diff --git a/pywb/binsearch.py b/pywb/binsearch.py
new file mode 100644
index 00000000..ef3171f1
--- /dev/null
+++ b/pywb/binsearch.py
@@ -0,0 +1,92 @@
+from collections import deque
+import os
+import itertools
+
+class FileReader:
+ def __init__(self, filename):
+ self.fh = open(filename, 'rb')
+ self.size = os.path.getsize(filename)
+
+ def getsize(self):
+ return self.size
+
+ def readline(self):
+ return self.fh.readline()
+
+ def seek(self, offset):
+ return self.fh.seek(offset)
+
+
+def binsearch_offset(reader, key, compare_func = cmp, block_size = 8192):
+ min = 0
+ max = reader.getsize() / block_size
+
+ while (max - min > 1):
+ mid = min + ((max - min) / 2)
+ reader.seek(mid * block_size)
+
+ if mid > 0:
+ reader.readline() # skip partial line
+
+ line = reader.readline()
+
+ if compare_func(key, line) > 0:
+ min = mid
+ else:
+ max = mid
+
+ return (min * block_size)
+
+
+def search(reader, key, prev_size = 0, compare_func = cmp, block_size = 8192):
+ min = binsearch_offset(reader, key, compare_func, block_size)
+
+ reader.seek(min)
+
+ if min > 0:
+ reader.readline() # skip partial line
+
+ if prev_size > 1:
+ prev_deque = deque(maxlen = prev_size)
+
+ line = None
+
+ while True:
+ line = reader.readline()
+ if not line:
+ break
+ if compare_func(line, key) >= 0:
+ break
+
+ if prev_size == 1:
+ prev = line
+ elif prev_size > 1:
+ prev_deque.append(line)
+
+ def gen_iter(line):
+ if prev_size == 1:
+ yield prev
+ elif prev_size > 1:
+ for i in prev_deque:
+ yield i
+
+ while line:
+ yield line
+ line = reader.readline()
+
+ return gen_iter(line)
+
+
+# Iterate over exact matches
+def iter_exact(reader, key):
+ lines = search(reader, key)
+ for x in lines:
+ if not x.startswith(key):
+ break
+
+ yield x
+
+
+
+
+
diff --git a/pywb/html_rewriter.py b/pywb/html_rewriter.py
index 800f5dda..0ee6bf2a 100644
--- a/pywb/html_rewriter.py
+++ b/pywb/html_rewriter.py
@@ -23,13 +23,14 @@ class WBHtml(HTMLParser):
>>> parse('![]()
')
- ![]()
+ ![]()
>>> parse('')
+ # HTML Entities
>>> parse('› >')
- › >
+ › >
# Don't rewrite anchors
>>> parse('Text')
@@ -47,7 +48,7 @@ class WBHtml(HTMLParser):
>>> parse('')
-
+
# Script tag
>>> parse('')
@@ -61,7 +62,7 @@ class WBHtml(HTMLParser):
>>> parse('')
-
+
>>> parse('')
@@ -218,11 +219,12 @@ class WBHtml(HTMLParser):
if rwMod is not None:
attrValue = self._rewriteURL(attrValue, rwMod)
- if attrValue is not None:
- #self.out.write(' {0}="{1}"'.format(attrName, attrValue))
+ # parser doesn't differentiate between 'attr=""' and just 'attr'
+ # 'attr=""' is more common, so use that form
+ if attrValue:
self.out.write(' ' + attrName + '="' + attrValue + '"')
else:
- self.out.write(' ' + attrName)
+ self.out.write(' ' + attrName + '=""')
self.out.write('/>' if isStartEnd else '>')