From da0623fbbb378e761421ceb4a05dae31407b3657 Mon Sep 17 00:00:00 2001
From: Ilya Kreymer <ikreymer@gmail.com>
Date: Wed, 26 Mar 2014 14:05:02 -0700
Subject: [PATCH] lxml: ensure lxml support is optional: if not available,
 use_lxml_parser() will return false and doctests/pytest collection won't test
 the lxml parser

---
 pywb/rewrite/lxml_html_rewriter.py           |  8 +++++++-
 pywb/rewrite/rewriterules.py                 | 13 ++++++++-----
 pywb/rewrite/test/test_lxml_html_rewriter.py | 17 +++++++++++------
 3 files changed, 26 insertions(+), 12 deletions(-)

diff --git a/pywb/rewrite/lxml_html_rewriter.py b/pywb/rewrite/lxml_html_rewriter.py
index ef809cf8..3a8e05dd 100644
--- a/pywb/rewrite/lxml_html_rewriter.py
+++ b/pywb/rewrite/lxml_html_rewriter.py
@@ -1,4 +1,10 @@
-import lxml.etree
+try:
+    import lxml.etree
+    LXML_SUPPORTED = True
+except ImportError:
+    LXML_SUPPORTED = False
+    pass
+
 import cgi
 import re
 
diff --git a/pywb/rewrite/rewriterules.py b/pywb/rewrite/rewriterules.py
index 007e925a..083bf129 100644
--- a/pywb/rewrite/rewriterules.py
+++ b/pywb/rewrite/rewriterules.py
@@ -13,14 +13,17 @@ HTML = HTMLRewriter
 
 #=================================================================
 def use_lxml_parser():
-    try:
-        import logging
-        from lxml_html_rewriter import LXMLHTMLRewriter
+    import logging
+    from lxml_html_rewriter import LXMLHTMLRewriter, LXML_SUPPORTED
+
+    if LXML_SUPPORTED:
         global HTML
         HTML = LXMLHTMLRewriter
         logging.debug('Using LXML Parser')
-    except ImportError:
-        logging.debug('Error Loading LXML Parser')
+        return True
+    else:
+        logging.debug('LXML Parser not available')
+        return False
 
 
 #=================================================================
diff --git a/pywb/rewrite/test/test_lxml_html_rewriter.py b/pywb/rewrite/test/test_lxml_html_rewriter.py
index 2d6df8f7..6beb7c7c 100644
--- a/pywb/rewrite/test/test_lxml_html_rewriter.py
+++ b/pywb/rewrite/test/test_lxml_html_rewriter.py
@@ -114,10 +114,7 @@ ur"""
 
 from pywb.rewrite.url_rewriter import UrlRewriter
 
-try:
-    from pywb.rewrite.lxml_html_rewriter import LXMLHTMLRewriter
-except ImportError:
-    pass
+from pywb.rewrite.lxml_html_rewriter import LXMLHTMLRewriter, LXML_SUPPORTED
 
 urlrewriter = UrlRewriter('20131226101010/http://example.com/some/path/index.html', '/web/')
 
@@ -127,5 +124,13 @@ def parse(data, head_insert=None):
     print parser.rewrite(data) + parser.close()
 
 if __name__ == "__main__":
-    import doctest
-    doctest.testmod()
+    if LXML_SUPPORTED:
+        import doctest
+        doctest.testmod()
+else:
+    # skip if not supported and lxml not available
+    if not LXML_SUPPORTED:
+        import pytest
+        lxml = pytest.importorskip('lxml.etree')
+
+