Bump version

.github/workflows/package.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     strategy:
       matrix:
-        os: [ubuntu-latest, windows-latest]
+        os: [ubuntu-latest, windows-latest, macos-latest]
 
     steps:
     - uses: actions/checkout@v2
@@ -28,6 +28,9 @@ jobs:
     - name: "Install sync_url.py"
       run: "pyinstaller sync_url.py -F"
 
+    - name: "Move artifact"
+      run: "mv dist/sync_url* dist/sync_url-${{ matrix.os }}"
+
     - uses: actions/upload-artifact@v2
       with:
         name: "Pferd Sync URL"
@@ -55,13 +58,17 @@ jobs:
     - name: "look at folder structure"
       run: "ls -lah"
 
+    - name: "Rename releases"
+      run: "mv sync_url-macos-latest pferd_sync_url_mac && mv sync_url-ubuntu-latest pferd_sync_url_linux && mv sync_url-windows-latest pferd_sync_url.exe"
+
     - 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."
+        body: "Download the correct sync_url for your platform and run it in the terminal or CMD. You might need to make it executable on Linux/Mac with `chmod +x <file>`. Also please enclose the *url you pass to the program in double quotes* or your shell might silently screw it up!"
         files: |
-          sync_url
+          pferd_sync_url_mac
-          sync_url.exe
+          pferd_sync_url_linux
+          pferd_sync_url.exe

PFERD/download_summary.py

@@ -5,6 +5,12 @@ from pathlib import Path
 from typing import List
 
 
+def _mergeNoDuplicate(first: List[Path], second: List[Path]) -> List[Path]:
+    tmp = list(set(first + second))
+    tmp.sort(key=lambda x: str(x.resolve()))
+    return tmp
+
+
 class DownloadSummary:
     """
     Keeps track of all new, modified or deleted files and provides a summary.
@@ -40,9 +46,9 @@ class DownloadSummary:
         """
         Merges ourselves with the passed summary. Modifies this object, but not the passed one.
         """
-        self._new_files += summary.new_files
+        self._new_files = _mergeNoDuplicate(self._new_files, summary.new_files)
-        self._modified_files += summary.modified_files
+        self._modified_files = _mergeNoDuplicate(self._modified_files, summary.modified_files)
-        self._deleted_files += summary.deleted_files
+        self._deleted_files = _mergeNoDuplicate(self._deleted_files, summary.deleted_files)
 
     def add_deleted_file(self, path: Path) -> None:
         """

PFERD/ilias/crawler.py

@@ -26,6 +26,10 @@ LOGGER = logging.getLogger(__name__)
 PRETTY = PrettyLogger(LOGGER)
 
 
+def _sanitize_path_name(name: str) -> str:
+    return name.replace("/", "-")
+
+
 class IliasElementType(Enum):
     """
     The type of an ilias element.
@@ -260,7 +264,7 @@ class IliasCrawler:
         links: List[bs4.Tag] = soup.select("a.il_ContainerItemTitle")
         for link in links:
             abs_url = self._abs_url_from_link(link)
-            element_path = Path(folder_path, link.getText().strip())
+            element_path = Path(folder_path, _sanitize_path_name(link.getText().strip()))
             element_type = self._find_type_from_link(element_path, link, abs_url)
 
             if element_type == IliasElementType.REGULAR_FILE:
@@ -377,7 +381,7 @@ class IliasCrawler:
             modification_date = demangle_date(modification_date_str)
 
         # Grab the name from the link text
-        name = link_element.getText()
+        name = _sanitize_path_name(link_element.getText())
         full_path = Path(path, name + "." + file_type)
 
         return [
@@ -508,7 +512,7 @@ class IliasCrawler:
         ).getText().strip()
         title += ".mp4"
 
-        video_path: Path = Path(parent_path, title)
+        video_path: Path = Path(parent_path, _sanitize_path_name(title))
 
         video_url = self._abs_url_from_link(link)
 
@@ -580,6 +584,7 @@ class IliasCrawler:
                 # Two divs, side by side. Left is the name, right is the link ==> get left
                 # sibling
                 file_name = file_link.parent.findPrevious(name="div").getText().strip()
+                file_name = _sanitize_path_name(file_name)
                 url = self._abs_url_from_link(file_link)
 
                 LOGGER.debug("Found file %r at %r", file_name, url)
@@ -593,10 +598,17 @@ class IliasCrawler:
 
         return results
 
-    def _get_page(self, url: str, params: Dict[str, Any]) -> bs4.BeautifulSoup:
+    def _get_page(self, url: str, params: Dict[str, Any],
+                  retry_count: int = 0) -> bs4.BeautifulSoup:
         """
         Fetches a page from ILIAS, authenticating when needed.
         """
+
+        if retry_count >= 4:
+            raise FatalException("Could not get a proper page after 4 tries. "
+                                 "Maybe your URL is wrong, authentication fails continuously, "
+                                 "your ILIAS connection is spotty or ILIAS is not well.")
+
         LOGGER.debug("Fetching %r", url)
 
         response = self._session.get(url, params=params)
@@ -617,7 +629,7 @@ class IliasCrawler:
 
         self._authenticator.authenticate(self._session)
 
-        return self._get_page(url, params)
+        return self._get_page(url, params, retry_count + 1)
 
     @staticmethod
     def _is_logged_in(soup: bs4.BeautifulSoup) -> bool:

PFERD/ipd.py

@@ -0,0 +1,151 @@
+"""
+Utility functions and a scraper/downloader for the IPD pages.
+"""
+import datetime
+import logging
+import math
+import os
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Callable, List, Optional
+from urllib.parse import urljoin
+
+import bs4
+import requests
+
+from PFERD.errors import FatalException
+from PFERD.utils import soupify
+
+from .logging import PrettyLogger
+from .organizer import Organizer
+from .tmp_dir import TmpDir
+from .transform import Transformable
+from .utils import stream_to_path
+
+LOGGER = logging.getLogger(__name__)
+PRETTY = PrettyLogger(LOGGER)
+
+
+@dataclass
+class IpdDownloadInfo(Transformable):
+    """
+    Information about an ipd entry.
+    """
+    url: str
+    modification_date: Optional[datetime.datetime]
+
+
+IpdDownloadStrategy = Callable[[Organizer, IpdDownloadInfo], bool]
+
+
+def ipd_download_new_or_modified(organizer: Organizer, info: IpdDownloadInfo) -> bool:
+    """
+    Accepts new files or files with a more recent modification date.
+    """
+    resolved_file = organizer.resolve(info.path)
+    if not resolved_file.exists():
+        return True
+    if not info.modification_date:
+        PRETTY.ignored_file(info.path, "could not find modification time, file exists")
+        return False
+
+    resolved_mod_time_seconds = resolved_file.stat().st_mtime
+
+    # Download if the info is newer
+    if info.modification_date.timestamp() > resolved_mod_time_seconds:
+        return True
+
+    PRETTY.ignored_file(info.path, "local file has newer or equal modification time")
+    return False
+
+
+class IpdCrawler:
+    # pylint: disable=too-few-public-methods
+    """
+    A crawler for IPD pages.
+    """
+
+    def __init__(self, base_url: str):
+        self._base_url = base_url
+
+    def _abs_url_from_link(self, link_tag: bs4.Tag) -> str:
+        """
+        Create an absolute url from an <a> tag.
+        """
+        return urljoin(self._base_url, link_tag.get("href"))
+
+    def crawl(self) -> List[IpdDownloadInfo]:
+        """
+        Crawls the playlist given in the constructor.
+        """
+        page = soupify(requests.get(self._base_url))
+
+        items: List[IpdDownloadInfo] = []
+
+        for link in page.findAll(name="a", attrs={"href": lambda x: x and x.endswith("pdf")}):
+            href: str = link.attrs.get("href")
+            name = href.split("/")[-1]
+
+            modification_date: Optional[datetime.datetime] = None
+            try:
+                enclosing_row: bs4.Tag = link.findParent(name="tr")
+                if enclosing_row:
+                    date_text = enclosing_row.find(name="td").text
+                    modification_date = datetime.datetime.strptime(date_text, "%d.%m.%Y")
+            except ValueError:
+                modification_date = None
+
+            items.append(IpdDownloadInfo(
+                Path(name),
+                url=self._abs_url_from_link(link),
+                modification_date=modification_date
+            ))
+
+        return items
+
+
+class IpdDownloader:
+    """
+    A downloader for ipd files.
+    """
+
+    def __init__(self, tmp_dir: TmpDir, organizer: Organizer, strategy: IpdDownloadStrategy):
+        self._tmp_dir = tmp_dir
+        self._organizer = organizer
+        self._strategy = strategy
+        self._session = requests.session()
+
+    def download_all(self, infos: List[IpdDownloadInfo]) -> None:
+        """
+        Download multiple files one after the other.
+        """
+        for info in infos:
+            self.download(info)
+
+    def download(self, info: IpdDownloadInfo) -> None:
+        """
+        Download a single file.
+        """
+        if not self._strategy(self._organizer, info):
+            self._organizer.mark(info.path)
+            return
+
+        with self._session.get(info.url, stream=True) as response:
+            if response.status_code == 200:
+                tmp_file = self._tmp_dir.new_path()
+                stream_to_path(response, tmp_file, info.path.name)
+                dst_path = self._organizer.accept_file(tmp_file, info.path)
+
+                if dst_path and info.modification_date:
+                    os.utime(
+                        dst_path,
+                        times=(
+                            math.ceil(info.modification_date.timestamp()),
+                            math.ceil(info.modification_date.timestamp())
+                        )
+                    )
+
+            elif response.status_code == 403:
+                raise FatalException("Received 403. Are you not using the KIT VPN?")
+            else:
+                PRETTY.warning(f"Could not download file, got response {response.status_code}")

PFERD/organizer.py

@@ -124,6 +124,8 @@ class Organizer(Location):
         self._cleanup(self.path)
 
     def _cleanup(self, start_dir: Path) -> None:
+        if not start_dir.exists():
+            return
         paths: List[Path] = list(start_dir.iterdir())
 
         # Recursively clean paths

PFERD/pferd.py

@@ -14,6 +14,8 @@ from .errors import FatalException, swallow_and_print_errors
 from .ilias import (IliasAuthenticator, IliasCrawler, IliasDirectoryFilter,
                     IliasDownloader, IliasDownloadInfo, IliasDownloadStrategy,
                     KitShibbolethAuthenticator, download_modified_or_new)
+from .ipd import (IpdCrawler, IpdDownloader, IpdDownloadInfo,
+                  IpdDownloadStrategy, ipd_download_new_or_modified)
 from .location import Location
 from .logging import PrettyLogger, enable_logging
 from .organizer import Organizer
@@ -294,6 +296,60 @@ class Pferd(Location):
 
         return organizer
 
+    @swallow_and_print_errors
+    def ipd_kit(
+            self,
+            target: Union[PathLike, Organizer],
+            url: str,
+            transform: Transform = lambda x: x,
+            download_strategy: IpdDownloadStrategy = ipd_download_new_or_modified,
+            clean: bool = True
+    ) -> Organizer:
+        """
+        Synchronizes a folder with a DIVA playlist.
+
+        Arguments:
+            target {Union[PathLike, Organizer]} -- The organizer / target folder to use.
+            url {str} -- the url to the page
+
+        Keyword Arguments:
+            transform {Transform} -- A transformation function for the output paths. Return None
+                to ignore a file. (default: {lambdax:x})
+            download_strategy {DivaDownloadStrategy} -- A function to determine which files need to
+                be downloaded. Can save bandwidth and reduce the number of requests.
+                (default: {diva_download_new})
+            clean {bool} -- Whether to clean up when the method finishes.
+        """
+        tmp_dir = self._tmp_dir.new_subdir()
+
+        if target is None:
+            PRETTY.starting_synchronizer("None", "IPD", url)
+            raise FatalException("Got 'None' as target directory, aborting")
+
+        if isinstance(target, Organizer):
+            organizer = target
+        else:
+            organizer = Organizer(self.resolve(to_path(target)))
+
+        PRETTY.starting_synchronizer(organizer.path, "IPD", url)
+
+        elements: List[IpdDownloadInfo] = IpdCrawler(url).crawl()
+        transformed = apply_transform(transform, elements)
+
+        if self._test_run:
+            self._print_transformables(transformed)
+            return organizer
+
+        downloader = IpdDownloader(tmp_dir=tmp_dir, organizer=organizer, strategy=download_strategy)
+        downloader.download_all(transformed)
+
+        if clean:
+            organizer.cleanup()
+
+        self._download_summary.merge(organizer.download_summary)
+
+        return organizer
+
     @swallow_and_print_errors
     def diva_kit(
             self,
@@ -352,4 +408,6 @@ class Pferd(Location):
         if clean:
             organizer.cleanup()
 
+        self._download_summary.merge(organizer.download_summary)
+
         return organizer

PFERD/transform.py

@@ -5,6 +5,8 @@ only files whose names match a regex, or renaming files from one numbering
 scheme to another.
 """
 
+import os
+import re
 from dataclasses import dataclass
 from pathlib import PurePath
 from typing import Callable, List, Optional, TypeVar
@@ -45,7 +47,8 @@ def apply_transform(
 
 # Transform combinators
 
-keep = lambda path: path
+def keep(path: PurePath) -> Optional[PurePath]:
+    return path
 
 def attempt(*args: Transform) -> Transform:
     def inner(path: PurePath) -> Optional[PurePath]:
@@ -125,3 +128,15 @@ def re_rename(regex: Regex, target: str) -> Transform:
             return path.with_name(target.format(*groups))
         return None
     return inner
+
+
+def sanitize_windows_path(path: PurePath) -> Optional[PurePath]:
+    """
+    A small function to escape characters that are forbidden in windows path names.
+    This method is a no-op on other operating systems.
+    """
+    # Escape windows illegal path characters
+    if os.name == 'nt':
+        sanitized_parts = [re.sub(r'[<>:"/|?]', "_", x) for x in list(path.parts)]
+        return PurePath(*sanitized_parts)
+    return path

README.md

@@ -25,7 +25,10 @@ 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
+2. Recognize that you most likely need to enclose the URL in `""` quotes to prevent your shell from interpreting `&` and other symbols
+3. Run the binary in your terminal (`./sync_url` or `sync_url.exe` in the CMD) to see the help and use it. I'd recommend using the `--cookies` option.  
+  If you are on **Linux/Mac**, you need to *make the file executable* using `chmod +x <file>`.  
+  If you are on **Mac**, you need to allow this unverified program to run (see e.g. [here](https://www.switchingtomac.com/tutorials/osx/how-to-run-unverified-apps-on-macos/))
 
 ## Installation
 
@@ -34,7 +37,7 @@ Ensure that you have at least Python 3.8 installed.
 To install PFERD or update your installation to the latest version, run this
 wherever you want to install or have already installed PFERD:
 ```
-$ pip install git+https://github.com/Garmelon/PFERD@v2.4.0
+$ pip install git+https://github.com/Garmelon/PFERD@v2.4.5
 ```
 
 The use of [venv] is recommended.
@@ -57,8 +60,8 @@ $ mkdir Vorlesungen
 $ cd Vorlesungen
 $ python3 -m venv .venv
 $ .venv/bin/activate
-$ pip install git+https://github.com/Garmelon/PFERD@v2.4.0
+$ pip install git+https://github.com/Garmelon/PFERD@v2.4.5
-$ curl -O https://raw.githubusercontent.com/Garmelon/PFERD/v2.4.0/example_config.py
+$ curl -O https://raw.githubusercontent.com/Garmelon/PFERD/v2.4.5/example_config.py
 $ python3 example_config.py
 $ deactivate
 ```

setup.py

@@ -2,7 +2,7 @@ from setuptools import find_packages, setup
 
 setup(
     name="PFERD",
-    version="2.4.0",
+    version="2.4.5",
     packages=find_packages(),
     install_requires=[
         "requests>=2.21.0",

sync_url.py

@@ -12,6 +12,7 @@ from PFERD import Pferd
 from PFERD.cookie_jar import CookieJar
 from PFERD.ilias import (IliasCrawler, IliasElementType,
                          KitShibbolethAuthenticator)
+from PFERD.transform import sanitize_windows_path
 from PFERD.utils import to_path
 
 
@@ -19,7 +20,7 @@ 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('--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()
@@ -59,7 +60,8 @@ def main() -> None:
         target=folder,
         full_url=args.url,
         cookies=args.cookies,
-        dir_filter=dir_filter
+        dir_filter=dir_filter,
+        transform=sanitize_windows_path
     )