From 292d71edc52009fff973784134799d05d7a7141e Mon Sep 17 00:00:00 2001
From: Lars Noack <lars@webcontact.de>
Date: Wed, 16 Apr 2025 15:26:43 +0200
Subject: [PATCH] feat: added proper content

---
 .gitignore    |  3 ++-
 stsg/build.py | 54 ++++++++++++++++++++++++++++++++++++++++++++++++---
 2 files changed, 53 insertions(+), 4 deletions(-)

diff --git a/.gitignore b/.gitignore
index 961b463..ede09e9 100644
--- a/.gitignore
+++ b/.gitignore
@@ -173,4 +173,5 @@ cython_debug/
 # PyPI configuration file
 .pypirc
 
-dist
\ No newline at end of file
+dist
+context.json
\ No newline at end of file
diff --git a/stsg/build.py b/stsg/build.py
index fe9f615..feac937 100644
--- a/stsg/build.py
+++ b/stsg/build.py
@@ -30,6 +30,44 @@ def get_first_header_content(content, fallback: str = ""):
     return fallback
 
 
+def shorten_text_and_clean(html_string, max_length=config.formatting.article_preview_length):
+    soup = BeautifulSoup(html_string, 'html.parser')
+
+    # Keep track of total characters added
+    total_chars = 0
+    finished = False
+
+    # Function to recursively trim and clean text
+    def process_element(element):
+        nonlocal total_chars, finished
+
+        for child in list(element.children):
+            if finished:
+                child.extract()
+                continue
+
+            if isinstance(child, str):
+                remaining = max_length - total_chars
+                if remaining <= 0:
+                    child.extract()
+                    finished = True
+                elif len(child) > remaining:
+                    child.replace_with(child[:remaining] + '...')
+                    total_chars = max_length
+                    finished = True
+                else:
+                    total_chars += len(child)
+            elif hasattr(child, 'children'):
+                process_element(child)
+                # Remove empty tags
+                if not child.text.strip():
+                    child.decompose()
+
+    process_element(soup)
+
+    return str(soup)
+
+
 def stem_to_language_code(stem: str) -> str:
     language_code = stem.lower().replace("-", "_")
     
@@ -98,8 +136,16 @@ class ArticleTranslation:
         
     def __init_context__(self):
         self.context["meta"] = self.article.context_meta
+        self.context["url"] = self.url
         self.context["language"] = LANGUAGES[self.language_code]
 
+        html_content = self.file.read_text()
+        if self.file.suffix == ".md":
+            html_content = markdown.markdown(html_content)
+
+        self.context["content"] = html_content
+        self.context["preview"] = shorten_text_and_clean(html_string=html_content)
+
     def build(self):
         self.dist_path.mkdir(parents=True, exist_ok=True)
 
@@ -110,8 +156,6 @@ class ArticleTranslation:
         r = {
             "article_content": self.article_content,
             "article_preview": self.article_preview,
-            "article_url": self.url,
-            "article_slug": self.article.slug,
             "article_title": self.title,
         }
 
@@ -269,11 +313,15 @@ def build():
 
     logger.info("building page tree...")
     tree = Article(directory=Path(config.setup.source_directory, "articles"), is_root=True)
+    
+    print(ARTICLE_LAKE.keys())
+    
     logger.info("compiling tree context...")
     tree.__init_context__()
 
     import json
-    print(json.dumps(tree.context, indent=4))
+    with Path("context.json").open("w") as f:
+        json.dump(tree.context, f, indent=4)
 
     # build article reverence values
     for article_overview in ARTICLE_LAKE.values():