From 70f652deb27d5e7b11ed54e9e418331ebb1bbae7 Mon Sep 17 00:00:00 2001
From: Bryan Newbold <bnewbold@robocracy.org>
Date: Tue, 9 Apr 2019 12:14:56 -0700
Subject: temporary hack to remove ext-link XML in refs

---
 python/fatcat_web/hacks.py  | 15 +++++++++++++++
 python/fatcat_web/routes.py | 10 ++++++++++
 2 files changed, 25 insertions(+)
 create mode 100644 python/fatcat_web/hacks.py

(limited to 'python')

diff --git a/python/fatcat_web/hacks.py b/python/fatcat_web/hacks.py
new file mode 100644
index 00000000..51b97142
--- /dev/null
+++ b/python/fatcat_web/hacks.py
@@ -0,0 +1,15 @@
+
+import re
+
+STRIP_EXTLINK_XML_RE = re.compile(r"<ext-link.*xlink:type=\"simple\">")
+
+def strip_extlink_xml(unstr):
+    unstr = unstr.replace("</ext-link>", "")
+    unstr = STRIP_EXTLINK_XML_RE.sub("", unstr)
+    return unstr
+
+def test_strip_extlink_xml():
+    assert strip_extlink_xml("asdf") == "asdf"
+    assert strip_extlink_xml("""LOCKSS (2014) Available: <ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="http://lockss.org/" xlink:type="simple">http://lockss.org/</ext-link>. Accessed: 2014 November 1.""") == \
+    """LOCKSS (2014) Available: http://lockss.org/. Accessed: 2014 November 1."""
+
diff --git a/python/fatcat_web/routes.py b/python/fatcat_web/routes.py
index c3467a61..eb53c2ef 100644
--- a/python/fatcat_web/routes.py
+++ b/python/fatcat_web/routes.py
@@ -13,6 +13,7 @@ from fatcat_web import app, api, auth_api, priv_api
 from fatcat_web.auth import handle_token_login, handle_logout, load_user, handle_ia_xauth
 from fatcat_web.cors import crossdomain
 from fatcat_web.search import *
+from fatcat_web.hacks import strip_extlink_xml
 
 
 ### Views ###################################################################
@@ -280,6 +281,15 @@ def release_view(ident):
         entity.es = release_to_elasticsearch(entity, force_bool=False)
     for fs in filesets:
         fs.total_size = sum([f.size for f in fs.manifest])
+    for ref in entity.refs:
+        # this is a UI hack to get rid of XML crud in unstructured refs like:
+        # LOCKSS (2014) Available: <ext-link
+        # xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri"
+        # xlink:href="http://lockss.org/"
+        # xlink:type="simple">http://lockss.org/</ext-link>. Accessed: 2014
+        # November 1.
+        if ref.extra and ref.extra.get('unstructured'):
+            ref.extra['unstructured'] = strip_extlink_xml(ref.extra['unstructured'])
     entity.filesets = filesets
     authors = [c for c in entity.contribs if c.role in ('author', None)]
     authors = sorted(authors, key=lambda c: c.index or 99999999)
-- 
cgit v1.2.3