def __init__(self, url):
#SEED VARIABLES
self.url = url
self.baseurl = '/'.join(url.split('/')[0:3])
self.domain = url.split("/")[2]
self.regex = self.baseurl + r'/.*'
print(self.regex)
#INIT AND SETUP VARIABLES
self.programName = 'hyperScraper'
self.profileFolder = self.programName + "_profiles"
self.currentDomainPath = './{}/{}/'.format(self.profileFolder,
self.domain)
self.domainDataFolder = self.currentDomainPath + "data/"
self.domainExploreFile = self.currentDomainPath + 'explored.url'
self.domainVisitFile = self.currentDomainPath + 'visited.url'
self.profile_check_make(self.domain) #MAKE A PROFILE FOR DOMAIN
#RUNTIME VARIABLES
self.exploredURLMemory = mx.setload(self.domainExploreFile)
self.visitedURLMemory = mx.setload(self.domainVisitFile)
self.pendingURLMemory = list(self.exploredURLMemory -
self.visitedURLMemory)
#INITIAL SEED
self.visit(url) #initialize seed url and build links
def __init__(self, url, myregex=r'/.*'):
#SEED VARIABLES
self.url = url #same as input while initializing
self.domain = url.split("/")[2] #==www.asdasd.com
self.topLevelDomain = ".".join(self.domain.split('.')[1:])
self.baseurl = '/'.join(url.split('/')[:3]) #==https://www.asdasd.com
self.regex = self.baseurl + myregex
print("LOG: regex url matching pattern is >>", self.regex)
#INIT AND SETUP VARIABLES
self.programName = 'hyperScraper'
self.profileFolder = self.programName + "_profiles/"
self.currentDomainPath = self.profileFolder + f'{self.domain}/'
self.domainDataFolder = self.currentDomainPath + "data/"
self.domainExploreFile = self.currentDomainPath + 'explored.url'
self.domainVisitFile = self.currentDomainPath + 'visited.url'
self.profile_check_make(self.domain) #MAKE A PROFILE FOR DOMAIN
#INIT RUNTIME VARIABLES
self.exploredURLMemory = mx.setload(self.domainExploreFile)
self.visitedURLMemory = mx.setload(self.domainVisitFile)
self.pendingURLMemory = list(self.exploredURLMemory -
self.visitedURLMemory)
#INITIAL SEED
try:
self.visit(url) #initialize seed url and build links
except Exception as e:
print(
"ERROR: initial Seed encountered serious error change to {raise e} in this line to track it"
)
raise e
lexico_index_path='./Dictionary/lexico_index.set'; mx.touch(lexico_index_path)
lexico_dict_path='./Dictionary/lexico_dict.jsonl'; mx.touch(lexico_dict_path)
#NLP---------------------------------------
def get_lemma(WORD):
from spacy import load as spacy_load
try:
Cache.nlpengine
except:
Cache.nlpengine = spacy_load('en_core_web_sm',disable=['parser','ner', 'tagger', 'textcat'])
w=Cache.nlpengine(WORD)[0]
return w.lemma_.lower() if w.lemma_ != '-PRON-' else w.lower_
#CACHE LAYERS------------------------------
Cache.base_dict=mx.jload(basedict_path)
Cache.lexico_index=mx.setload(lexico_index_path)
Cache.lexico_dict=mx.jloadlines(lexico_dict_path)
#CORE--------------------------------------
def get_definition(WORD):
base_dict=Cache.base_dict
defn=base_dict.get(WORD) if base_dict.get(WORD) else ''
finds=re.findall(r'[\d].*?;',defn)
if finds:
return finds
else:
return [defn]
#--------------------------------------
def lexico_fetch(WORD): #lower Level
WORD=get_lemma(WORD)
while True:
def seed_dict_with_words():
#do seeding of dict with standard most common words.
words={POOL.apply_async(get_lemma, (x.lower(),)) for x in mx.setload('./Dictionary/3000common.set')} ;
words={x.get() for x in words}
add_words_to_dictionary(words)
import os
from mxproxy import mx
import re
def get_body(url, bodyclass=''):
if bodyclass:
return mx.get_page(url).find(class_=bodyclass)
else:
print('define bodyclass as parameter of this function')
if __name__ == '__main__':
SUBJECTFOLDER = 'SEM-5/DIP/'
url = 'https://www.sanfoundry.com/1000-digital-image-processing-questions-answers/'
page = mx.get_page_soup(url)
links = {x['href'] for x in page.select('.entry-content a')}
mx.touch(SUBJECTFOLDER + 'visited.url')
visited = mx.setload(SUBJECTFOLDER + 'visited.url')
pendinglinks = links - visited
for x in pendinglinks:
text = mx.get_page_soup(x).select_one('.entry-content').text
mx.fappend('SEM-5/DIP/BigData.txt', text)
mx.fappend(SUBJECTFOLDER + 'visited.url', x)
# print(text)
