[GRDM-57869] Improve Maintenance Message Display

admin/base/templatetags/filters.py

@@ -3,6 +3,7 @@
 from django.utils.safestring import mark_safe
 import json
 from django.utils.translation import ugettext_lazy as _
+from osf.utils.text_rendering import osf_urlize as custom_osf_urlize
 
 register = template.Library()
 
@@ -13,3 +14,7 @@ def jsonify(o):
 @register.filter
 def transValue(value1):
     return _(str(value1))
+
+@register.filter
+def osf_urlize(text):
+    return custom_osf_urlize(text)

admin/templates/maintenance/display.html

@@ -3,7 +3,7 @@
 {% load render_bundle from webpack_loader %}
 {% load spam_extras %}
 {% load static %}
-{% load render_bundle from webpack_loader %}
+{% load filters %}
 {% block title %}
     <title>{% trans "Maintenance State" %}</title>
 {% endblock title %}
@@ -26,7 +26,7 @@ <h4>{% trans "Current alert:" %}</h4>
             <td>{{ current_alert.start }} - {{ current_alert.end }} UTC</td>
             <td>
                 {% if current_alert.message %}
-                    {{ current_alert.message }}
+                    {{ current_alert.message|osf_urlize|linebreaksbr }}
                 {% else %}
                     {% trans "The site will undergo maintenance between &lt;localized start time&gt;-&lt;localized end time&gt;. Thank you for your patience." %}
                 {% endif %}

osf/utils/text_rendering.py

@@ -0,0 +1,113 @@
+from django.template.defaultfilters import linebreaksbr
+from django.utils.html import escape
+from django.utils.safestring import mark_safe
+import re
+from urllib.parse import urlparse
+
+_LEADING_BRACKETS = frozenset('([{【（「『〔')
+_CLOSER_MAP = {')': '(', ']': '[', '}': '{', '）': '（', '】': '【', '」': '「', '』': '『', '〕': '〔'}
+_TRAILING_PUNCT = frozenset('.,!?、。')
+
+# Matches URL portion only — stops at whitespace, HTML chars, and Japanese brackets
+_URL_IN_TEXT = re.compile(r'https?://[^\s<>"\'【（「『〔】）」』〕、。]*')
+
+
+def is_valid_domain(host: str) -> bool:
+    if not host:
+        return False
+    if not re.match(r'^[a-zA-Z0-9.-]+$', host):
+        return False
+    if '..' in host or '.' not in host:
+        return False
+
+    labels = host.split('.')
+    for label in labels:
+        if label.startswith('-') or label.endswith('-') or not label:
+            return False
+
+    tld = labels[-1]
+    if not re.match(r'^[a-zA-Z]{2,}$', tld):
+        return False
+
+    return True
+
+
+def _trim_edges(s: str):
+    core = s
+    leading = ''
+    trailing = ''
+
+    while core and core[0] in _LEADING_BRACKETS:
+        leading += core[0]
+        core = core[1:]
+
+    changed = True
+    while changed and core:
+        changed = False
+        if core[-1] in _TRAILING_PUNCT:
+            trailing = core[-1] + trailing
+            core = core[:-1]
+            changed = True
+        elif core[-1] in _CLOSER_MAP:
+            closer = core[-1]
+            opener = _CLOSER_MAP[closer]
+            if core.count(opener) < core.count(closer):
+                trailing = closer + trailing
+                core = core[:-1]
+                changed = True
+
+    return leading, core, trailing
+
+
+def osf_urlize(text: str) -> str:
+    if not text:
+        return ''
+
+    result = []
+    for part in re.split(r'(\s+)', text):
+        if not part:
+            continue
+        if re.match(r'^\s+$', part):
+            result.append(part)
+            continue
+
+        for chunk in re.split(r'([<>"])', part):
+            if not chunk:
+                continue
+            if chunk in ('<', '>', '"'):
+                result.append(escape(chunk))
+                continue
+
+            leading, core, trailing = _trim_edges(chunk)
+
+            m = _URL_IN_TEXT.search(core)
+            if m:
+                text_before = core[:m.start()]
+                url_raw = m.group(0)
+                text_after = core[m.start() + len(url_raw):]
+
+                # Strip trailing ASCII punctuation and unbalanced brackets from URL
+                _, url_core, url_trailing = _trim_edges(url_raw)
+                text_after = url_trailing + text_after
+
+                try:
+                    host = urlparse(url_core).hostname
+                    if host and is_valid_domain(host):
+                        result.append(
+                            escape(leading + text_before) +
+                            f'<a href="{escape(url_core)}" rel="nofollow">{escape(url_core)}</a>' +
+                            escape(text_after + trailing)
+                        )
+                        continue
+                except Exception:
+                    pass
+
+            result.append(escape(leading + core + trailing))
+
+    return mark_safe(''.join(result))
+
+
+def render_text(text: str) -> str:
+    if not text:
+        return ''
+    return linebreaksbr(osf_urlize(text))

tests/test_text_rendering.py

@@ -0,0 +1,100 @@
+import unittest
+from osf.utils.text_rendering import render_text, osf_urlize
+
+
+class TestOsfUrlize(unittest.TestCase):
+
+    def test_https_url_linkified(self):
+        result = osf_urlize('Visit https://example.com')
+        assert '<a href="https://example.com"' in result
+        assert 'rel="nofollow"' in result
+
+    def test_http_url_linkified(self):
+        result = osf_urlize('Visit http://example.com')
+        assert '<a href="http://example.com"' in result
+
+    def test_bare_domain_not_linkified(self):
+        result = osf_urlize('Visit google.com')
+        assert '<a' not in result
+
+    def test_www_domain_not_linkified(self):
+        result = osf_urlize('Visit www.google.com')
+        assert '<a' not in result
+
+    def test_ftp_url_not_linkified(self):
+        result = osf_urlize('ftp://example.com')
+        assert '<a' not in result
+
+    def test_html_escaped(self):
+        result = osf_urlize('<script>alert("xss")</script>')
+        assert '<script>' not in result
+
+    def test_url_in_square_brackets_linkified(self):
+        result = osf_urlize('[https://example.com]')
+        assert '<a href="https://example.com"' in result
+        assert 'https://example.com]' not in result
+
+    def test_url_in_parentheses_linkified(self):
+        result = osf_urlize('(https://example.com)')
+        assert '<a href="https://example.com"' in result
+        assert 'https://example.com)' not in result
+
+    def test_url_with_trailing_slash_in_square_brackets(self):
+        result = osf_urlize('[https://example.com/]')
+        assert '<a href="https://example.com/"' in result
+        assert 'https://example.com/]' not in result
+
+    def test_url_with_trailing_slash_in_parentheses(self):
+        result = osf_urlize('(https://example.com/)')
+        assert '<a href="https://example.com/"' in result
+        assert 'https://example.com/)' not in result
+
+    def test_trailing_dots(self):
+        result = osf_urlize('https://google.com...')
+        assert '<a href="https://google.com"' in result
+        assert 'https://google.com...' not in result
+
+    def test_url_in_parens_with_trailing_period(self):
+        result = osf_urlize('(https://google.com).')
+        assert '<a href="https://google.com"' in result
+        assert 'https://google.com).' not in result
+
+    def test_trailing_comma(self):
+        result = osf_urlize('https://google.com,')
+        assert '<a href="https://google.com"' in result
+        assert 'https://google.com,' not in result
+
+    def test_url_in_double_quotes(self):
+        result = osf_urlize('"https://google.com"')
+        assert '<a href="https://google.com"' in result
+        assert 'https://google.com&quot;' not in result
+
+
+class TestRenderText(unittest.TestCase):
+
+    def test_empty_string(self):
+        assert render_text('') == ''
+
+    def test_none_input(self):
+        assert render_text(None) == ''
+
+    def test_linebreaks(self):
+        result = render_text('line1\nline2')
+        assert '<br' in result
+
+    def test_https_url_linkified(self):
+        result = render_text('Visit https://abc-test.com')
+        assert '<a href="https://abc-test.com"' in result
+
+    def test_bare_domain_not_linkified(self):
+        result = render_text('Visit google.com')
+        assert '<a' not in result
+
+    def test_plain_text_not_linkified(self):
+        result = render_text('version 1.0 etc. Co.,Ltd.')
+        assert '<a' not in result
+
+    def test_xss_not_rendered(self):
+        result = render_text('<script>alert("xss")</script>')
+        assert '<script>' not in result
+

website/templates/nav.mako

@@ -1,4 +1,5 @@
 <%def name="nav(service_name, service_url, service_support_url, service_support_target='_self')">
+<% from osf.utils.text_rendering import render_text %>
 <link rel="stylesheet" href='/static/css/nav.css'>
 <div class="osf-nav-wrapper">
 
@@ -139,7 +140,7 @@
                         <span aria-hidden="true">&times;</span></button>
                     <strong>${_('Notice:')}</strong>
                     % if maintenance['message']:
-                        ${maintenance['message']}
+                        ${render_text(maintenance['message']) | n}
                     % else:
                         ${_('The site will undergo maintenance between <span id="maintenanceTime"></span>.') | n}
                         ${_("Thank you for your patience.")}