From 269276003d599c3d3f5b1b1ed7eafe694d2ea964 Mon Sep 17 00:00:00 2001
From: "Thomas G. Lopes" <thomasgl@pm.me>
Date: Fri, 13 Mar 2026 12:46:08 +0000
Subject: [PATCH] preserve markdown formatting in scraped changelog

---
 .gitea/scripts/update-appimage-nix.sh | 157 +++++++++++++++++++++-----
 1 file changed, 127 insertions(+), 30 deletions(-)

diff --git a/.gitea/scripts/update-appimage-nix.sh b/.gitea/scripts/update-appimage-nix.sh
index 8a78795..affc347 100644
--- a/.gitea/scripts/update-appimage-nix.sh
+++ b/.gitea/scripts/update-appimage-nix.sh
@@ -22,7 +22,7 @@ release_tag_template="${release_tag_template//$'\r'/}"
 current_version=$($PYTHON_BIN - <<'PY'
 import re
 import os
-p=os.environ['FILE']
+p=os.environ["FILE"]
 s=open(p).read()
 m=re.search(r'version\s*=\s*"([^"]+)";', s)
 print(m.group(1) if m else "")
@@ -53,11 +53,11 @@ export NEW_HASH="$new_hash"
 "$PYTHON_BIN" - <<'PY'
 import os
 import re
-p=os.environ['FILE']
+p=os.environ["FILE"]
 s=open(p).read()
 s=re.sub(r'version\s*=\s*"[^"]+"', f'version = "{os.environ["LATEST_VERSION"]}"', s, count=1)
 s=re.sub(r'hash\s*=\s*"[^"]+"', f'hash = "{os.environ["NEW_HASH"]}"', s, count=1)
-open(p,'w').write(s)
+open(p,"w").write(s)
 PY
 
 echo "updated=true" >> "$GITHUB_OUTPUT"
@@ -76,7 +76,8 @@ release_html=$(curl -fsSL "$release_url" || true)
 release_notes=""
 if [ -n "$release_html" ]; then
   release_notes=$(printf '%s' "$release_html" | "$PYTHON_BIN" -c '
-import re, sys
+import re
+import sys
 from html.parser import HTMLParser
 
 html = sys.stdin.read()
@@ -88,46 +89,142 @@ if not m:
 
 fragment = m.group(1)
 
-class Extractor(HTMLParser):
+class MdExtractor(HTMLParser):
   def __init__(self):
     super().__init__()
     self.out = []
-    self.in_code = False
+    self.list_depth = 0
+    self.in_pre = False
+    self.in_code_inline = False
+    self.link_stack = []
+
+  def _append(self, text):
+    self.out.append(text)
+
+  def _ensure_newline(self):
+    if not self.out:
+      return
+    if not self.out[-1].endswith("\n"):
+      self.out.append("\n")
+
+  def _ensure_blank_line(self):
+    if not self.out:
+      return
+    joined = "".join(self.out)
+    if joined.endswith("\n\n"):
+      return
+    if joined.endswith("\n"):
+      self.out.append("\n")
+    else:
+      self.out.append("\n\n")
 
   def handle_starttag(self, tag, attrs):
-    if tag in ("br",):
-      self.out.append("\n")
-    elif tag in ("p", "div", "h1", "h2", "h3", "h4", "h5", "h6", "li"):
-      if self.out and not self.out[-1].endswith("\n"):
-        self.out.append("\n")
-    elif tag in ("pre",):
-      if self.out and not self.out[-1].endswith("\n"):
-        self.out.append("\n")
-      self.out.append("```\n")
-      self.in_code = True
-    elif tag in ("code",) and not self.in_code:
-      self.out.append("`")
+    attrs_d = dict(attrs)
+
+    if tag in ("h1", "h2", "h3", "h4", "h5", "h6"):
+      self._ensure_blank_line()
+      level = int(tag[1])
+      self._append("#" * level + " ")
+      return
+
+    if tag in ("p", "div"):
+      self._ensure_blank_line()
+      return
+
+    if tag in ("ul", "ol"):
+      self._ensure_blank_line()
+      self.list_depth += 1
+      return
+
+    if tag == "li":
+      self._ensure_newline()
+      indent = "  " * max(self.list_depth - 1, 0)
+      self._append(f"{indent}- ")
+      return
+
+    if tag == "br":
+      self._append("\n")
+      return
+
+    if tag == "pre":
+      self._ensure_blank_line()
+      self._append("```\n")
+      self.in_pre = True
+      return
+
+    if tag == "code":
+      if not self.in_pre:
+        self._append("`")
+        self.in_code_inline = True
+      return
+
+    if tag == "a":
+      href = attrs_d.get("href", "")
+      self.link_stack.append(href)
+      self._append("[")
+      return
+
+    if tag in ("strong", "b"):
+      self._append("**")
+      return
+
+    if tag in ("em", "i"):
+      self._append("*")
+      return
 
   def handle_endtag(self, tag):
-    if tag in ("p", "div", "li", "h1", "h2", "h3", "h4", "h5", "h6"):
-      if not self.out or not self.out[-1].endswith("\n"):
-        self.out.append("\n")
-    elif tag in ("pre",):
-      if not self.out or not self.out[-1].endswith("\n"):
-        self.out.append("\n")
-      self.out.append("```\n")
-      self.in_code = False
-    elif tag in ("code",) and not self.in_code:
-      self.out.append("`")
+    if tag in ("h1", "h2", "h3", "h4", "h5", "h6", "p", "div"):
+      self._ensure_blank_line()
+      return
+
+    if tag in ("ul", "ol"):
+      self.list_depth = max(self.list_depth - 1, 0)
+      self._ensure_blank_line()
+      return
+
+    if tag == "li":
+      self._ensure_newline()
+      return
+
+    if tag == "pre":
+      self._ensure_newline()
+      self._append("```\n\n")
+      self.in_pre = False
+      return
+
+    if tag == "code":
+      if not self.in_pre and self.in_code_inline:
+        self._append("`")
+        self.in_code_inline = False
+      return
+
+    if tag == "a":
+      href = self.link_stack.pop() if self.link_stack else ""
+      if href:
+        self._append("]("
+          + href
+          + ")")
+      else:
+        self._append("]")
+      return
+
+    if tag in ("strong", "b"):
+      self._append("**")
+      return
+
+    if tag in ("em", "i"):
+      self._append("*")
+      return
 
   def handle_data(self, data):
     if data:
-      self.out.append(data)
+      self._append(data)
 
-parser = Extractor()
+parser = MdExtractor()
 parser.feed(fragment)
 text = "".join(parser.out)
 text = re.sub(r"\n{3,}", "\n\n", text)
+text = re.sub(r"[ \t]+\n", "\n", text)
 print(text.strip())
 ' || true)
 else