From 50b50513c6d8bb01200104633d7ce312e17a0ba7 Mon Sep 17 00:00:00 2001
From: I-Al-Istannen <i-al-istannen@users.noreply.github.com>
Date: Tue, 29 Aug 2023 13:51:19 +0200
Subject: [PATCH] Ignore SCORM learning modules

---
 CHANGELOG.md                               |  1 +
 PFERD/crawl/ilias/kit_ilias_html.py        |  7 +++++++
 PFERD/crawl/ilias/kit_ilias_web_crawler.py | 10 +++++++++-
 3 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 3f318b2..47df846 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -30,6 +30,7 @@ ambiguous situations.
 - Remove size suffix for files in content pages
 - Sanitize ascii control characters on Windows
 - Crawling of paginated past meetings
+- Ignore SCORM learning modules
 
 ### Added
 - `no-delete-prompt-override` conflict resolution strategy
diff --git a/PFERD/crawl/ilias/kit_ilias_html.py b/PFERD/crawl/ilias/kit_ilias_html.py
index 5a94a0b..2c37816 100644
--- a/PFERD/crawl/ilias/kit_ilias_html.py
+++ b/PFERD/crawl/ilias/kit_ilias_html.py
@@ -27,6 +27,7 @@ class IliasElementType(Enum):
     BOOKING = "booking"
     MEETING = "meeting"
     SURVEY = "survey"
+    SCORM_LEARNING_MODULE = "scorm_learning_module"
     MEDIACAST_VIDEO_FOLDER = "mediacast_video_folder"
     MEDIACAST_VIDEO = "mediacast_video"
     OPENCAST_VIDEO = "opencast_video"
@@ -953,6 +954,9 @@ class IliasPage:
         if "baseClass=ilMediaCastHandlerGUI" in parsed_url.query:
             return IliasElementType.MEDIACAST_VIDEO_FOLDER
 
+        if "baseClass=ilSAHSPresentationGUI" in parsed_url.query:
+            return IliasElementType.SCORM_LEARNING_MODULE
+
         # Booking and Meeting can not be detected based on the link. They do have a ref_id though, so
         # try to guess it from the image.
 
@@ -1031,6 +1035,9 @@ class IliasPage:
         if str(img_tag["src"]).endswith("icon_mcst.svg"):
             return IliasElementType.MEDIACAST_VIDEO_FOLDER
 
+        if str(img_tag["src"]).endswith("icon_sahs.svg"):
+            return IliasElementType.SCORM_LEARNING_MODULE
+
         return IliasElementType.FOLDER
 
     @staticmethod
diff --git a/PFERD/crawl/ilias/kit_ilias_web_crawler.py b/PFERD/crawl/ilias/kit_ilias_web_crawler.py
index 4f6cc74..d5f6809 100644
--- a/PFERD/crawl/ilias/kit_ilias_web_crawler.py
+++ b/PFERD/crawl/ilias/kit_ilias_web_crawler.py
@@ -403,6 +403,14 @@ instance's greatest bottleneck.
                 "[bright_black](surveys contain no relevant data)"
             )
             return None
+        elif element.type == IliasElementType.SCORM_LEARNING_MODULE:
+            log.status(
+                "[bold bright_black]",
+                "Ignored",
+                fmt_path(element_path),
+                "[bright_black](scorm learning modules are not supported)"
+            )
+            return None
         elif element.type == IliasElementType.LEARNING_MODULE:
             return await self._handle_learning_module(element, element_path)
         elif element.type == IliasElementType.LINK:
@@ -897,7 +905,7 @@ instance's greatest bottleneck.
             soup = soupify(await request.read())
             if self._is_logged_in(soup):
                 return self._verify_page(soup, url, root_page_allowed)
-        raise CrawlError("get_page failed even after authenticating")
+        raise CrawlError(f"get_page failed even after authenticating on {url!r}")
 
     def _verify_page(self, soup: BeautifulSoup, url: str, root_page_allowed: bool) -> BeautifulSoup:
         if IliasPage.is_root_page(soup) and not root_page_allowed: