diff options
| author | Bryan Newbold <bnewbold@robocracy.org> | 2019-04-09 12:14:56 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@robocracy.org> | 2019-04-09 12:14:56 -0700 |
| commit | 70f652deb27d5e7b11ed54e9e418331ebb1bbae7 (patch) | |
| tree | 4cf1a3c01b23d1b47ec8a2b3d35fbccbd6ff7529 /python/fatcat_web/hacks.py | |
| parent | 3c5bd4d21f4744ae444819adaffa20e3b52ea61c (diff) | |
| download | fatcat-70f652deb27d5e7b11ed54e9e418331ebb1bbae7.tar.gz fatcat-70f652deb27d5e7b11ed54e9e418331ebb1bbae7.zip | |
temporary hack to remove ext-link XML in refs
Diffstat (limited to 'python/fatcat_web/hacks.py')
| -rw-r--r-- | python/fatcat_web/hacks.py | 15 |
1 files changed, 15 insertions, 0 deletions
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.""" + |