Fix html transformation for about + syntax hl changes

1 files changed, 83 insertions, 0 deletions

diff --git a/cgit/filters/html-transform.py b/cgit/filters/html-transform.py
new file mode 100755
index 0000000..9fc7aec
--- /dev/null
+++ b/cgit/filters/html-transform.py
@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+# Rewrite image urls in about section
+
+import sys
+from html.parser import HTMLParser
+import html
+from urllib.parse import urljoin, urlparse
+import re
+
+def is_absolute_url(url):
+  if not url: return False
+  parsed = urlparse(url)
+  return bool(parsed.netloc or parsed.scheme)
+
+def is_relative_path(url):
+  if not url: return False
+  return not url.startswith("/") and not is_absolute_url(url)
+
+def linkify(text, make_link):
+  url_pattern = r'(?<!\S)(https?://\S+?)([.,!?;:]?(?=\s|$))'
+  def replace(m):
+    url, trail = m.group(1), m.group(2)
+    return make_link(url) + trail
+  return re.sub(url_pattern, replace, text)
+
+class LinkRewriter(HTMLParser):
+  def __init__(self):
+    super().__init__(convert_charrefs=False)
+    self.output = []
+
+  def get_html(self):
+    return "".join(self.output)
+
+  def _build_tag(self, tag, attrs, self_closing = False):
+    new_tag = tag
+    attr_map = {}
+    for name, value in attrs:
+      if name in ("href", "src") and is_relative_path(value) and not value.startswith('#'):
+        if tag == 'a':
+          value = "../" + urljoin("tree/", value)
+        else:
+          value = "../" + urljoin("plain/", value)
+      attr_map[name] = value
+      # Open in new tab
+      if tag == "a" and name == "href" and is_absolute_url(value):
+        if not 'target' in attr_map: attr_map['target'] = '_blank _parent'
+        if not 'rel' in attr_map: attr_map['rel'] = 'noopener'
+
+    attr_str = "".join(
+      f' {name}="{html.escape(value, quote=True)}"' if value is not None else f" {name}"
+      for name, value in attr_map.items()
+    )
+    return f"<{new_tag}{attr_str}{' /' if self_closing else ''}>"
+
+  def handle_starttag(self, tag, attrs):
+    self.output.append(self._build_tag(tag, attrs))
+
+  def handle_startendtag(self, tag, attrs):
+    self.output.append(self._build_tag(tag, attrs, self_closing=True))
+
+  def handle_endtag(self, tag):
+    self.output.append(f"</{tag}>")
+
+  def handle_data(self, data):
+    data_with_links = linkify(data,
+      lambda url: self._build_tag("a", [('href', url)]) + html.escape(url) + "</a>")
+    self.output.append(data_with_links)
+
+  def handle_entityref(self, name):
+    self.output.append(f"&{name};")
+
+  def handle_charref(self, name):
+    self.output.append(f"&#{name};")
+
+  def handle_comment(self, data):
+    self.output.append(f"<!--{data}-->")
+
+html_in = sys.stdin.read()
+
+parser = LinkRewriter()
+parser.feed(html_in)
+
+print(parser.get_html())