path
def main() -> None:
parser = argparse.ArgumentParser()
parser.add_argument("--test-run", action="store_true")
parser.add_argument('-c', '--cookies', nargs='?', default=None, help="File to store cookies in")
parser.add_argument('-f', '--no-videos', nargs='?', default=None, help="Don't download videos")
parser.add_argument('url', help="URL to the course page")
parser.add_argument('folder', nargs='?', default=None, help="Folder to put stuff into")
args = parser.parse_args()
url = urlparse(args.url)
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:
folder = args.folder
# Initialize pferd at the *parent of the passed folder*
# This is needed so Pferd's internal protections against escaping the working directory
# do not trigger (e.g. if somebody names a file in ILIAS '../../bad thing.txt')
pferd = Pferd(Path(Path(__file__).parent, folder).parent, test_run=args.test_run)
else:
# fetch course name from ilias
folder = crawler.find_element_name(args.url)
cookie_jar.save_cookies()
# Initialize pferd at the location of the script
pferd = Pferd(Path(__file__).parent, test_run=args.test_run)
def dir_filter(_: Path, element: IliasElementType) -> bool:
if args.no_videos:
return element not in [IliasElementType.VIDEO_FILE, IliasElementType.VIDEO_FOLDER]
return True
pferd.enable_logging()
# fetch
pferd.ilias_kit_folder(
target=folder,
full_url=args.url,
cookies=args.cookies,
dir_filter=dir_filter
)
if __name__ == "__main__":
main()