From f6c713d62198f9970d81b56dbf86fcb04f760629 Mon Sep 17 00:00:00 2001 From: I-Al-Istannen Date: Sun, 19 Oct 2025 15:08:21 +0200 Subject: [PATCH] Fix mypy errors --- PFERD/crawl/ilias/ilias_web_crawler.py | 2 -- PFERD/crawl/ilias/kit_ilias_html.py | 4 ++-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/PFERD/crawl/ilias/ilias_web_crawler.py b/PFERD/crawl/ilias/ilias_web_crawler.py index 2eb8e9c..ee1de9c 100644 --- a/PFERD/crawl/ilias/ilias_web_crawler.py +++ b/PFERD/crawl/ilias/ilias_web_crawler.py @@ -985,8 +985,6 @@ instance's greatest bottleneck. """ log.explain_topic("Internalizing images") for elem in tag.find_all(recursive=True): - if not isinstance(elem, Tag): - continue if elem.name == "img": if src := elem.attrs.get("src", None): url = urljoin(self._base_url, cast(str, src)) diff --git a/PFERD/crawl/ilias/kit_ilias_html.py b/PFERD/crawl/ilias/kit_ilias_html.py index 70ec3d7..5b88e8d 100644 --- a/PFERD/crawl/ilias/kit_ilias_html.py +++ b/PFERD/crawl/ilias/kit_ilias_html.py @@ -435,7 +435,7 @@ class IliasPage: return None def get_description(self) -> Optional[BeautifulSoup]: - def is_interesting_class(name: str) -> bool: + def is_interesting_class(name: str | None) -> bool: return name in [ "ilCOPageSection", "ilc_Paragraph", "ilc_va_ihcap_VAccordIHeadCap", "ilc_va_ihcap_AccordIHeadCap", "ilc_media_cont_MediaContainer" @@ -1243,7 +1243,7 @@ class IliasPage: # In a series of divs. # Find the parent containing all those divs, so we can filter our what we need properties_parent = cast(Tag, cast(Tag, link_element.find_parent( - "div", {"class": lambda x: "il_ContainerListItem" in x} + "div", {"class": lambda x: x is not None and "il_ContainerListItem" in x} )).select_one(".il_ItemProperties")) # The first one is always the filetype file_type = cast(Tag, properties_parent.select_one("span.il_ItemProperty")).get_text().strip()