Automatically decoding url encoded data in response

This commit is contained in:
Miroslav Stampar 2013-03-05 17:32:10 +01:00
parent 9e49d8c68f
commit 62980d7d5a

View File

@ -227,23 +227,27 @@ def decodePage(page, contentEncoding, contentType):
kb.pageEncoding = conf.charset
# can't do for all responses because we need to support binary files too
if contentType and not isinstance(page, unicode) and any(map(lambda _: _ in contentType.lower(), ("text/txt", "text/raw", "text/html", "text/xml"))):
if contentType and not isinstance(page, unicode) and "text/" in contentType.lower():
# e.g. Ãëàâà
if "&#" in page:
page = re.sub('&#(\d{1,3});', lambda _: chr(int(_.group(1))) if int(_.group(1)) < 256 else _.group(0), page)
page = re.sub(r"&#(\d{1,3});", lambda _: chr(int(_.group(1))) if int(_.group(1)) < 256 else _.group(0), page)
# e.g. %20%28%29
if "%" in page:
page = re.sub(r"%([0-9a-fA-F]{2})", lambda _: _.group(1).decode("hex"), page)
# e.g. &amp;
page = re.sub('&([^;]+);', lambda _: chr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 256) < 256 else _.group(0), page)
page = re.sub(r"&([^;]+);", lambda _: chr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 256) < 256 else _.group(0), page)
kb.pageEncoding = kb.pageEncoding or checkCharEncoding(getHeuristicCharEncoding(page))
page = getUnicode(page, kb.pageEncoding)
# e.g. &#8217;&#8230;&#8482;
if "&#" in page:
page = re.sub('&#(\d+);', lambda _: unichr(int(_.group(1))), page)
page = re.sub(r"&#(\d+);", lambda _: unichr(int(_.group(1))), page)
# e.g. &zeta;
page = re.sub('&([^;]+);', lambda _: unichr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 0) > 255 else _.group(0), page)
page = re.sub(r"&([^;]+);", lambda _: unichr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 0) > 255 else _.group(0), page)
return page