# can't do for all responses because we need to support binary files too
if not isinstance(page, unicode) and "text/" in contentType:
if kb.heuristicMode:
kb.pageEncoding = kb.pageEncoding or checkCharEncoding(getHeuristicCharEncoding(page))
page = getUnicode(page, kb.pageEncoding)
else:
# e.g. Ãëàâà
if "&#" in 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. &
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))
if (kb.pageEncoding or "").lower() == "utf-8-sig":
kb.pageEncoding = "utf-8"
if page and page.startswith("\xef\xbb\xbf"): # Reference: https://docs.python.org/2/library/codecs.html (Note: noticed problems when "utf-8-sig" is left to Python for handling)
page = page[3:]
page = getUnicode(page, kb.pageEncoding)
# e.g. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
# can't do for all responses because we need to support binary files too
if contentType and not isinstance(page, unicode) and "text/" in contentType.lower():
if kb.heuristicMode:
kb.pageEncoding = kb.pageEncoding or checkCharEncoding(getHeuristicCharEncoding(page))
page = getUnicode(page, kb.pageEncoding)
else:
# e.g. Ãëàâà
if "&#" in 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. &
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. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
retVal = unichr(int(match.group(1)))
except ValueError:
pass
return retVal
page = re.sub(r"&#(\d+);", _, page)
else:
# e.g. Ãëàâà
if "&#" in 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. &
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. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
retVal = unichr(int(match.group(1)))
except ValueError:
pass
def decodePage(page, contentEncoding, contentType):
"""
Decode compressed/charset HTTP response
"""
if not page or (conf.nullConnection and len(page) < 2):
return getUnicode(page)
if hasattr(contentEncoding, "lower"):
contentEncoding = contentEncoding.lower()
else:
contentEncoding = ""
if hasattr(contentType, "lower"):
contentType = contentType.lower()
else:
contentType = ""
if contentEncoding in ("gzip", "x-gzip", "deflate"):
if not kb.pageCompress:
return None
try:
if contentEncoding == "deflate":
data = io.BytesIO(zlib.decompress(page, -15)) # Reference: http://stackoverflow.com/questions/1089662/python-inflate-and-deflate-implementations
else:
data = gzip.GzipFile("", "rb", 9, io.BytesIO(page))
size = struct.unpack("<l", page[-4:])[0] # Reference: http://pydoc.org/get.cgi/usr/local/lib/python2.5/gzip.py
if size > MAX_CONNECTION_TOTAL_SIZE:
raise Exception("size too large")
page = data.read()
except Exception as ex:
if "<html" not in page: # in some cases, invalid "Content-Encoding" appears for plain HTML (should be ignored)
errMsg = "detected invalid data for declared content "
errMsg += "encoding '%s' ('%s')" % (contentEncoding, getSafeExString(ex))
singleTimeLogMessage(errMsg, logging.ERROR)
warnMsg = "turning off page compression"
singleTimeWarnMessage(warnMsg)
kb.pageCompress = False
raise SqlmapCompressionException
if not conf.encoding:
httpCharset, metaCharset = None, None
# Reference: http://stackoverflow.com/questions/1020892/python-urllib2-read-to-unicode
if contentType.find("charset=") != -1:
httpCharset = checkCharEncoding(contentType.split("charset=")[-1])
metaCharset = checkCharEncoding(extractRegexResult(META_CHARSET_REGEX, page))
if (any((httpCharset, metaCharset)) and not all((httpCharset, metaCharset))) or (httpCharset == metaCharset and all((httpCharset, metaCharset))):
kb.pageEncoding = httpCharset or metaCharset # Reference: http://bytes.com/topic/html-css/answers/154758-http-equiv-vs-true-header-has-precedence
debugMsg = "declared web page charset '%s'" % kb.pageEncoding
singleTimeLogMessage(debugMsg, logging.DEBUG, debugMsg)
else:
kb.pageEncoding = None
else:
kb.pageEncoding = conf.encoding
# can't do for all responses because we need to support binary files too
if isinstance(page, six.binary_type) and "text/" in contentType:
# e.g. 	Ãëàâà
if b"&#" in page:
page = re.sub(b"&#x([0-9a-f]{1,2});", lambda _: decodeHex(_.group(1) if len(_.group(1)) == 2 else "0%s" % _.group(1)), page)
page = re.sub(b"&#(\d{1,3});", lambda _: chr(int(_.group(1))) if int(_.group(1)) < 256 else _.group(0), page)
# e.g. %20%28%29
if b"%" in page:
page = re.sub(b"%([0-9a-fA-F]{2})", lambda _: decodeHex(_.group(1)), page)
# e.g. &
page = re.sub(b"&([^;]+);", lambda _: chr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 256) < 256 else _.group(0), page)
kb.pageEncoding = kb.pageEncoding or checkCharEncoding(getHeuristicCharEncoding(page))
if (kb.pageEncoding or "").lower() == "utf-8-sig":
kb.pageEncoding = "utf-8"
if page and page.startswith("\xef\xbb\xbf"): # Reference: https://docs.python.org/2/library/codecs.html (Note: noticed problems when "utf-8-sig" is left to Python for handling)
page = page[3:]
page = getUnicode(page, kb.pageEncoding)
# e.g. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
retVal = unichr(int(match.group(1)))
except (ValueError, OverflowError):
pass
return retVal
page = re.sub(r"&#(\d+);", _, page)
# e.g. ζ
page = re.sub(r"&([^;]+);", lambda _: unichr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 0) > 255 else _.group(0), page)
return page
page = getUnicode(page, kb.pageEncoding)
else:
# e.g. Ãëàâà
if "&#" in 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. &
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. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
retVal = unichr(int(match.group(1)))
except ValueError:
pass
def decodePage(page, contentEncoding, contentType):
"""
Decode compressed/charset HTTP response
"""
if not page or (conf.nullConnection and len(page) < 2):
return getUnicode(page)
if hasattr(contentEncoding, "lower"):
contentEncoding = contentEncoding.lower()
else:
contentEncoding = ""
if hasattr(contentType, "lower"):
contentType = contentType.lower()
else:
contentType = ""
if contentEncoding in ("gzip", "x-gzip", "deflate"):
if not kb.pageCompress:
return None
try:
if contentEncoding == "deflate":
data = io.BytesIO(zlib.decompress(page, -15)) # Reference: http://stackoverflow.com/questions/1089662/python-inflate-and-deflate-implementations
else:
data = gzip.GzipFile("", "rb", 9, io.BytesIO(page))
size = struct.unpack("<l", page[-4:])[0] # Reference: http://pydoc.org/get.cgi/usr/local/lib/python2.5/gzip.py
if size > MAX_CONNECTION_TOTAL_SIZE:
raise Exception("size too large")
page = data.read()
except Exception as ex:
if "<html" not in page: # in some cases, invalid "Content-Encoding" appears for plain HTML (should be ignored)
errMsg = "detected invalid data for declared content "
errMsg += "encoding '%s' ('%s')" % (contentEncoding, getSafeExString(ex))
singleTimeLogMessage(errMsg, logging.ERROR)
warnMsg = "turning off page compression"
singleTimeWarnMessage(warnMsg)
kb.pageCompress = False
raise SqlmapCompressionException
if not conf.encoding:
httpCharset, metaCharset = None, None
# Reference: http://stackoverflow.com/questions/1020892/python-urllib2-read-to-unicode
if contentType.find("charset=") != -1:
httpCharset = checkCharEncoding(contentType.split("charset=")[-1])
metaCharset = checkCharEncoding(extractRegexResult(META_CHARSET_REGEX, page))
if (any((httpCharset, metaCharset)) and not all((httpCharset, metaCharset))) or (httpCharset == metaCharset and all((httpCharset, metaCharset))):
kb.pageEncoding = httpCharset or metaCharset # Reference: http://bytes.com/topic/html-css/answers/154758-http-equiv-vs-true-header-has-precedence
debugMsg = "declared web page charset '%s'" % kb.pageEncoding
singleTimeLogMessage(debugMsg, logging.DEBUG, debugMsg)
else:
kb.pageEncoding = None
else:
kb.pageEncoding = conf.encoding
# can't do for all responses because we need to support binary files too
if isinstance(page, six.binary_type) and "text/" in contentType:
# e.g. 	Ãëàâà
if "&#" in page:
page = re.sub(r"&#x([0-9a-f]{1,2});", lambda _: (_.group(1) if len(_.group(1)) == 2 else "0%s" % _.group(1)).decode("hex"), 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. &
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))
if (kb.pageEncoding or "").lower() == "utf-8-sig":
kb.pageEncoding = "utf-8"
if page and page.startswith("\xef\xbb\xbf"): # Reference: https://docs.python.org/2/library/codecs.html (Note: noticed problems when "utf-8-sig" is left to Python for handling)
page = page[3:]
page = getUnicode(page, kb.pageEncoding)
# e.g. ’…™
if "&#" in page:
def _(match):
retVal = match.group(0)
try:
retVal = unichr(int(match.group(1)))
except (ValueError, OverflowError):
pass
return retVal
page = re.sub(r"&#(\d+);", _, page)
# e.g. ζ
page = re.sub(r"&([^;]+);", lambda _: unichr(htmlEntities[_.group(1)]) if htmlEntities.get(_.group(1), 0) > 255 else _.group(0), page)
return page
