def canVisit(self, url):
topDomain = utils.getTopDomain(url)
retrRules = self.__rulesDict__.get(topDomain, None)
if retrRules is None: # Cache miss
robotsUrl = utils.robotsTxt(url)
roboFileBuf = utils.dlAndDecode(robotsUrl)
if not self.parseRobotFile(topDomain, roboFileBuf):
return False
retr = self.__rulesDict__[topDomain]
sp = tuple(filter(lambda a: a, url.split(topDomain)))
if sp:
firstCh = firstLetterCompile.search(sp[0])
if firstCh:
# Time to probe
fCh = firstCh.groups(1)[0]
retr = self.__rulesDict__[topDomain]['disallow']
compList = retr.get(fCh, None)
if compList:
for comp in compList:
if comp.search(sp[0]):
return False
return True
return True
def canVisit(self, url):
topDomain = utils.getTopDomain(url)
retrRules = self.__rulesDict__.get(topDomain, None)
if retrRules is None: # Cache miss
robotsUrl = utils.robotsTxt(url)
roboFileBuf = utils.dlAndDecode(robotsUrl)
if not self.parseRobotFile(topDomain, roboFileBuf):
return False
retr = self.__rulesDict__[topDomain]
sp = tuple(filter(lambda a: a, url.split(topDomain)))
if sp:
firstCh = firstLetterCompile.search(sp[0])
if firstCh:
# Time to probe
fCh = firstCh.groups(1)[0]
retr = self.__rulesDict__[topDomain]['disallow']
compList = retr.get(fCh, None)
if compList:
for comp in compList:
if comp.search(sp[0]):
return False
return True
return True
def extractFileUrls(url,
extCompile,
router,
depth=5,
httpDomain=utils.HTTPS_DOMAIN):
# Args: url, extCompile=> A pattern object of the extension(s) to match
# depth => An integer that indicates how deep to scrap
# Note: A negative recursion depth indicates that you want
# to keep crawling as far as the program can go
if not depth:
return
elif not restDriver.isCallableAttr(extCompile, 'search'):
utils.streamPrintFlush(
"Expecting a regex compiled object/result as arg 'extCompile'\n",
sys.stderr)
return
if not utils.httpHeadCompile.search(url):
url = "%s%s" % (httpDomain, url)
if not robotParser.canVisit(url):
print('Cannot visit %s due to /robots.txt rules' % (url))
return
decodedData = utils.dlAndDecode(url)
if not decodedData:
return
else:
urls = utils.urlCompile.findall(decodedData)
urls = [
utils.repeatHttpHeadCompile.sub(utils.HTTP_HEAD_REGEX, s)
for s in urls
]
plainUrls = []
matchedFileUrls = []
for u in urls:
pathSelector = plainUrls
regSearch = extCompile.search(u)
if regSearch:
rGroup = regSearch.groups(1)
u = '%s.%s' % (rGroup[0], rGroup[1])
pathSelector = matchedFileUrls
pathSelector.append(u)
uniqFileUrls = set(matchedFileUrls)
dlResults = [
pushUpJob(eachUrl, router, url) for eachUrl in uniqFileUrls
]
resultsList = [val for val in dlResults if val]
depth -= 1
for eachUrl in plainUrls:
extractFileUrls(eachUrl, extCompile, router, depth)
def extractFileUrls(url, extCompile, router, depth=5, httpDomain=utils.HTTPS_DOMAIN):
# Args: url, extCompile=> A pattern object of the extension(s) to match
# depth => An integer that indicates how deep to scrap
# Note: A negative recursion depth indicates that you want
# to keep crawling as far as the program can go
if not depth:
return
elif not restDriver.isCallableAttr(extCompile, 'search'):
utils.streamPrintFlush(
"Expecting a regex compiled object/result as arg 'extCompile'\n", sys.stderr
)
return
if not utils.httpHeadCompile.search(url):
url = "%s%s"%(httpDomain, url)
if not robotParser.canVisit(url):
print('Cannot visit %s due to /robots.txt rules'%(url))
return
decodedData = utils.dlAndDecode(url)
if not decodedData:
return
else:
urls = utils.urlCompile.findall(decodedData)
urls = [utils.repeatHttpHeadCompile.sub(utils.HTTP_HEAD_REGEX, s) for s in urls]
plainUrls = []
matchedFileUrls = []
for u in urls:
pathSelector = plainUrls
regSearch = extCompile.search(u)
if regSearch:
rGroup = regSearch.groups(1)
u = '%s.%s'%(rGroup[0], rGroup[1])
pathSelector = matchedFileUrls
pathSelector.append(u)
uniqFileUrls = set(matchedFileUrls)
dlResults = [pushUpJob(eachUrl, router, url) for eachUrl in uniqFileUrls]
resultsList = [val for val in dlResults if val]
depth -= 1
for eachUrl in plainUrls:
extractFileUrls(eachUrl, extCompile, router, depth)
def getFiles(url,
extCompile,
recursionDepth=5,
httpDomain=utils.HTTPS_DOMAIN,
baseDir=None):
# Args: url, extCompile=> A pattern object of the extension(s) to match
# recursionDepth => An integer that indicates how deep to scrap
# Note: A negative recursion depth indicates that you want
# to keep crawling as far as the program can go
if not recursionDepth:
return
elif not hasattr(extCompile, 'search'):
utils.streamPrintFlush(
"Expecting a pattern object/result of re.compile(..) for arg 'extCompile'\n",
sys.stderr)
return
if not utils.httpHeadCompile.search(url):
url = "%s%s" % (httpDomain, url)
decodedData = utils.dlAndDecode(url)
if not decodedData:
return
else:
urls = utils.urlCompile.findall(decodedData)
urls = list(
map(
lambda s: utils.repeatHttpHeadCompile.sub(
utils.HTTP_HEAD_REGEX, s), urls))
if not urls:
capableUrls = utils.urlCapableCompile.findall(decodedData)
trimmedHeadUrl = url.strip('/')
for capableUrl in capableUrls:
trimmed = capableUrl.strip('/')
fixedUrl = '%s/%s' % (trimmedHeadUrl, trimmed)
urls.append(fixedUrl)
plainUrls = []
matchedFileUrls = []
for u in urls:
pathSelector = plainUrls
regSearch = extCompile.search(u)
if regSearch:
g = regSearch.groups(1)
u = '%s.%s' % (g[0], g[1])
pathSelector = matchedFileUrls
pathSelector.append(u)
if not baseDir:
baseDir = os.path.abspath(".")
fullUrlToMemPath = os.path.join(baseDir,
utils.pathCleanseCompile.sub('_', url))
utils.createDir(fullUrlToMemPath)
# Time to download all the matched files
dlResults = []
for eachUrl in matchedFileUrls:
dlResults.append(dlData(eachUrl, fullUrlToMemPath))
resultsList = list(filter(lambda val: val, dlResults))
# Report to user successful saves
downloadCount = len(resultsList)
# print(downloadCount)
if not downloadCount:
# Mark this url as a bad one/miss and for the sake of crawling
# not hitting dead ends, we won't crawl it anymore unless otherwise specified
urlHash = getHash(url)
urlScoreTuple = missesDict.get(urlHash, None)
badCrawlCount = 0
if urlScoreTuple and len(urlScoreTuple) != 2:
badCrawlCount = (
urlScoreTuple[1]) + 1 # Increment the bad crawl score
missesDict[urlHash] = (url, badCrawlCount, time.time())
return # Cut this journey short
else:
utils.streamPrintFlush(
"For url %s downloaded %d files\n" % (url, downloadCount),
sys.stderr)
recursionDepth -= 1
for eachUrl in plainUrls:
getFiles(eachUrl,
extCompile,
recursionDepth,
baseDir=fullUrlToMemPath)
def getFiles(url, extCompile, recursionDepth=5, httpDomain=utils.HTTPS_DOMAIN, baseDir=None):
# Args: url, extCompile=> A pattern object of the extension(s) to match
# recursionDepth => An integer that indicates how deep to scrap
# Note: A negative recursion depth indicates that you want
# to keep crawling as far as the program can go
if not recursionDepth:
return
elif not hasattr(extCompile, 'search'):
utils.streamPrintFlush(
"Expecting a pattern object/result of re.compile(..) for arg 'extCompile'\n"
, sys.stderr)
return
if not utils.httpHeadCompile.search(url):
url = "%s%s"%(httpDomain, url)
decodedData = utils.dlAndDecode(url)
if not decodedData:
return
else:
urls = utils.urlCompile.findall(decodedData)
urls = list(
map(lambda s: utils.repeatHttpHeadCompile.sub(utils.HTTP_HEAD_REGEX, s), urls)
)
if not urls:
capableUrls = utils.urlCapableCompile.findall(decodedData)
trimmedHeadUrl = url.strip('/')
for capableUrl in capableUrls:
trimmed = capableUrl.strip('/')
fixedUrl = '%s/%s'%(trimmedHeadUrl, trimmed)
urls.append(fixedUrl)
plainUrls = []
matchedFileUrls = []
for u in urls:
pathSelector = plainUrls
regSearch = extCompile.search(u)
if regSearch:
g = regSearch.groups(1)
u = '%s.%s'%(g[0], g[1])
pathSelector = matchedFileUrls
pathSelector.append(u)
if not baseDir:
baseDir = os.path.abspath(".")
fullUrlToMemPath = os.path.join(baseDir, utils.pathCleanseCompile.sub('_', url))
utils.createDir(fullUrlToMemPath)
# Time to download all the matched files
dlResults = []
for eachUrl in matchedFileUrls:
dlResults.append(dlData(eachUrl, fullUrlToMemPath))
resultsList = list(filter(lambda val: val, dlResults))
# Report to user successful saves
downloadCount = len(resultsList)
# print(downloadCount)
if not downloadCount:
# Mark this url as a bad one/miss and for the sake of crawling
# not hitting dead ends, we won't crawl it anymore unless otherwise specified
urlHash = getHash(url)
urlScoreTuple = missesDict.get(urlHash, None)
badCrawlCount = 0
if urlScoreTuple and len(urlScoreTuple) != 2:
badCrawlCount = (urlScoreTuple[1]) + 1 # Increment the bad crawl score
missesDict[urlHash] = (url, badCrawlCount, time.time())
return # Cut this journey short
else:
utils.streamPrintFlush(
"For url %s downloaded %d files\n"%(url, downloadCount), sys.stderr
)
recursionDepth -= 1
for eachUrl in plainUrls:
getFiles(eachUrl, extCompile, recursionDepth, baseDir=fullUrlToMemPath)
