def showAdmin():
ipfsEndpoint = '{0}:{1}'.format(IPFSAPI_HOST, IPFSAPI_PORT)
status = {
'ipwbVersion': ipwbVersion,
'ipfsEndpoint': ipfsEndpoint,
'ipfsAlive': ipwbUtils.isDaemonAlive(ipfsEndpoint)
}
# TODO: Calculate actual URI-R/M counts
indexes = [{
'path': ipwbUtils.getIPWBReplayIndexPath(),
'enabled': True,
'urimCount': '#URI-M',
'urirCount': '#URI-R'
}]
# TODO: Calculate actual values
summary = {
'urimCount': '#URI-M',
'urirCount': '#URI-R',
'htmlCount': '#HTML',
'earliest': 'Date1',
'latest': 'Date2'
}
return render_template('admin.html',
status=status,
indexes=indexes,
summary=summary)
def showWebUI(path):
webuiPath = '/'.join(('webui', path)).replace('ipwb.replay', 'ipwb')
content = pkg_resources.resource_string(__name__, webuiPath)
if 'index.html' in path:
iFile = ipwbConfig.getIPWBReplayIndexPath()
if iFile is None or iFile == '':
iFile = pkg_resources.resource_filename(__name__, INDEX_FILE)
if not os.path.isabs(iFile): # Convert rel to abs path
iFileAbs = pkg_resources.resource_filename(__name__, iFile)
if os.path.exists(iFileAbs):
iFile = iFileAbs # Local file
content = content.replace('MEMCOUNT', str(retrieveMemCount(iFile)))
content = content.replace(
'var uris = []',
'var uris = {0}'.format(getURIsAndDatetimesInCDXJ(iFile)))
content = content.replace('INDEXSRC', iFile)
fileExtension = os.path.splitext(path)[1]
mimeType = 'text/html'
if fileExtension == '.js':
mimeType = 'application/javascript'
elif fileExtension == '.css':
mimeType = 'text/css'
resp = Response(content, mimetype=mimeType)
resp.headers['Service-Worker-Allowed'] = '/'
return resp
def resolveMemento(urir, datetime):
""" Request a URI-R at a supplied datetime from the CDXJ """
urir = getCompleteURI(urir)
if ipwbUtils.isLocalHosty(urir):
urir = urir.split('/', 4)[4]
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbUtils.getIPWBReplayIndexPath()
print('Getting CDXJ Lines with the URI-R {0} from {1}'.format(
urir, indexPath))
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
closestLine = getCDXJLineClosestTo(datetime, cdxjLinesWithURIR)
if closestLine is None:
msg = '<h1>ERROR 404</h1>'
msg += 'No capture found for {0} at {1}.'.format(urir, datetime)
return Response(msg, status=404)
uri = unsurt(closestLine.split(' ')[0])
newDatetime = closestLine.split(' ')[1]
linkHeader = getLinkHeaderAbbreviatedTimeMap(urir, newDatetime)
return (newDatetime, linkHeader, uri)
def showTimeMap(urir, format):
urir = getCompleteURI(urir)
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbUtils.getIPWBReplayIndexPath()
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
tmContentType = ''
hostAndPort = ipwbUtils.getIPWBReplayConfig()
tgURI = 'http://{0}:{1}/timegate/{2}'.format(hostAndPort[0],
hostAndPort[1], urir)
tm = '' # Initialize for usage beyond below conditionals
if format == 'link':
tm = generateLinkTimeMapFromCDXJLines(cdxjLinesWithURIR, s,
request.url, tgURI)
tmContentType = 'application/link-format'
elif format == 'cdxj':
tm = generateCDXJTimeMapFromCDXJLines(cdxjLinesWithURIR, s,
request.url, tgURI)
tmContentType = 'application/cdxj+ors'
resp = Response(tm)
resp.headers['Content-Type'] = tmContentType
return resp
def getCDXJLinesWithURIR(urir, indexPath):
""" Get all CDXJ records corresponding to a URI-R """
if not indexPath:
indexPath = ipwbConfig.getIPWBReplayIndexPath()
indexPath = getIndexFileFullPath(indexPath)
print('Getting CDXJ Lines with {0} in {1}'.format(urir, indexPath))
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
cdxjLinesWithURIR = []
cdxjLineIndex = getCDXJLine_binarySearch(s, indexPath, True, True) # get i
if cdxjLineIndex is None:
return []
cdxjLines = []
with open(indexPath, 'r') as f:
cdxjLines = f.read().split('\n')
baseCDXJLine = cdxjLines[cdxjLineIndex] # via binsearch
cdxjLinesWithURIR.append(baseCDXJLine)
# Get lines before pivot that match surt
sI = cdxjLineIndex - 1
while sI >= 0:
if cdxjLines[sI].split(' ')[0] == s:
cdxjLinesWithURIR.append(cdxjLines[sI])
sI -= 1
# Get lines after pivot that match surt
sI = cdxjLineIndex + 1
while sI < len(cdxjLines):
if cdxjLines[sI].split(' ')[0] == s:
cdxjLinesWithURIR.append(cdxjLines[sI])
sI += 1
return cdxjLinesWithURIR
def showMementosForURIRs(urir):
urir = getCompleteURI(urir)
if ipwbConfig.isLocalHosty(urir):
urir = urir.split('/', 4)[4]
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
print('Getting CDXJ Lines with the URI-R {0} from {1}'.format(
urir, indexPath))
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
if len(cdxjLinesWithURIR) == 1:
fields = cdxjLinesWithURIR[0].split(' ', 2)
redirectURI = '/{1}/{0}'.format(unsurt(fields[0]), fields[1])
return redirect(redirectURI, code=302)
msg = ''
if cdxjLinesWithURIR:
msg += '<p>{0} capture(s) available:</p><ul>'.format(
len(cdxjLinesWithURIR))
for line in cdxjLinesWithURIR:
fields = line.split(' ', 2)
dt14 = fields[1]
dtrfc1123 = ipwbConfig.datetimeToRFC1123(fields[1])
msg += ('<li><a href="/{1}/{0}">{0} at {2}</a></li>'.format(
unsurt(fields[0]), dt14, dtrfc1123))
msg += '</ul>'
return Response(msg)
def getCDXJLinesWithURIR(urir, indexPath=ipwbConfig.getIPWBReplayIndexPath()):
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
cdxjLinesWithURIR = []
cdxjLineIndex = getCDXJLine_binarySearch(s, indexPath, True, True) # get i
if cdxjLineIndex is None:
return []
cdxjLines = []
with open(indexPath, 'r') as f:
cdxjLines = f.read().split('\n')
baseCDXJLine = cdxjLines[cdxjLineIndex] # via binsearch
cdxjLinesWithURIR.append(baseCDXJLine)
# Get lines before pivot that match surt
sI = cdxjLineIndex - 1
while sI >= 0:
if cdxjLines[sI].split(' ')[0] == s:
cdxjLinesWithURIR.append(cdxjLines[sI])
sI -= 1
# Get lines after pivot that match surt
sI = cdxjLineIndex + 1
while sI < len(cdxjLines):
if cdxjLines[sI].split(' ')[0] == s:
cdxjLinesWithURIR.append(cdxjLines[sI])
sI += 1
return cdxjLinesWithURIR
def showTimeMap(urir, format):
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
tm = generateTimeMapFromCDXJLines(cdxjLinesWithURIR, s, request.url)
return Response(tm)
def getLinkHeaderAbbreviatedTimeMap(urir, pivotDatetime):
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
hostAndPort = ipwbConfig.getIPWBReplayConfig()
tmURI = 'http://{0}:{1}/timemap/link/{2}'.format(
'localhost', # hostAndPort[0],
hostAndPort[1],
urir)
tm = generateLinkTimeMapFromCDXJLines(cdxjLinesWithURIR, s, tmURI)
# Fix base TM relation when viewing abbrev version in Link resp
tm = tm.replace('rel="self"', 'rel="timemap"')
# Only one memento in TimeMap
if 'rel="first last memento"' in tm:
return tm
tmLines = tm.split('\n')
for idx, line in enumerate(tmLines):
if len(re.findall('rel=.*memento"', line)) == 0:
continue # Not a memento
if pivotDatetime in line:
addBothNextAndPrev = False
if idx > 0 and idx < len(tmLines) - 1:
addBothNextAndPrev = True
if addBothNextAndPrev or idx == 0:
tmLines[idx + 1] = \
tmLines[idx + 1].replace('memento"', 'next memento"')
if addBothNextAndPrev or idx == len(tmLines) - 1:
tmLines[idx - 1] = \
tmLines[idx - 1].replace('memento"', 'prev memento"')
break
# Remove all mementos in abbrev TM that are not:
# first, last, prev, next, or pivot
for idx, line in enumerate(tmLines):
if len(re.findall('rel=.*memento"', line)) == 0:
continue # Not a memento
if pivotDatetime in line:
continue
if len(re.findall('rel=.*(next|prev|first|last)', line)) == 0:
tmLines[idx] = ''
tm = '\n'.join(tmLines)
return tm
def showTimeMap(urir, format):
urir = getCompleteURI(urir)
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
tmContentType = ''
if format == 'link':
tm = generateLinkTimeMapFromCDXJLines(cdxjLinesWithURIR, s,
request.url)
tmContentType = 'application/link-format'
elif format == 'cdxj':
tm = generateCDXJTimeMapFromCDXJLines(cdxjLinesWithURIR, s,
request.url)
tmContentType = 'application/cdxj+ors'
resp = Response(tm)
resp.headers['Content-Type'] = tmContentType
return resp
def showMemento(urir, datetime):
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
print('Resolving request for {0} at {1}'.format(urir, datetime))
print('Found {0} cdxj entrie(s) for '.format(len(cdxjLinesWithURIR)))
print('MEMENTOS:')
print(cdxjLinesWithURIR)
closestLine = getCDXJLineClosestTo(datetime, cdxjLinesWithURIR)
print "best line: " + closestLine
if closestLine is None:
msg = '<h1>ERROR 404</h1>'
msg += 'No capture found for {0} at {1}.'.format(path, datetime)
return Response(msg, status=404)
uri = unsurt(closestLine.split(' ')[0])
return show_uri(uri, datetime)
def showWebUI(path):
webuiPath = '/'.join(('webui', path)).replace('ipwb.replay', 'ipwb')
content = pkg_resources.resource_string(__name__, webuiPath)
if 'index.html' in path:
iFile = ipwbConfig.getIPWBReplayIndexPath()
if iFile is None or iFile == '':
iFile = pkg_resources.resource_filename(__name__, INDEX_FILE)
if not os.path.isabs(iFile): # Convert rel to abs path
iFileAbs = pkg_resources.resource_filename(__name__, iFile)
if os.path.exists(iFileAbs):
iFile = iFileAbs # Local file
content = content.replace('MEMCOUNT', str(retrieveMemCount(iFile)))
content = content.replace(
'var uris = []',
'var uris = {0}'.format(getURIsAndDatetimesInCDXJ(iFile)))
content = content.replace('INDEXSRC', iFile)
return Response(content)
def showMemento(urir, datetime):
""" Request a URI-R at a supplied datetime from the CDXJ """
urir = getCompleteURI(urir)
if ipwbConfig.isLocalHosty(urir):
urir = urir.split('/', 4)[4]
s = surt.surt(urir, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
print('Getting CDXJ Lines with the URI-R {0} from {1}'.format(
urir, indexPath))
cdxjLinesWithURIR = getCDXJLinesWithURIR(urir, indexPath)
closestLine = getCDXJLineClosestTo(datetime, cdxjLinesWithURIR)
if closestLine is None:
msg = '<h1>ERROR 404</h1>'
msg += 'No capture found for {0} at {1}.'.format(urir, datetime)
return Response(msg, status=404)
uri = unsurt(closestLine.split(' ')[0])
newDatetime = closestLine.split(' ')[1]
if newDatetime != datetime:
return redirect('/memento/{0}/{1}'.format(newDatetime, urir), code=302)
return show_uri(uri, newDatetime)
def show_uri(path, datetime=None):
global IPFS_API
if len(path) == 0:
return showWebUI('index.html')
# TODO: Use a better approach to serve static contents
# instead of using the same logic for every JS file as the SW script
localScripts = [
'serviceWorker.js', 'reconstructive.js', 'reconstructive-banner.js'
]
if path in localScripts:
return getServiceWorker(path)
daemonAddress = '{0}:{1}'.format(IPFSAPI_IP, IPFSAPI_PORT)
if not ipwbConfig.isDaemonAlive(daemonAddress):
errStr = ('IPFS daemon not running. '
'Start it using $ ipfs daemon on the command-line '
' or from the <a href="/">'
'IPWB replay homepage</a>.')
return Response(errStr, status=503)
path = getCompleteURI(path)
cdxjLine = ''
try:
surtedURI = surt.surt(path,
path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
searchString = surtedURI
if datetime is not None:
searchString = surtedURI + ' ' + datetime
cdxjLine = getCDXJLine_binarySearch(searchString, indexPath)
print('CDXJ Line: {0}'.format(cdxjLine))
except Exception as e:
print(sys.exc_info()[0])
respString = ('{0} not found :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
if cdxjLine is None: # Resource not found in archives
return generateNoMementosInterface(path, datetime)
cdxjParts = cdxjLine.split(" ", 2)
jObj = json.loads(cdxjParts[2])
datetime = cdxjParts[1]
digests = jObj['locator'].split('/')
class HashNotFoundError(Exception):
pass
try:
def handler(signum, frame):
raise HashNotFoundError()
signal.signal(signal.SIGALRM, handler)
signal.alarm(10)
payload = IPFS_API.cat(digests[-1])
header = IPFS_API.cat(digests[-2])
signal.alarm(0)
except ipfsapi.exceptions.TimeoutError:
print("{0} not found at {1}".format(cdxjParts[0], digests[-1]))
respString = ('{0} not found in IPFS :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
except TypeError:
print('A type error occurred')
print(traceback.format_exc())
print(sys.exc_info()[0])
except HashNotFoundError:
print("Hashes not found")
return '', 404
except Exception as e:
print('Unknown exception occurred while fetching from ipfs.')
print(sys.exc_info()[0])
sys.exit()
if 'encryption_method' in jObj:
keyString = None
while keyString is None:
if 'encryption_key' in jObj:
keyString = jObj['encryption_key']
else:
askForKey = ('Enter a path for file',
' containing decryption key: \n> ')
keyString = raw_input(askForKey)
encryptionMethod = None
if jObj['encryption_method'] == 'xor':
encryptionMethod = XOR
pKey = encryptionMethod.new(keyString)
payload = pKey.decrypt(base64.b64decode(payload))
hKey = encryptionMethod.new(keyString)
header = hKey.decrypt(base64.b64decode(header))
hLines = header.split('\n')
hLines.pop(0)
resp = Response(payload)
for idx, hLine in enumerate(hLines):
k, v = hLine.split(': ', 1)
if k.lower() == 'transfer-encoding' and v.lower() == 'chunked':
try:
unchunkedPayload = extractResponseFromChunkedData(payload)
except Exception as e:
continue # Data may have no actually been chunked
resp.set_data(unchunkedPayload)
if k.lower() not in ["content-type", "content-encoding"]:
k = "X-Archive-Orig-" + k
resp.headers[k] = v
# Add ipwb header for additional SW logic
newPayload = resp.get_data()
ipwbjsinject = """<script src="/webui/webui.js"></script>
<script>injectIPWBJS()</script>"""
newPayload = newPayload.replace('</html>', ipwbjsinject + '</html>')
resp.set_data(newPayload)
resp.headers['Memento-Datetime'] = ipwbConfig.datetimeToRFC1123(datetime)
# Get TimeMap for Link response header
respWithLinkHeader = getLinkHeaderAbbreviatedTimeMap(path, datetime)
resp.headers['Link'] = respWithLinkHeader.replace('\n', ' ')
return resp
def show_uri(path):
global IPFS_API
if len(path) == 0:
return showWebUI('index.html')
sys.exit()
daemonAddress = '{0}:{1}'.format(IPFSAPI_IP, IPFSAPI_PORT)
if not ipwbConfig.isDaemonAlive(daemonAddress):
errStr = ('IPFS daemon not running. '
'Start it using $ ipfs daemon on the command-line '
' or from the <a href="/">'
'IPWB replay homepage</a>.')
return Response(errStr)
# show the user profile for that user
cdxjLine = ''
try:
s = surt.surt(path, path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
cdxjLine = getCDXJLine_binarySearch(s, indexPath)
except:
print sys.exc_info()[0]
respString = ('{0} not found :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
if cdxjLine is None: # Resource not found in archives
return Response(status=404)
cdxjParts = cdxjLine.split(" ", 2)
jObj = json.loads(cdxjParts[2])
digests = jObj['locator'].split('/')
try:
payload = IPFS_API.cat(digests[-1], timeout=1)
header = IPFS_API.cat(digests[-2])
except ipfsapi.exceptions.TimeoutError:
print "{0} not found at {1}".format(cdxjParts[0], digests[-1])
respString = ('{0} not found in IPFS :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
except:
print sys.exc_info()[0]
print "general error"
sys.exit()
if 'encryption_method' in jObj:
pKey = XOR.new(jObj['encryption_key'])
payload = pKey.decrypt(base64.b64decode(payload))
hKey = XOR.new(jObj['encryption_key'])
header = hKey.decrypt(base64.b64decode(header))
hLines = header.split('\n')
hLines.pop(0)
resp = Response(payload)
for idx, hLine in enumerate(hLines):
k, v = hLine.split(': ', 1)
if k.lower() != "content-type":
k = "X-Archive-Orig-" + k
resp.headers[k] = v
return resp
def show_uri(path, datetime=None):
global IPFS_API
if len(path) == 0:
return showWebUI('index.html')
sys.exit()
if path == 'serviceWorker.js':
return getServiceWorker(path)
sys.exit()
daemonAddress = '{0}:{1}'.format(IPFSAPI_IP, IPFSAPI_PORT)
if not ipwbConfig.isDaemonAlive(daemonAddress):
errStr = ('IPFS daemon not running. '
'Start it using $ ipfs daemon on the command-line '
' or from the <a href="/">'
'IPWB replay homepage</a>.')
return Response(errStr)
cdxjLine = ''
try:
surtedURI = surt.surt( # Good ol' pep8 line length
path,
path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbConfig.getIPWBReplayIndexPath()
searchString = surtedURI
if datetime is not None:
searchString = surtedURI + ' ' + datetime
cdxjLine = getCDXJLine_binarySearch(searchString, indexPath)
except:
print sys.exc_info()[0]
respString = ('{0} not found :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
if cdxjLine is None: # Resource not found in archives
msg = '<h1>ERROR 404</h1>'
msg += 'No capture found for {0} at {1}.'.format(path, datetime)
linesWithSameURIR = getCDXJLinesWithURIR(path)
if linesWithSameURIR:
msg += '<p>{0} capture(s) available:</p><ul>'.format(
len(linesWithSameURIR))
for line in linesWithSameURIR:
fields = line.split(' ', 2)
msg += ('<li><a href="/{1}/{0}">{0} at {1}</a></li>'.format(
unsurt(fields[0]), fields[1]))
msg += '</ul>'
return Response(msg, status=404)
cdxjParts = cdxjLine.split(" ", 2)
jObj = json.loads(cdxjParts[2])
datetime = cdxjParts[1]
digests = jObj['locator'].split('/')
try:
payload = IPFS_API.cat(digests[-1], timeout=1)
header = IPFS_API.cat(digests[-2])
except ipfsapi.exceptions.TimeoutError:
print "{0} not found at {1}".format(cdxjParts[0], digests[-1])
respString = ('{0} not found in IPFS :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_IP, IPWBREPLAY_PORT)
return Response(respString)
except:
print sys.exc_info()[0]
print "general error"
sys.exit()
if 'encryption_method' in jObj:
pKey = XOR.new(jObj['encryption_key'])
payload = pKey.decrypt(base64.b64decode(payload))
hKey = XOR.new(jObj['encryption_key'])
header = hKey.decrypt(base64.b64decode(header))
hLines = header.split('\n')
hLines.pop(0)
resp = Response(payload)
for idx, hLine in enumerate(hLines):
k, v = hLine.split(': ', 1)
if k.lower() == 'transfer-encoding' and v.lower() == 'chunked':
resp.set_data(extractResponseFromChunkedData(payload))
if k.lower() != "content-type":
k = "X-Archive-Orig-" + k
resp.headers[k] = v
resp.headers['Memento-Datetime'] = ipwbConfig.datetimeToRFC1123(datetime)
return resp
def show_uri(path, datetime=None):
global IPFS_API
if len(path) == 0:
return showWebUI('index.html')
# TODO: Use a better approach to serve static contents
# instead of using the same logic for every JS file as the SW script
localScripts = [
'serviceWorker.js', 'reconstructive.js', 'reconstructive-banner.js'
]
if path in localScripts:
return getServiceWorker(path)
daemonAddress = '{0}:{1}'.format(IPFSAPI_HOST, IPFSAPI_PORT)
if not ipwbUtils.isDaemonAlive(daemonAddress):
errStr = ('IPFS daemon not running. '
'Start it using $ ipfs daemon on the command-line '
' or from the <a href="/">'
'IPWB replay homepage</a>.')
return Response(errStr, status=503)
path = getCompleteURI(path)
cdxjLine = ''
try:
surtedURI = surt.surt(path,
path_strip_trailing_slash_unless_empty=False)
indexPath = ipwbUtils.getIPWBReplayIndexPath()
searchString = surtedURI
if datetime is not None:
searchString = surtedURI + ' ' + datetime
cdxjLine = getCDXJLine_binarySearch(searchString, indexPath)
print('CDXJ Line: {0}'.format(cdxjLine))
except Exception as e:
print(sys.exc_info()[0])
respString = ('{0} not found :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_HOST, IPWBREPLAY_PORT)
return Response(respString)
if cdxjLine is None: # Resource not found in archives
return generateNoMementosInterface(path, datetime)
cdxjParts = cdxjLine.split(" ", 2)
jObj = json.loads(cdxjParts[2])
datetime = cdxjParts[1]
digests = jObj['locator'].split('/')
class HashNotFoundError(Exception):
pass
payload = None
header = None
try:
def handler(signum, frame):
raise HashNotFoundError()
if os.name != 'nt': # Bug #310
signal.signal(signal.SIGALRM, handler)
signal.alarm(10)
payload = IPFS_API.cat(digests[-1])
header = IPFS_API.cat(digests[-2])
if os.name != 'nt': # Bug #310
signal.alarm(0)
except ipfsapi.exceptions.TimeoutError:
print("{0} not found at {1}".format(cdxjParts[0], digests[-1]))
respString = ('{0} not found in IPFS :(' +
' <a href="http://{1}:{2}">Go home</a>').format(
path, IPWBREPLAY_HOST, IPWBREPLAY_PORT)
return Response(respString)
except TypeError as e:
print('A type error occurred')
print(e)
abort(500)
except HTTPError as e:
print("Fetching from the IPFS failed")
print(e)
abort(503)
except HashNotFoundError:
if payload is None:
print("Hashes not found:\n\t{0}\n\t{1}".format(
digests[-1], digests[-2]))
abort(404)
else: # payload found but not header, fabricate header
print("HTTP header not found, fabricating for resp replay")
header = ''
except Exception as e:
print('Unknown exception occurred while fetching from ipfs.')
print(e)
abort(500)
if 'encryption_method' in jObj:
keyString = None
while keyString is None:
if 'encryption_key' in jObj:
keyString = jObj['encryption_key']
else:
askForKey = ('Enter a path for file',
' containing decryption key: \n> ')
keyString = raw_input(askForKey)
paddedEncryptionKey = pad(keyString, AES.block_size)
key = base64.b64encode(paddedEncryptionKey)
nonce = b64decode(jObj['encryption_nonce'])
cipher = AES.new(key, AES.MODE_CTR, nonce=nonce)
header = cipher.decrypt(base64.b64decode(header))
payload = cipher.decrypt(base64.b64decode(payload))
hLines = header.split('\n')
hLines.pop(0)
status = 200
if 'status_code' in jObj:
status = jObj['status_code']
resp = Response(payload, status=status)
for idx, hLine in enumerate(hLines):
k, v = hLine.split(': ', 1)
if k.lower() == 'transfer-encoding' and v.lower() == 'chunked':
try:
unchunkedPayload = extractResponseFromChunkedData(payload)
except Exception as e:
continue # Data may have no actually been chunked
resp.set_data(unchunkedPayload)
if k.lower() not in ["content-type", "content-encoding", "location"]:
k = "X-Archive-Orig-" + k
resp.headers[k] = v
# Add ipwb header for additional SW logic
newPayload = resp.get_data()
ipwbjsinject = """<script src="/webui/webui.js"></script>
<script>injectIPWBJS()</script>"""
newPayload = newPayload.replace('</html>', ipwbjsinject + '</html>')
resp.set_data(newPayload)
resp.headers['Memento-Datetime'] = ipwbUtils.digits14ToRFC1123(datetime)
if header is None:
resp.headers['X-Headers-Generated-By'] = 'InterPlanetary Wayback'
# Get TimeMap for Link response header
# respWithLinkHeader = getLinkHeaderAbbreviatedTimeMap(path, datetime)
# resp.headers['Link'] = respWithLinkHeader.replace('\n', ' ')
if status[0] == '3' and isUri(resp.headers.get('Location')):
# Bad assumption that the URI-M will contain \d14 but works for now.
uriBeforeURIR = request.url[:re.search(r'/\d{14}/', request.url).end()]
newURIM = uriBeforeURIR + resp.headers['Location']
resp.headers['Location'] = newURIM
return resp
