Bump version to 3.8.0

Co-authored-by: Tim <me@scriptim.dev>

.github/workflows/build-and-release.yml

@@ -14,7 +14,7 @@ jobs:
       fail-fast: false
       matrix:
         os: [ubuntu-latest, windows-latest, macos-13, macos-latest]
-        python: ["3.9"]
+        python: ["3.11"]
     steps:
       - uses: actions/checkout@v4
 

CHANGELOG.md

@@ -22,6 +22,20 @@ ambiguous situations.
 
 ## Unreleased
 
+## 3.8.0 - 2025-04-15
+
+### Added
+- Support for ILIAS 9
+
+### Changed
+- Added prettier CSS to forum threads
+- Increase minimum supported Python version to 3.11
+
+## Fixed
+- File links in report on Windows
+- TOTP authentication in KIT Shibboleth
+- Forum crawling only considering the first 20 entries
+
 ## 3.7.0 - 2024-11-13
 
 ### Added

PFERD/auth/keyring.py

@@ -1,4 +1,4 @@
-from typing import Optional, Tuple
+from typing import Optional, Tuple, cast
 
 import keyring
 
@@ -13,7 +13,7 @@ class KeyringAuthSection(AuthSection):
         return self.s.get("username")
 
     def keyring_name(self) -> str:
-        return self.s.get("keyring_name", fallback=NAME)
+        return cast(str, self.s.get("keyring_name", fallback=NAME))
 
 
 class KeyringAuthenticator(Authenticator):

PFERD/crawl/crawler.py

@@ -149,9 +149,7 @@ class CrawlerSection(Section):
         return self.s.getboolean("skip", fallback=False)
 
     def output_dir(self, name: str) -> Path:
-        # TODO Use removeprefix() after switching to 3.9
-        if name.startswith("crawl:"):
-            name = name[len("crawl:"):]
+        name = name.removeprefix("crawl:")
         return Path(self.s.get("output_dir", name)).expanduser()
 
     def redownload(self) -> Redownload:
@@ -294,6 +292,35 @@ class Crawler(ABC):
         log.explain("Answer: Yes")
         return CrawlToken(self._limiter, path)
 
+    def should_try_download(
+            self,
+            path: PurePath,
+            *,
+            etag_differs: Optional[bool] = None,
+            mtime: Optional[datetime] = None,
+            redownload: Optional[Redownload] = None,
+            on_conflict: Optional[OnConflict] = None,
+    ) -> bool:
+        log.explain_topic(f"Decision: Should Download {fmt_path(path)}")
+
+        if self._transformer.transform(path) is None:
+            log.explain("Answer: No (ignored)")
+            return False
+
+        should_download = self._output_dir.should_try_download(
+            path,
+            etag_differs=etag_differs,
+            mtime=mtime,
+            redownload=redownload,
+            on_conflict=on_conflict
+        )
+        if should_download:
+            log.explain("Answer: Yes")
+            return True
+        else:
+            log.explain("Answer: No")
+            return False
+
     async def download(
             self,
             path: PurePath,

PFERD/crawl/http_crawler.py

@@ -3,7 +3,7 @@ import http.cookies
 import ssl
 from datetime import datetime
 from pathlib import Path, PurePath
-from typing import Any, Dict, List, Optional, Tuple
+from typing import Any, Dict, List, Optional, Tuple, cast
 
 import aiohttp
 import certifi
@@ -22,7 +22,7 @@ ETAGS_CUSTOM_REPORT_VALUE_KEY = "etags"
 
 class HttpCrawlerSection(CrawlerSection):
     def http_timeout(self) -> float:
-        return self.s.getfloat("http_timeout", fallback=20)
+        return self.s.getfloat("http_timeout", fallback=30)
 
 
 class HttpCrawler(Crawler):
@@ -187,12 +187,12 @@ class HttpCrawler(Crawler):
             if level == 0 or (level == 1 and drop_h1):
                 return PurePath()
 
-            level_heading = tag.find_previous(name=f"h{level}")
+            level_heading = cast(Optional[Tag], tag.find_previous(name=f"h{level}"))
 
             if level_heading is None:
                 return find_associated_headings(tag, level - 1)
 
-            folder_name = level_heading.getText().strip()
+            folder_name = level_heading.get_text().strip()
             return find_associated_headings(level_heading, level - 1) / folder_name
 
         # start at level <h3> because paragraph-level headings are usually too granular for folder names
@@ -231,6 +231,7 @@ class HttpCrawler(Crawler):
 
                 etag_header = resp.headers.get("ETag")
                 last_modified_header = resp.headers.get("Last-Modified")
+                last_modified = None
 
                 if last_modified_header:
                     try:

PFERD/crawl/ilias/file_templates.py

@@ -1,5 +1,5 @@
 from enum import Enum
-from typing import Optional
+from typing import Optional, cast
 
 import bs4
 
@@ -126,6 +126,88 @@ _learning_module_template = """
 </html>
 """
 
+_forum_thread_template = """
+<!DOCTYPE html>
+<html lang="en">
+    <head>
+        <meta charset="UTF-8">
+        <title>ILIAS - Forum: {{name}}</title>
+        <style>
+            * {
+                box-sizing: border-box;
+            }
+            body {
+                font-family: 'Open Sans', Verdana, Arial, Helvetica, sans-serif;
+                padding: 8px;
+            }
+            ul, ol, p {
+                margin: 1.2em 0;
+            }
+            p {
+                margin-top: 8px;
+                margin-bottom: 8px;
+            }
+            a {
+                color: #00876c;
+                text-decoration: none;
+                cursor: pointer;
+            }
+            a:hover {
+                text-decoration: underline;
+            }
+            body > p:first-child > span:first-child {
+                font-size: 1.6em;
+            }
+            body > p:first-child > span:first-child ~ span.default {
+                display: inline-block;
+                font-size: 1.2em;
+                padding-bottom: 8px;
+            }
+            .ilFrmPostContent {
+                margin-top: 8px;
+                max-width: 64em;
+            }
+            .ilFrmPostContent > *:first-child {
+                margin-top: 0px;
+            }
+            .ilFrmPostTitle {
+                margin-top: 24px;
+                color: #00876c;
+                font-weight: bold;
+            }
+            #ilFrmPostList {
+                list-style: none;
+                padding-left: 0;
+            }
+            li.ilFrmPostRow {
+                padding: 3px 0 3px 3px;
+                margin-bottom: 24px;
+                border-left: 6px solid #dddddd;
+            }
+            .ilFrmPostRow > div {
+                display: flex;
+            }
+            .ilFrmPostImage img {
+                margin: 0 !important;
+                padding: 6px 9px 9px 6px;
+            }
+            .ilUserIcon {
+                width: 115px;
+            }
+            .small {
+                text-decoration: none;
+                font-size: 0.75rem;
+                color: #6f6f6f;
+            }
+        </style>
+    </head>
+    <body>
+    {{heading}}
+    {{content}}
+    </body>
+</html>
+""".strip()  # noqa: E501 line too long
+
 
 def learning_module_template(body: bs4.Tag, name: str, prev: Optional[str], next: Optional[str]) -> str:
     # Seems to be comments, ignore those.
@@ -139,13 +221,13 @@ def learning_module_template(body: bs4.Tag, name: str, prev: Optional[str], next
         </div>
     """
     if prev and body.select_one(".ilc_page_lnav_LeftNavigation"):
-        text = body.select_one(".ilc_page_lnav_LeftNavigation").getText().strip()
+        text = cast(bs4.Tag, body.select_one(".ilc_page_lnav_LeftNavigation")).get_text().strip()
         left = f'<a href="{prev}">{text}</a>'
     else:
         left = "<span></span>"
 
     if next and body.select_one(".ilc_page_rnav_RightNavigation"):
-        text = body.select_one(".ilc_page_rnav_RightNavigation").getText().strip()
+        text = cast(bs4.Tag, body.select_one(".ilc_page_rnav_RightNavigation")).get_text().strip()
         right = f'<a href="{next}">{text}</a>'
     else:
         right = "<span></span>"
@@ -160,8 +242,17 @@ def learning_module_template(body: bs4.Tag, name: str, prev: Optional[str], next
             "{{left}}", left).replace("{{right}}", right).encode())
         )
 
-    body = body.prettify()
-    return _learning_module_template.replace("{{body}}", body).replace("{{name}}", name)
+    body_str = cast(str, body.prettify())
+    return _learning_module_template.replace("{{body}}", body_str).replace("{{name}}", name)
+
+
+def forum_thread_template(name: str, url: str, heading: bs4.Tag, content: bs4.Tag) -> str:
+    if title := cast(Optional[bs4.Tag], heading.find(name="b")):
+        title.wrap(bs4.Tag(name="a", attrs={"href": url}))
+    return _forum_thread_template \
+        .replace("{{name}}", name) \
+        .replace("{{heading}}", cast(str, heading.prettify())) \
+        .replace("{{content}}", cast(str, content.prettify()))
 
 
 class Links(Enum):

PFERD/crawl/ilias/ilias_html_cleaner.py

@@ -1,3 +1,5 @@
+from typing import cast
+
 from bs4 import BeautifulSoup, Comment, Tag
 
 _STYLE_TAG_CONTENT = """
@@ -70,18 +72,18 @@ def insert_base_markup(soup: BeautifulSoup) -> BeautifulSoup:
 
 
 def clean(soup: BeautifulSoup) -> BeautifulSoup:
-    for block in soup.find_all(class_=lambda x: x in _ARTICLE_WORTHY_CLASSES):
+    for block in cast(list[Tag], soup.find_all(class_=lambda x: x in _ARTICLE_WORTHY_CLASSES)):
         block.name = "article"
 
-    for block in soup.find_all("h3"):
+    for block in cast(list[Tag], soup.find_all("h3")):
         block.name = "div"
 
-    for block in soup.find_all("h1"):
+    for block in cast(list[Tag], soup.find_all("h1")):
         block.name = "h3"
 
-    for block in soup.find_all(class_="ilc_va_ihcap_VAccordIHeadCap"):
+    for block in cast(list[Tag], soup.find_all(class_="ilc_va_ihcap_VAccordIHeadCap")):
         block.name = "h3"
-        block["class"] += ["accordion-head"]
+        block["class"] += ["accordion-head"]  # type: ignore
 
     for dummy in soup.select(".ilc_text_block_Standard.ilc_Paragraph"):
         children = list(dummy.children)
@@ -97,7 +99,7 @@ def clean(soup: BeautifulSoup) -> BeautifulSoup:
         if figure := video.find_parent("figure"):
             figure.decompose()
 
-    for hrule_imposter in soup.find_all(class_="ilc_section_Separator"):
+    for hrule_imposter in cast(list[Tag], soup.find_all(class_="ilc_section_Separator")):
         hrule_imposter.insert(0, soup.new_tag("hr"))
 
     return soup

PFERD/crawl/ilias/ilias_web_crawler.py

@@ -19,10 +19,10 @@ from ...utils import fmt_path, soupify, url_set_query_param
 from ..crawler import CrawlError, CrawlToken, CrawlWarning, DownloadToken, anoncritical
 from ..http_crawler import HttpCrawler, HttpCrawlerSection
 from .async_helper import _iorepeat
-from .file_templates import Links, learning_module_template
+from .file_templates import Links, forum_thread_template, learning_module_template
 from .ilias_html_cleaner import clean, insert_base_markup
 from .kit_ilias_html import (IliasElementType, IliasForumThread, IliasLearningModulePage, IliasPage,
-                             IliasPageElement, _sanitize_path_name, parse_ilias_forum_export)
+                             IliasPageElement, IliasSoup, _sanitize_path_name, parse_ilias_forum_export)
 from .shibboleth_login import ShibbolethLogin
 
 TargetType = Union[str, int]
@@ -105,7 +105,6 @@ class IliasWebCrawlerSection(HttpCrawlerSection):
 
 
 _DIRECTORY_PAGES: Set[IliasElementType] = {
-    IliasElementType.COURSE,
     IliasElementType.EXERCISE,
     IliasElementType.EXERCISE_FILES,
     IliasElementType.FOLDER,
@@ -257,6 +256,7 @@ instance's greatest bottleneck.
             async with cl:
                 next_stage_url: Optional[str] = url
                 current_parent = current_element
+                page = None
 
                 while next_stage_url:
                     soup = await self._get_page(next_stage_url)
@@ -266,18 +266,19 @@ instance's greatest bottleneck.
                     # If we expect to find a root course, enforce it
                     if current_parent is None and expected_course_id is not None:
                         perma_link = IliasPage.get_soup_permalink(soup)
-                        if not perma_link or "crs_" not in perma_link:
+                        if not perma_link or "crs/" not in perma_link:
                             raise CrawlError("Invalid course id? Didn't find anything looking like a course")
                         if str(expected_course_id) not in perma_link:
                             raise CrawlError(f"Expected course id {expected_course_id} but got {perma_link}")
 
-                    page = IliasPage(soup, next_stage_url, current_parent)
+                    page = IliasPage(soup, current_parent)
                     if next_element := page.get_next_stage_element():
                         current_parent = next_element
                         next_stage_url = next_element.url
                     else:
                         next_stage_url = None
 
+                page = cast(IliasPage, page)
                 elements.extend(page.get_child_elements())
                 if description_string := page.get_description():
                     description.append(description_string)
@@ -360,6 +361,54 @@ instance's greatest bottleneck.
                 "[bright_black](scorm learning modules are not supported)"
             )
             return None
+        elif element.type == IliasElementType.LITERATURE_LIST:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](literature lists are not currently supported)"
+            )
+            return None
+        elif element.type == IliasElementType.LEARNING_MODULE_HTML:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](HTML learning modules are not supported)"
+            )
+            return None
+        elif element.type == IliasElementType.BLOG:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](blogs are not currently supported)"
+            )
+            return None
+        elif element.type == IliasElementType.DCL_RECORD_LIST:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](dcl record lists are not currently supported)"
+            )
+            return None
+        elif element.type == IliasElementType.MEDIA_POOL:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](media pools are not currently supported)"
+            )
+            return None
+        elif element.type == IliasElementType.COURSE:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](not descending into linked course, download it separately)"
+            )
+            return None
         elif element.type == IliasElementType.LEARNING_MODULE:
             return await self._handle_learning_module(element, element_path)
         elif element.type == IliasElementType.LINK:
@@ -461,10 +510,10 @@ instance's greatest bottleneck.
         if not dl:
             return
 
-        async with dl as (bar, sink):
+        async with dl as (_bar, sink):
             description = clean(insert_base_markup(description))
-            description = await self.internalize_images(description)
-            sink.file.write(description.prettify().encode("utf-8"))
+            description_tag = await self.internalize_images(description)
+            sink.file.write(cast(str, description_tag.prettify()).encode("utf-8"))
             sink.done()
 
     @anoncritical
@@ -483,7 +532,7 @@ instance's greatest bottleneck.
             async with self.session.get(export_url, allow_redirects=False) as resp:
                 # No redirect means we were authenticated
                 if hdrs.LOCATION not in resp.headers:
-                    return soupify(await resp.read()).select_one("a").get("href").strip()
+                    return soupify(await resp.read()).select_one("a").get("href").strip()  # type: ignore
                 # We are either unauthenticated or the link is not active
                 new_url = resp.headers[hdrs.LOCATION].lower()
                 if "baseclass=illinkresourcehandlergui" in new_url and "cmd=infoscreen" in new_url:
@@ -588,7 +637,7 @@ instance's greatest bottleneck.
             )
 
         async with dl as (bar, sink):
-            page = IliasPage(await self._get_page(element.url), element.url, element)
+            page = IliasPage(await self._get_page(element.url), element)
             stream_elements = page.get_child_elements()
 
             if len(stream_elements) > 1:
@@ -598,7 +647,7 @@ instance's greatest bottleneck.
                 stream_element = stream_elements[0]
 
                 # We do not have a local cache yet
-                await self._stream_from_url(stream_element.url, sink, bar, is_video=True)
+                await self._stream_from_url(stream_element, sink, bar, is_video=True)
                 add_to_report([str(self._transformer.transform(dl.path))])
                 return
 
@@ -613,7 +662,7 @@ instance's greatest bottleneck.
             async with maybe_dl as (bar, sink):
                 log.explain(f"Streaming video from real url {stream_element.url}")
                 contained_video_paths.append(str(self._transformer.transform(maybe_dl.path)))
-                await self._stream_from_url(stream_element.url, sink, bar, is_video=True)
+                await self._stream_from_url(stream_element, sink, bar, is_video=True)
 
         add_to_report(contained_video_paths)
 
@@ -635,12 +684,19 @@ instance's greatest bottleneck.
     async def _download_file(self, element: IliasPageElement, dl: DownloadToken, is_video: bool) -> None:
         assert dl  # The function is only reached when dl is not None
         async with dl as (bar, sink):
-            await self._stream_from_url(element.url, sink, bar, is_video)
+            await self._stream_from_url(element, sink, bar, is_video)
+
+    async def _stream_from_url(
+        self,
+        element: IliasPageElement,
+        sink: FileSink,
+        bar: ProgressBar,
+        is_video: bool
+    ) -> None:
+        url = element.url
 
-    async def _stream_from_url(self, url: str, sink: FileSink, bar: ProgressBar, is_video: bool) -> None:
         async def try_stream() -> bool:
             next_url = url
-
             # Normal files redirect to the magazine if we are not authenticated. As files could be HTML,
             # we can not match on the content type here. Instead, we disallow redirects and inspect the
             # new location. If we are redirected anywhere but the ILIAS 8 "sendfile" command, we assume
@@ -688,7 +744,7 @@ instance's greatest bottleneck.
         await self.authenticate(auth_id)
 
         if not await try_stream():
-            raise CrawlError("File streaming failed after authenticate()")
+            raise CrawlError(f"File streaming failed after authenticate() {element!r}")
 
     async def _handle_forum(
         self,
@@ -707,32 +763,66 @@ instance's greatest bottleneck.
 
         async with cl:
             next_stage_url = element.url
+            page = None
+
             while next_stage_url:
                 log.explain_topic(f"Parsing HTML page for {fmt_path(cl.path)}")
                 log.explain(f"URL: {next_stage_url}")
 
                 soup = await self._get_page(next_stage_url)
-                page = IliasPage(soup, next_stage_url, element)
+                page = IliasPage(soup, element)
 
                 if next := page.get_next_stage_element():
                     next_stage_url = next.url
                 else:
                     break
 
-            download_data = page.get_download_forum_data()
-            if not download_data:
-                raise CrawlWarning("Failed to extract forum data")
-            if download_data.empty:
+            forum_threads: list[tuple[IliasPageElement, bool]] = []
+            for entry in cast(IliasPage, page).get_forum_entries():
+                path = cl.path / (_sanitize_path_name(entry.name) + ".html")
+                forum_threads.append((entry, self.should_try_download(path, mtime=entry.mtime)))
+
+            # Sort the ids. The forum download will *preserve* this ordering
+            forum_threads.sort(key=lambda elem: elem[0].id())
+
+            if not forum_threads:
                 log.explain("Forum had no threads")
                 return
+
+            download_data = cast(IliasPage, page).get_download_forum_data(
+                [thread.id() for thread, download in forum_threads if download]
+            )
+            if not download_data:
+                raise CrawlWarning("Failed to extract forum data")
+
+            if not download_data.empty:
                 html = await self._post_authenticated(download_data.url, download_data.form_data)
                 elements = parse_ilias_forum_export(soupify(html))
+            else:
+                elements = []
 
-        elements.sort(key=lambda elem: elem.title)
+        # Verify that ILIAS does not change the order, as we depend on it later. Otherwise, we could not call
+        # download in the correct order, potentially messing up duplication handling.
+        expected_element_titles = [thread.name for thread, download in forum_threads if download]
+        actual_element_titles = [_sanitize_path_name(thread.name) for thread in elements]
+        if expected_element_titles != actual_element_titles:
+            raise CrawlWarning(
+                f"Forum thread order mismatch: {expected_element_titles} != {actual_element_titles}"
+            )
 
         tasks: List[Awaitable[None]] = []
-        for elem in elements:
-            tasks.append(asyncio.create_task(self._download_forum_thread(cl.path, elem)))
+        for thread, download in forum_threads:
+            if download:
+                # This only works because ILIAS keeps the order in the export
+                elem = elements.pop(0)
+                tasks.append(asyncio.create_task(self._download_forum_thread(cl.path, elem, thread)))
+            else:
+                # We only downloaded the threads we "should_try_download"ed. This can be an
+                # over-approximation and all will be fine.
+                # If we selected too few, e.g. because there was a duplicate title and the mtime of the
+                # original is newer than the update of the duplicate.
+                # This causes stale data locally, but I consider this problem acceptable right now.
+                tasks.append(asyncio.create_task(self._download_forum_thread(cl.path, thread, thread)))
 
         # And execute them
         await self.gather(tasks)
@@ -742,18 +832,22 @@ instance's greatest bottleneck.
     async def _download_forum_thread(
         self,
         parent_path: PurePath,
-        element: IliasForumThread,
+        thread: Union[IliasForumThread, IliasPageElement],
+        element: IliasPageElement
     ) -> None:
-        path = parent_path / (_sanitize_path_name(element.title) + ".html")
-        maybe_dl = await self.download(path, mtime=element.mtime)
-        if not maybe_dl:
+        path = parent_path / (_sanitize_path_name(thread.name) + ".html")
+        maybe_dl = await self.download(path, mtime=thread.mtime)
+        if not maybe_dl or not isinstance(thread, IliasForumThread):
             return
 
         async with maybe_dl as (bar, sink):
-            content = "<!DOCTYPE html>\n"
-            content += element.title_tag.prettify()
-            content += element.content_tag.prettify()
-            sink.file.write(content.encode("utf-8"))
+            rendered = forum_thread_template(
+                thread.name,
+                element.url,
+                thread.name_tag,
+                await self.internalize_images(thread.content_tag)
+            )
+            sink.file.write(rendered.encode("utf-8"))
             sink.done()
 
     async def _handle_learning_module(
@@ -777,7 +871,7 @@ instance's greatest bottleneck.
             log.explain_topic(f"Parsing initial HTML page for {fmt_path(cl.path)}")
             log.explain(f"URL: {element.url}")
             soup = await self._get_page(element.url)
-            page = IliasPage(soup, element.url, element)
+            page = IliasPage(soup, element)
             if next := page.get_learning_module_data():
                 elements.extend(await self._crawl_learning_module_direction(
                     cl.path, next.previous_url, "left", element
@@ -820,7 +914,7 @@ instance's greatest bottleneck.
             log.explain_topic(f"Parsing HTML page for {fmt_path(path)} ({dir}-{counter})")
             log.explain(f"URL: {next_element_url}")
             soup = await self._get_page(next_element_url)
-            page = IliasPage(soup, next_element_url, parent_element)
+            page = IliasPage(soup, parent_element)
             if next := page.get_learning_module_data():
                 elements.append(next)
                 if dir == "left":
@@ -851,13 +945,13 @@ instance's greatest bottleneck.
         if prev:
             prev_p = self._transformer.transform(parent_path / (_sanitize_path_name(prev) + ".html"))
             if prev_p:
-                prev = os.path.relpath(prev_p, my_path.parent)
+                prev = cast(str, os.path.relpath(prev_p, my_path.parent))
             else:
                 prev = None
         if next:
             next_p = self._transformer.transform(parent_path / (_sanitize_path_name(next) + ".html"))
             if next_p:
-                next = os.path.relpath(next_p, my_path.parent)
+                next = cast(str, os.path.relpath(next_p, my_path.parent))
             else:
                 next = None
 
@@ -877,15 +971,15 @@ instance's greatest bottleneck.
                 continue
             if elem.name == "img":
                 if src := elem.attrs.get("src", None):
-                    url = urljoin(self._base_url, src)
+                    url = urljoin(self._base_url, cast(str, src))
                     if not url.startswith(self._base_url):
                         continue
                     log.explain(f"Internalizing {url!r}")
                     img = await self._get_authenticated(url)
                     elem.attrs["src"] = "data:;base64," + base64.b64encode(img).decode()
-            if elem.name == "iframe" and elem.attrs.get("src", "").startswith("//"):
+            if elem.name == "iframe" and cast(str, elem.attrs.get("src", "")).startswith("//"):
                 # For unknown reasons the protocol seems to be stripped.
-                elem.attrs["src"] = "https:" + elem.attrs["src"]
+                elem.attrs["src"] = "https:" + cast(str, elem.attrs["src"])
         return tag
 
     def _ensure_not_seen(self, element: IliasPageElement, parent_path: PurePath) -> None:
@@ -897,10 +991,10 @@ instance's greatest bottleneck.
             )
         self._visited_urls[element.url] = parent_path
 
-    async def _get_page(self, url: str, root_page_allowed: bool = False) -> BeautifulSoup:
+    async def _get_page(self, url: str, root_page_allowed: bool = False) -> IliasSoup:
         auth_id = await self._current_auth_id()
         async with self.session.get(url) as request:
-            soup = soupify(await request.read())
+            soup = IliasSoup(soupify(await request.read()), str(request.url))
             if IliasPage.is_logged_in(soup):
                 return self._verify_page(soup, url, root_page_allowed)
 
@@ -909,13 +1003,13 @@ instance's greatest bottleneck.
 
         # Retry once after authenticating. If this fails, we will die.
         async with self.session.get(url) as request:
-            soup = soupify(await request.read())
+            soup = IliasSoup(soupify(await request.read()), str(request.url))
             if IliasPage.is_logged_in(soup):
                 return self._verify_page(soup, url, root_page_allowed)
         raise CrawlError(f"get_page failed even after authenticating on {url!r}")
 
     @staticmethod
-    def _verify_page(soup: BeautifulSoup, url: str, root_page_allowed: bool) -> BeautifulSoup:
+    def _verify_page(soup: IliasSoup, url: str, root_page_allowed: bool) -> IliasSoup:
         if IliasPage.is_root_page(soup) and not root_page_allowed:
             raise CrawlError(
                 "Unexpectedly encountered ILIAS root page. "
@@ -979,11 +1073,11 @@ instance's greatest bottleneck.
             async with self.session.get(urljoin(self._base_url, "/login.php"), params=params) as request:
                 login_page = soupify(await request.read())
 
-            login_form = login_page.find("form", attrs={"name": "formlogin"})
+            login_form = cast(Optional[Tag], login_page.find("form", attrs={"name": "formlogin"}))
             if login_form is None:
                 raise CrawlError("Could not find the login form! Specified client id might be invalid.")
 
-            login_url = login_form.attrs.get("action")
+            login_url = cast(Optional[str], login_form.attrs.get("action"))
             if login_url is None:
                 raise CrawlError("Could not find the action URL in the login form!")
 
@@ -997,34 +1091,6 @@ instance's greatest bottleneck.
 
             # do the actual login
             async with self.session.post(urljoin(self._base_url, login_url), data=login_data) as request:
-                soup = soupify(await request.read())
-                if not self._is_logged_in(soup):
+                soup = IliasSoup(soupify(await request.read()), str(request.url))
+                if not IliasPage.is_logged_in(soup):
                     self._auth.invalidate_credentials()
-
-    @staticmethod
-    def _is_logged_in(soup: BeautifulSoup) -> bool:
-        # Normal ILIAS pages
-        mainbar: Optional[Tag] = soup.find(class_="il-maincontrols-metabar")
-        if mainbar is not None:
-            login_button = mainbar.find(attrs={"href": lambda x: x and "login.php" in x})
-            shib_login = soup.find(id="button_shib_login")
-            return not login_button and not shib_login
-
-        # Personal Desktop
-        if soup.find("a", attrs={"href": lambda x: x and "block_type=pditems" in x}):
-            return True
-
-        # Video listing embeds do not have complete ILIAS html. Try to match them by
-        # their video listing table
-        video_table = soup.find(
-            recursive=True,
-            name="table",
-            attrs={"id": lambda x: x is not None and x.startswith("tbl_xoct")}
-        )
-        if video_table is not None:
-            return True
-        # The individual video player wrapper page has nothing of the above.
-        # Match it by its playerContainer.
-        if soup.select_one("#playerContainer") is not None:
-            return True
-        return False

PFERD/crawl/ilias/shibboleth_login.py

@@ -1,8 +1,8 @@
-from typing import Any, Optional
+from typing import Any, Optional, cast
 
 import aiohttp
 import yarl
-from bs4 import BeautifulSoup
+from bs4 import BeautifulSoup, Tag
 
 from ...auth import Authenticator, TfaAuthenticator
 from ...logging import log
@@ -48,8 +48,8 @@ class ShibbolethLogin:
         while not self._login_successful(soup):
             # Searching the form here so that this fails before asking for
             # credentials rather than after asking.
-            form = soup.find("form", {"method": "post"})
-            action = form["action"]
+            form = cast(Tag, soup.find("form", {"method": "post"}))
+            action = cast(str, form["action"])
 
             # Equivalent: Enter credentials in
             # https://idp.scc.kit.edu/idp/profile/SAML2/Redirect/SSO
@@ -59,9 +59,10 @@ class ShibbolethLogin:
                 "_eventId_proceed": "",
                 "j_username": username,
                 "j_password": password,
+                "fudis_web_authn_assertion_input": "",
             }
             if csrf_token_input := form.find("input", {"name": "csrf_token"}):
-                data["csrf_token"] = csrf_token_input["value"]
+                data["csrf_token"] = csrf_token_input["value"]  # type: ignore
             soup = await _post(sess, url, data)
 
             if soup.find(id="attributeRelease"):
@@ -78,14 +79,14 @@ class ShibbolethLogin:
 
         # Equivalent: Being redirected via JS automatically
         # (or clicking "Continue" if you have JS disabled)
-        relay_state = soup.find("input", {"name": "RelayState"})
-        saml_response = soup.find("input", {"name": "SAMLResponse"})
-        url = form = soup.find("form", {"method": "post"})["action"]
+        relay_state = cast(Tag, soup.find("input", {"name": "RelayState"}))
+        saml_response = cast(Tag, soup.find("input", {"name": "SAMLResponse"}))
+        url = form = soup.find("form", {"method": "post"})["action"]  # type: ignore
         data = {  # using the info obtained in the while loop above
-            "RelayState": relay_state["value"],
-            "SAMLResponse": saml_response["value"],
+            "RelayState": cast(str, relay_state["value"]),
+            "SAMLResponse": cast(str, saml_response["value"]),
         }
-        await sess.post(url, data=data)
+        await sess.post(cast(str, url), data=data)
 
     async def _authenticate_tfa(
         self, session: aiohttp.ClientSession, soup: BeautifulSoup, shib_url: yarl.URL
@@ -97,8 +98,8 @@ class ShibbolethLogin:
 
         # Searching the form here so that this fails before asking for
         # credentials rather than after asking.
-        form = soup.find("form", {"method": "post"})
-        action = form["action"]
+        form = cast(Tag, soup.find("form", {"method": "post"}))
+        action = cast(str, form["action"])
 
         # Equivalent: Enter token in
         # https://idp.scc.kit.edu/idp/profile/SAML2/Redirect/SSO
@@ -106,10 +107,10 @@ class ShibbolethLogin:
         username, password = await self._auth.credentials()
         data = {
             "_eventId_proceed": "",
-            "j_tokenNumber": tfa_token,
+            "fudis_otp_input": tfa_token,
         }
         if csrf_token_input := form.find("input", {"name": "csrf_token"}):
-            data["csrf_token"] = csrf_token_input["value"]
+            data["csrf_token"] = csrf_token_input["value"]  # type: ignore
         return await _post(session, url, data)
 
     @staticmethod
@@ -120,7 +121,7 @@ class ShibbolethLogin:
 
     @staticmethod
     def _tfa_required(soup: BeautifulSoup) -> bool:
-        return soup.find(id="j_tokenNumber") is not None
+        return soup.find(id="fudiscr-form") is not None
 
 
 async def _post(session: aiohttp.ClientSession, url: str, data: Any) -> BeautifulSoup:

PFERD/crawl/kit_ipd_crawler.py

@@ -3,7 +3,7 @@ import re
 from dataclasses import dataclass
 from datetime import datetime
 from pathlib import PurePath
-from typing import Any, Awaitable, Generator, Iterable, List, Optional, Pattern, Tuple, Union
+from typing import Any, Awaitable, Generator, Iterable, List, Optional, Pattern, Tuple, Union, cast
 from urllib.parse import urljoin
 
 from bs4 import BeautifulSoup, Tag
@@ -156,11 +156,11 @@ class KitIpdCrawler(HttpCrawler):
         name = os.path.basename(url)
         return KitIpdFile(name, url)
 
-    def _find_file_links(self, tag: Union[Tag, BeautifulSoup]) -> List[Tag]:
-        return tag.findAll(name="a", attrs={"href": self._file_regex})
+    def _find_file_links(self, tag: Union[Tag, BeautifulSoup]) -> list[Tag]:
+        return cast(list[Tag], tag.find_all(name="a", attrs={"href": self._file_regex}))
 
     def _abs_url_from_link(self, url: str, link_tag: Tag) -> str:
-        return urljoin(url, link_tag.get("href"))
+        return urljoin(url, cast(str, link_tag.get("href")))
 
     async def _stream_from_url(self, url: str, path: PurePath, sink: FileSink, bar: ProgressBar) -> None:
         async with self.session.get(url, allow_redirects=False) as resp:

PFERD/logging.py

@@ -1,9 +1,8 @@
 import asyncio
 import sys
 import traceback
-from contextlib import asynccontextmanager, contextmanager
-# TODO In Python 3.9 and above, ContextManager is deprecated
-from typing import AsyncIterator, ContextManager, Iterator, List, Optional
+from contextlib import AbstractContextManager, asynccontextmanager, contextmanager
+from typing import AsyncIterator, Iterator, List, Optional
 
 from rich.console import Console, Group
 from rich.live import Live
@@ -261,7 +260,7 @@ directly or as a GitHub issue: https://github.com/Garmelon/PFERD/issues/new
             action: str,
             text: str,
             total: Optional[float] = None,
-    ) -> ContextManager[ProgressBar]:
+    ) -> AbstractContextManager[ProgressBar]:
         """
         Allows markup in the "style" argument which will be applied to the
         "action" string.
@@ -277,7 +276,7 @@ directly or as a GitHub issue: https://github.com/Garmelon/PFERD/issues/new
             action: str,
             text: str,
             total: Optional[float] = None,
-    ) -> ContextManager[ProgressBar]:
+    ) -> AbstractContextManager[ProgressBar]:
         """
         Allows markup in the "style" argument which will be applied to the
         "action" string.

PFERD/output_dir.py

@@ -371,6 +371,22 @@ class OutputDirectory:
 
         raise OutputDirError("Failed to create temporary file")
 
+    def should_try_download(
+        self,
+        path: PurePath,
+        *,
+        etag_differs: Optional[bool] = None,
+        mtime: Optional[datetime] = None,
+        redownload: Optional[Redownload] = None,
+        on_conflict: Optional[OnConflict] = None,
+    ) -> bool:
+        heuristics = Heuristics(etag_differs, mtime)
+        redownload = self._redownload if redownload is None else redownload
+        on_conflict = self._on_conflict if on_conflict is None else on_conflict
+        local_path = self.resolve(path)
+
+        return self._should_download(local_path, heuristics, redownload, on_conflict)
+
     async def download(
             self,
             remote_path: PurePath,

PFERD/pferd.py

@@ -1,6 +1,5 @@
 from pathlib import Path, PurePath
 from typing import Dict, List, Optional
-from urllib.parse import quote
 
 from rich.markup import escape
 
@@ -171,9 +170,7 @@ class Pferd:
 
             def fmt_path_link(relative_path: PurePath) -> str:
                 # We need to URL-encode the path because it might contain spaces or special characters
-                absolute_path = str(crawler.output_dir.resolve(relative_path).absolute())
-                absolute_path = absolute_path.replace("\\\\?\\", "")
-                link = f"file://{quote(absolute_path)}"
+                link = crawler.output_dir.resolve(relative_path).absolute().as_uri()
                 return f"[link={link}]{fmt_path(relative_path)}[/link]"
 
             something_changed = False

PFERD/report.py

@@ -34,15 +34,6 @@ class MarkConflictError(Exception):
         self.collides_with = collides_with
 
 
-# TODO Use PurePath.is_relative_to when updating to 3.9
-def is_relative_to(a: PurePath, b: PurePath) -> bool:
-    try:
-        a.relative_to(b)
-        return True
-    except ValueError:
-        return False
-
-
 class Report:
     """
     A report of a synchronization. Includes all files found by the crawler, as
@@ -173,7 +164,7 @@ class Report:
             if path == other:
                 raise MarkDuplicateError(path)
 
-            if is_relative_to(path, other) or is_relative_to(other, path):
+            if path.is_relative_to(other) or other.is_relative_to(path):
                 raise MarkConflictError(path, other)
 
         self.known_files.add(path)

PFERD/version.py

@@ -1,2 +1,2 @@
 NAME = "PFERD"
-VERSION = "3.7.0"
+VERSION = "3.8.0"

README.md

@@ -17,7 +17,7 @@ Binaries for Linux, Windows and Mac can be downloaded directly from the
 
 ### With pip
 
-Ensure you have at least Python 3.9 installed. Run the following command to
+Ensure you have at least Python 3.11 installed. Run the following command to
 install PFERD or upgrade it to the latest version:
 
 ```

flake.lock

@@ -2,16 +2,16 @@
   "nodes": {
     "nixpkgs": {
       "locked": {
-        "lastModified": 1708979614,
-        "narHash": "sha256-FWLWmYojIg6TeqxSnHkKpHu5SGnFP5um1uUjH+wRV6g=",
+        "lastModified": 1744440957,
+        "narHash": "sha256-FHlSkNqFmPxPJvy+6fNLaNeWnF1lZSgqVCl/eWaJRc4=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "b7ee09cf5614b02d289cd86fcfa6f24d4e078c2a",
+        "rev": "26d499fc9f1d567283d5d56fcf367edd815dba1d",
         "type": "github"
       },
       "original": {
         "owner": "NixOS",
-        "ref": "nixos-23.11",
+        "ref": "nixos-24.11",
         "repo": "nixpkgs",
         "type": "github"
       }

flake.nix

@@ -2,7 +2,7 @@
   description = "Tool for downloading course-related files from ILIAS";
 
   inputs = {
-    nixpkgs.url = "github:NixOS/nixpkgs/nixos-23.11";
+    nixpkgs.url = "github:NixOS/nixpkgs/nixos-24.11";
   };
 
   outputs = { self, nixpkgs }:

pyproject.toml

@@ -12,7 +12,7 @@ dependencies = [
   "certifi>=2021.10.8"
 ]
 dynamic = ["version"]
-requires-python = ">=3.9"
+requires-python = ">=3.11"
 
 [project.scripts]
 pferd = "PFERD.__main__:main"