Replace footer links with cfpb-link

cfgov/core/middleware.py

@@ -11,7 +11,13 @@
 
 from wagtail.rich_text import expand_db_html
 
-from core.utils import add_link_markup, get_body_html, get_link_tags
+from core.utils import (
+    add_cfpb_link_markup,
+    add_link_markup,
+    get_body_html,
+    get_cfpb_link_tags,
+    get_link_tags,
+)
 
 
 class DownstreamCacheControlMiddleware:
@@ -52,6 +58,12 @@ def parse_links(html, request_path=None, encoding=None):
         if tag_with_markup:
             expanded_html = expanded_html.replace(tag, tag_with_markup)
 
+    cfpb_link_tags = get_cfpb_link_tags(body_html)
+    for tag in cfpb_link_tags:
+        tag_with_markup = add_cfpb_link_markup(tag, request_path)
+        if tag_with_markup:
+            expanded_html = expanded_html.replace(tag, tag_with_markup)
+
     return expanded_html
 
 

cfgov/core/tests/test_utils.py

@@ -7,10 +7,12 @@
     ASK_CFPB_LINKS,
     NON_CFPB_LINKS,
     UNSAFE_CHARACTERS,
+    add_cfpb_link_markup,
     add_link_markup,
     extract_answers_from_request,
     format_file_size,
     get_body_html,
+    get_cfpb_link_tags,
     get_link_tags,
     make_safe,
     text_matches_href,
@@ -77,6 +79,61 @@ def test_format_file_size_terabytes(self):
         self.assertEqual(format_file_size(1024 * 9000000000), "8 TB")
 
 
+class CfpbLinkUtilsTests(SimpleTestCase):
+    def test_get_cfpb_link_tags(self):
+        self.assertEqual(
+            get_cfpb_link_tags(
+                'outer <cfpb-link><a href="">inner</a></cfpb-link>'
+            ),
+            ['<cfpb-link><a href="">inner</a></cfpb-link>'],
+        )
+
+    def test_add_cfpb_link_markup_invalid(self):
+        tag = "not a valid tag"
+        path = "/about-us/blog/"
+        self.assertIsNone(add_cfpb_link_markup(tag, path))
+
+    def test_add_cfpb_link_markup_anchor(self):
+        tag = '<cfpb-link><a href="/about-us/blog/#anchor">bar</a></cfpb-link>'
+        path = "/about-us/blog/"
+        self.assertEqual(
+            add_cfpb_link_markup(tag, path),
+            '<cfpb-link><a href="#anchor">bar</a></cfpb-link>',
+        )
+
+    def check_external_cfpb_link(self, url):
+        tag = f'<cfpb-link><a href="{url}">foo</a></cfpb-link>'
+        path = "/about-us/blog/"
+
+        expected_html = (
+            '<cfpb-link link-variant="external">'
+            f'<a href="{url}">foo</a></cfpb-link>'
+        )
+        expected_tag = BeautifulSoup(expected_html, "html.parser")
+
+        self.assertEqual(add_cfpb_link_markup(tag, path), str(expected_tag))
+
+    def check_download_cfpb_link(self, url):
+        tag = f'<cfpb-link><a href="{url}">foo</a></cfpb-link>'
+        path = "/about-us/blog/"
+
+        expected_html = (
+            '<cfpb-link link-variant="download">'
+            f'<a href="{url}">foo</a></cfpb-link>'
+        )
+        expected_tag = BeautifulSoup(expected_html, "html.parser")
+
+        self.assertEqual(add_cfpb_link_markup(tag, path), str(expected_tag))
+
+    def test_usa_gov(self):
+        url = "https://www.usa.gov"
+        self.check_external_cfpb_link(url)
+
+    def test_test_pdf(self):
+        url = "test.pdf"
+        self.check_download_cfpb_link(url)
+
+
 class LinkUtilsTests(SimpleTestCase):
     @classmethod
     def setUpClass(cls):
@@ -138,6 +195,12 @@ def test_get_link_tags_spacing(self):
             ],
         )
 
+    def test_get_link_tags_does_not_match_wc(self):
+        self.assertEqual(
+            get_link_tags('outer <cfpb-link><a href="">inner</a></cfpb-link>'),
+            [],
+        )
+
     def test_add_link_markup_invalid(self):
         tag = "not a valid tag"
         path = "/about-us/blog/"

cfgov/core/utils.py

@@ -59,6 +59,9 @@
 # Match <a…>…</a>
 A_TAG_RE = re.compile(TAG_RE.format(tag_name="a"))
 
+# Match <cfpb-link>…</cfpb-link>
+CFPB_LINK_TAG_RE = re.compile(TAG_RE.format(tag_name="cfpb-link"))
+
 # If a link contains these elements, it should *not* get an icon
 ICONLESS_LINK_CHILD_ELEMENTS = [
     "img",
@@ -90,6 +93,8 @@
 
 MAX_CHARS = 75
 
+BEAUTIFUL_SOUP_PARSER = "html.parser"
+
 
 def make_safe(term):
     for char in UNSAFE_CHARACTERS:
@@ -128,7 +133,18 @@
 
 
 def get_link_tags(html):
-    return A_TAG_RE.findall(html)
+    without_components = re.sub(
+        r"<cfpb-link\b[^>]*>.*?</cfpb-link>",
+        "",
+        html,
+        flags=re.DOTALL | re.IGNORECASE,
+    )
+    links = A_TAG_RE.findall(without_components)
+    return links
+
+
+def get_cfpb_link_tags(html):
+    return CFPB_LINK_TAG_RE.findall(html)
 
 
 def text_matches_href(text, href):
@@ -177,6 +193,42 @@
     return normalize((text or "").strip()) == normalize((href or "").strip())
 
 
+def add_cfpb_link_markup(tag, request_path):
+    """Add necessary attributes to the given cfpb-link.
+
+    Return cfpb-link if modified.
+    """
+    soup = BeautifulSoup(tag, BEAUTIFUL_SOUP_PARSER)
+    tag = soup.find("cfpb-link")
+
+    if tag is None:
+        return None
+
+    a_tag = tag.find("a")
+    href = a_tag["href"]
+
+    if request_path is not None:
+        # Strips the path of the current page from hrefs that are internal page
+        # anchor links.
+        # TODO: Remove that functionality when we get to Wagtail>=2.7, which
+        # adds the ability to create anchor links.
+        in_page_anchor_pattern = request_path + "#"
+        if a_tag["href"].startswith(in_page_anchor_pattern):
+            # Strip current path from in-page anchor links
+            a_tag["href"] = href.replace(request_path, "")
+            return str(tag)
+
+    if NON_CFPB_LINKS.match(href):
+        # Sets the icon to indicate you're leaving consumerfinance.gov
+        tag["link-variant"] = "external"
+
+    elif DOWNLOAD_LINKS.search(href):
+        # Sets the icon to indicate you're downloading a file
+        tag["link-variant"] = "download"
+
+    return str(tag)
+
+
 def add_link_markup(tag, request_path):
     """Add necessary markup to the given link and return if modified.
 
@@ -190,7 +242,7 @@
     """
     icon = False
 
-    soup = BeautifulSoup(tag, "html.parser")
+    soup = BeautifulSoup(tag, BEAUTIFUL_SOUP_PARSER)
     tag = soup.find("a", href=True)
 
     if tag is None:
@@ -252,7 +304,7 @@
     icon_classes = {"class": LINK_ICON_TEXT_CLASSES}
     spans = tag.findAll("span", icon_classes)
 
-    icon_soup = BeautifulSoup(svg_icon(icon), "html.parser")
+    icon_soup = BeautifulSoup(svg_icon(icon), BEAUTIFUL_SOUP_PARSER)
 
     # If this is an <a class="a-btn"> tag without a span inside, we want to
     # add proper markup so that the link appears as a button with the icon on