FossilRepo

Fix wiki markdown/HTML detection, proper rendering for both formats - Smart content detection: check for markdown patterns (# headings, code fences, [ref] links) BEFORE any transformations - Markdown pages get proper rendering through Python markdown library - Fossil wiki/HTML pages get link conversion and pass through as HTML - Fossil [/path|text] links converted to markdown links in MD mode, HTML anchors in HTML mode - <verbatim> blocks converted to code fences (MD) or <pre> (HTML) - Fixes Fossil-NG and other markdown wiki pages rendering as raw text

lmata 2026-04-06 13:57 trunk
Commit 50843fb4d000506a1a3eee51fd4eca43fe23c610f0e85e7e1670f79eabd328a3
Parent 0fb1c2724e23568…
1 file changed +34 -9
~ fossil/views.py
M fossil/views.py
+34 -9
--- fossil/views.py
+++ fossil/views.py
@@ -22,24 +22,49 @@
2222
 - Markdown (newer pages)
2323
 """
2424
 if not content:
2525
 return ""
2626
27
- # Convert Fossil-specific syntax
28
- # [/path|text] -> <a href="/path">text</a>
27
+ # Detect format from the raw content BEFORE any transformations
28
+ is_markdown = _is_markdown(content)
29
+
30
+ if is_markdown:
31
+ # Markdown: convert Fossil [/path|text] links to markdown links first
32
+ content = re.sub(r"\[(/[^|\]]+)\|([^\]]+)\]", r"[\2](\1)", content)
33
+ content = re.sub(r"<verbatim>(.*?)</verbatim>", r"```\n\1\n```", content, flags=re.DOTALL)
34
+ return md.markdown(content, extensions=["fenced_code", "tables", "toc"])
35
+
36
+ # Fossil wiki / HTML: convert Fossil-specific syntax to HTML
2937
 content = re.sub(r"\[(/[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
30
- # [url|text] -> <a href="url">text</a>
3138
 content = re.sub(r"\[(https?://[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
32
- # <verbatim>...</verbatim> -> <pre><code>...</code></pre>
3339
 content = re.sub(r"<verbatim>(.*?)</verbatim>", r"<pre><code>\1</code></pre>", content, flags=re.DOTALL)
40
+ return content
3441
35
- # If content looks like it has HTML tags, treat as HTML (Fossil wiki)
36
- if re.search(r"<(h[1-6]|p|ol|ul|li|div|table|pre|a|br)\b", content, re.IGNORECASE):
37
- return content
3842
39
- # Otherwise try markdown
40
- return md.markdown(content, extensions=["fenced_code", "tables", "toc"])
43
+def _is_markdown(content: str) -> bool:
44
+ """Detect if content is Markdown vs Fossil wiki/HTML.
45
+
46
+ Heuristic: if the content starts with markdown-style headings (#),
47
+ or has significant markdown syntax patterns, treat as markdown.
48
+ """
49
+ stripped = content.strip()
50
+ # Starts with markdown heading
51
+ if re.match(r"^#{1,6}\s", stripped):
52
+ return True
53
+ # Has multiple markdown headings
54
+ if len(re.findall(r"^#{1,6}\s", stripped, re.MULTILINE)) >= 2:
55
+ return True
56
+ # Has markdown link references [text][ref]
57
+ if re.search(r"\[.+\]\[.+\]", stripped):
58
+ return True
59
+ # Has markdown code fences
60
+ if "```" in stripped:
61
+ return True
62
+ # Starts with HTML block element — it's Fossil wiki/HTML
63
+ if re.match(r"<(h[1-6]|p|ol|ul|div|table)\b", stripped, re.IGNORECASE):
64
+ return False
65
+ return False
4166
4267
4368
 def _get_repo_and_reader(slug):
4469
 """Return (project, fossil_repo, reader) or raise 404."""
4570
 project = get_object_or_404(Project, slug=slug, deleted_at__isnull=True)
4671
--- fossil/views.py
+++ fossil/views.py
@@ -22,24 +22,49 @@
22 - Markdown (newer pages)
23 """
24 if not content:
25 return ""
26
27 # Convert Fossil-specific syntax
28 # [/path|text] -> <a href="/path">text</a>
 
 
 
 
 
 
 
 
29 content = re.sub(r"\[(/[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
30 # [url|text] -> <a href="url">text</a>
31 content = re.sub(r"\[(https?://[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
32 # <verbatim>...</verbatim> -> <pre><code>...</code></pre>
33 content = re.sub(r"<verbatim>(.*?)</verbatim>", r"<pre><code>\1</code></pre>", content, flags=re.DOTALL)
 
34
35 # If content looks like it has HTML tags, treat as HTML (Fossil wiki)
36 if re.search(r"<(h[1-6]|p|ol|ul|li|div|table|pre|a|br)\b", content, re.IGNORECASE):
37 return content
38
39 # Otherwise try markdown
40 return md.markdown(content, extensions=["fenced_code", "tables", "toc"])
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41
42
43 def _get_repo_and_reader(slug):
44 """Return (project, fossil_repo, reader) or raise 404."""
45 project = get_object_or_404(Project, slug=slug, deleted_at__isnull=True)
46
--- fossil/views.py
+++ fossil/views.py
@@ -22,24 +22,49 @@
22 - Markdown (newer pages)
23 """
24 if not content:
25 return ""
26
27 # Detect format from the raw content BEFORE any transformations
28 is_markdown = _is_markdown(content)
29
30 if is_markdown:
31 # Markdown: convert Fossil [/path|text] links to markdown links first
32 content = re.sub(r"\[(/[^|\]]+)\|([^\]]+)\]", r"[\2](\1)", content)
33 content = re.sub(r"<verbatim>(.*?)</verbatim>", r"```\n\1\n```", content, flags=re.DOTALL)
34 return md.markdown(content, extensions=["fenced_code", "tables", "toc"])
35
36 # Fossil wiki / HTML: convert Fossil-specific syntax to HTML
37 content = re.sub(r"\[(/[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
 
38 content = re.sub(r"\[(https?://[^|\]]+)\|([^\]]+)\]", r'<a href="\1">\2</a>', content)
 
39 content = re.sub(r"<verbatim>(.*?)</verbatim>", r"<pre><code>\1</code></pre>", content, flags=re.DOTALL)
40 return content
41
 
 
 
42
43 def _is_markdown(content: str) -> bool:
44 """Detect if content is Markdown vs Fossil wiki/HTML.
45
46 Heuristic: if the content starts with markdown-style headings (#),
47 or has significant markdown syntax patterns, treat as markdown.
48 """
49 stripped = content.strip()
50 # Starts with markdown heading
51 if re.match(r"^#{1,6}\s", stripped):
52 return True
53 # Has multiple markdown headings
54 if len(re.findall(r"^#{1,6}\s", stripped, re.MULTILINE)) >= 2:
55 return True
56 # Has markdown link references [text][ref]
57 if re.search(r"\[.+\]\[.+\]", stripped):
58 return True
59 # Has markdown code fences
60 if "```" in stripped:
61 return True
62 # Starts with HTML block element — it's Fossil wiki/HTML
63 if re.match(r"<(h[1-6]|p|ol|ul|div|table)\b", stripped, re.IGNORECASE):
64 return False
65 return False
66
67
68 def _get_repo_and_reader(slug):
69 """Return (project, fossil_repo, reader) or raise 404."""
70 project = get_object_or_404(Project, slug=slug, deleted_at__isnull=True)
71

Keyboard Shortcuts

Open search /
Next entry (timeline) j
Previous entry (timeline) k
Open focused entry Enter
Show this help ?
Toggle theme Top nav button