Develop Reference

Python multiprocessing in for loop (requests and BeautifulSoup)

I have list of a lot of links and I want to use multiprocessing to speed the proccess, here is simplified version, I need it to be ordered like this:
I tried a lot of things, process, pool etc. I always had errors, I need to do it with 4 or 8 threads and make it ordered like this. Thank you for all help. Here is code:
from bs4 import BeautifulSoup
import requests
import time
links = ["http://www.tennisexplorer.com/match-detail/?id=1672704", "http://www.tennisexplorer.com/match-detail/?id=1699387", "http://www.tennisexplorer.com/match-detail/?id=1698990" "http://www.tennisexplorer.com/match-detail/?id=1696623", "http://www.tennisexplorer.com/match-detail/?id=1688719", "http://www.tennisexplorer.com/match-detail/?id=1686305"]
data = []
def essa(match, omega):
aaa = BeautifulSoup(requests.get(match).text, "lxml")
center = aaa.find("div", id="center")
p1_l = center.find_all("th", class_="plName")[0].find("a").get("href")
p2_l = center.find_all("th", class_="plName")[1].find("a").get("href")
return p1_l + " - " + p2_l + " - " + str(omega)
i = 1
start_time = time.clock()
for link in links:
data.append(essa(link, i))
i += 1
for d in data:
print(d)
print(time.clock() - start_time, "seconds")

Spawn several threads of the function and join them together:
from threading import Thread
def essa(match, omega):
aaa = BeautifulSoup(requests.get(match).text, "lxml")
center = aaa.find("div", id="center")
p1_l = center.find_all("th", class_="plName")[0].find("a").get("href")
p2_l = center.find_all("th", class_="plName")[1].find("a").get("href")
print p1_l + " - " + p2_l + " - " + str(omega)
if __name__ == '__main__':
threadlist = []
for index, url in enumerate(links):
t= Thread(target=essa,args=(url, index))
t.start()
threadlist.append(t)
for b in threadlist:
b.join()
You wont get them to print in order, for the simple reason that some http responses take longer than others.

As far I can understand you have the list of links and make requests concurrently to make the process faster. Here is the sample code for multithreading. I hope this will help you. Read the documentation for concurrent futures.
import concurrent.futures
import urllib.request
URLS = ['http://www.foxnews.com/',
'http://www.cnn.com/',
'http://europe.wsj.com/',
'http://www.bbc.co.uk/',
'http://some-made-up-domain.com/']
# Retrieve a single page and report the URL and contents
def load_url(url, timeout):
with urllib.request.urlopen(url, timeout=timeout) as conn:
return conn.read()
# We can use a with statement to ensure threads are cleaned up promptly
with concurrent.futures.ThreadPoolExecutor(max_workers=5) as executor:
# Start the load operations and mark each future with its URL
future_to_url = {executor.submit(load_url, url, 60): url for url in URLS}
for future in concurrent.futures.as_completed(future_to_url):
url = future_to_url[future]
try:
data = future.result()
except Exception as exc:
print('%r generated an exception: %s' % (url, exc))
else:
print('%r page is %d bytes' % (url, len(data)))

Python : Manga parsing return empty file

i want to parse images from a "certain" manga and chapter. here's my code:
import requests, bs4, os, urllib.request
try:
url = "http://manganelo.com/chapter/read_one_punch_man_manga_online_free3/chapter_136"
res = requests.get(url)
print("[+] Asking a request to " + url)
# slice the url so it only contains the name and chapter
name = url[34:].replace("/", "_")
os.mkdir(name)
print("[+] Making '{}' directory".format(name))
os.chdir(os.path.join(os.getcwd(), name))
soup = bs4.BeautifulSoup(res.text, "html.parser")
for img in soup.findAll("img"):
manga_url = img.get("src")
manga_name = img.get("alt") + ".jpg"
urllib.request.urlretrieve(manga_url, manga_name)
print("[+] Downloading: " + manga_name)
except Exception as e:
print("[-] Error: " + str(e))
it works fine BUT only for a specific chapter, let's say i put chapter 130, when i try to run the code it returns blank file but if i put chapter 136 or others it works fine. How can this happen?

you can replace urllib.request.urlretrieve(manga_url, manga_name)
with :
r = requests.get(manga_url, stream=True)
if r.status_code == 200:
with open(manga_name, 'wb') as f:
r.raw.decode_content = True
shutil.copyfileobj(r.raw, f)
Actually Remote server is apparently checking the user agent header and rejecting requests from Python's urllib.
On the other hand you can use :
opener = urllib.request.URLopener()
opener.addheader('User-Agent', 'whatever')
opener.retrieve(manga_url, manga_name)
This works for me
Hope this helps

Accessing a classes variable in python

I understand that this is a duplicate, but I havent had that "ah-ha" moment where I understand HOW to access the a classes variable. In this code, I am crawling a website from a list of thousands of pages. Those jobs are submitted via concurrent.futures.
I want to be able to return the value of "results". I've used self.results within def __init__(self, url_list, threads) and I cant seem to pull that variable when I try print(example.results.
If self.results is returning a value, but example.results isn't pulling it from if __name__ == '__main__':, how can you access that? I know I've done something wrong, but I don't know what it is.
from concurrent.futures import ThreadPoolExecutor
from proxy_def import *
import requests
from bs4 import BeautifulSoup
from parsers import *
site = 0
class ConcurrentListCrawler(object):
def __init__(self, url_list, threads):
self.urls = url_list
self.results = {}
self.max_threads = threads
def __make_request(self, url):
try:
r = requests.get(url=url, timeout=20)
r.raise_for_status()
print(countit(), r.url)
except requests.exceptions.Timeout:
r = requests.get(url=url, timeout=60)
except requests.exceptions.ConnectionError:
r = requests.get(url=url, timeout=60)
except requests.exceptions.RequestException as e:
raise e
return r.url, r.text
def __parse_results(self, url, html):
try:
print(url)
trip_data = restaurant_parse(url)
except Exception as e:
raise e
if trip_data:
print('here we go')
self.results = trip_data
#print(self.results)
return self.results
def wrapper(self, url):
url, html = self.__make_request(url)
self.__parse_results(url, html)
def run_script(self):
with ThreadPoolExecutor(max_workers=min(len(self.urls),self.max_threads)) as Executor:
jobs = [Executor.submit(self.wrapper, u) for u in self.urls]
if __name__ == '__main__':
listo = loadit()
print(listo)
print(len(listo))
example = ConcurrentListCrawler(listo, 10)
example.run_script()
print(example.results)
Any pointers would be greatly appreciated.

I believe one of your methods is not returning the results.
Make the following change.
def wrapper(self, url):
url, html = self.__make_request(url)
return self.__parse_results(url, html)
After this, I suggest you utilize the self.results as a dictionary, like it was declared.
In the method "__parse_results(..)", append trip_data to self.results as follows, instead of assigning.
def __parse_results(self, url, html):
try:
print(url)
trip_data = restaurant_parse(url)
except Exception as e:
raise e
if trip_data:
print('here we go')
self.results[url] = trip_data
#print(self.results)
return self.results
When you append to self.results, it would retain the older values and you may avoid replacing by reassignment.

The issue was that I submitted all the jobs at once through a list. I was unable to pull the variable from the class because print(example.results) because that part of the code isnt access until all jobs are complete. With that I was able to resolve by getting rid of the class (even though the title of this posting indicates that this is the issue).
from concurrent.futures import ThreadPoolExecutor
import concurrent
from proxy_def import *
import requests
from bs4 import BeautifulSoup
from parsers import *
site = 0
def load_url(url):
try:
print(countit(), url)
trip_data = restaurant_parse(url)
return trip_data
except Exception as e:
raise e
if __name__ == '__main__':
URLs = loadit()
#print(URLs)
#print(len(URLs))
with ThreadPoolExecutor(max_workers=10) as executor:
# start the load operations and mark each future with its URL
future_to_url = {executor.submit(load_url, url): url for url in URLs}
for future in concurrent.futures.as_completed(future_to_url):
url = future_to_url[future]
try:
data = future.result()
print('this is data', data)
except Exception as exc:
print('%r generated an exception: %s' % (url, exc))
Here, I can pull the dictionary by grabbing data.
Thanks for the help, everyone.

python request urls parallel [duplicate]

This question already has an answer here:
How to send multiple http requests python
(1 answer)
Closed 6 years ago.
I created the following script to download images from an API endpoint which works as intended. Thing is that it is rather slow as all the requests have to wait on each other. What is the correct way to make it possible to still have the steps synchronously for each item I want to fetch, but make it parallel for each individual item. This from an online service called
servicem8
So what I hope to achieve is:
fetch all possible job ids => keep name/and other info
fetch name of the customer
fetch each attachment of a job
These three steps should be done for each job. So I could make things parallel for each job as they do not have to wait on each other.
Update:
Problem I do not understand is how can you make sure that you bundle for example the three calls per item in one call as its only per item that I can do things in parallel so for example when I want to
fetch item( fetch name => fetch description => fetch id)
so its the fetch item I want to make parallel?
The current code I have is working but rather slow:
import requests
import dateutil.parser
import shutil
import os
user = "test#test.com"
passw = "test"
print("Read json")
url = "https://api.servicem8.com/api_1.0/job.json"
r = requests.get(url, auth=(user, passw))
print("finished reading jobs.json file")
scheduled_jobs = []
if r.status_code == 200:
for item in r.json():
scheduled_date = item['job_is_scheduled_until_stamp']
try:
parsed_date = dateutil.parser.parse(scheduled_date)
if parsed_date.year == 2016:
if parsed_date.month == 10:
if parsed_date.day == 10:
url_customer = "https://api.servicem8.com/api_1.0/Company/{}.json".format(item[
'company_uuid'])
c = requests.get(url_customer, auth=(user, passw))
cus_name = c.json()['name']
scheduled_jobs.append(
[item['uuid'], item['generated_job_id'], cus_name])
except ValueError:
pass
for job in scheduled_jobs:
print("fetch for job {}".format(job))
url = "https://api.servicem8.com/api_1.0/Attachment.json?%24filter=related_object_uuid%20eq%20{}".format(job[
0])
r = requests.get(url, auth=(user, passw))
if r.json() == []:
pass
for attachment in r.json():
if attachment['active'] == 1 and attachment['file_type'] != '.pdf':
print("fetch for attachment {}".format(attachment))
url_staff = "https://api.servicem8.com/api_1.0/Staff.json?%24filter=uuid%20eq%20{}".format(
attachment['created_by_staff_uuid'])
s = requests.get(url_staff, auth=(user, passw))
for staff in s.json():
tech = "{}_{}".format(staff['first'], staff['last'])
url = "https://api.servicem8.com/api_1.0/Attachment/{}.file".format(attachment[
'uuid'])
r = requests.get(url, auth=(user, passw), stream=True)
if r.status_code == 200:
creation_date = dateutil.parser.parse(
attachment['timestamp']).strftime("%d.%m.%y")
if not os.path.exists(os.getcwd() + "/{}/{}".format(job[2], job[1])):
os.makedirs(os.getcwd() + "/{}/{}".format(job[2], job[1]))
path = os.getcwd() + "/{}/{}/SC -O {} {}{}".format(
job[2], job[1], creation_date, tech.upper(), attachment['file_type'])
print("writing file to path {}".format(path))
with open(path, 'wb') as f:
r.raw.decode_content = True
shutil.copyfileobj(r.raw, f)
else:
print(r.text)
Update [14/10]
I updated the code in the following way with some hints given. Thanks a lot for that. Only thing I could optimize I guess is the attachment downloading but it is working fine now. Funny thing I learned is that you cannot create a CON folder on a windows machine :-) did not know that.
I use pandas as well just to try to avoid some loops in my list of dicts but not sure if I am already most performant. Longest is actually reading in the full json files. I fully read them in as I could not find an API way of just telling the api, return me only the jobs from september 2016. The api query function seems to work on eq/lt/ht.
import requests
import dateutil.parser
import shutil
import os
import pandas as pd
user = ""
passw = ""
FOLDER = os.getcwd()
headers = {"Accept-Encoding": "gzip, deflate"}
import grequests
urls = [
'https://api.servicem8.com/api_1.0/job.json',
'https://api.servicem8.com/api_1.0/Attachment.json',
'https://api.servicem8.com/api_1.0/Staff.json',
'https://api.servicem8.com/api_1.0/Company.json'
]
#Create a set of unsent Requests:
print("Read json files")
rs = (grequests.get(u, auth=(user, passw), headers=headers) for u in urls)
#Send them all at the same time:
jobs,attachments,staffs,companies = grequests.map(rs)
#create dataframes
df_jobs = pd.DataFrame(jobs.json())
df_attachments = pd.DataFrame(attachments.json())
df_staffs = pd.DataFrame(staffs.json())
df_companies = pd.DataFrame(companies.json())
#url_customer = "https://api.servicem8.com/api_1.0/Company/{}.json".format(item['company_uuid'])
#c = requests.get(url_customer, auth=(user, passw))
#url = "https://api.servicem8.com/api_1.0/job.json"
#jobs = requests.get(url, auth=(user, passw), headers=headers)
#print("Reading attachments json")
#url = "https://api.servicem8.com/api_1.0/Attachment.json"
#attachments = requests.get(url, auth=(user, passw), headers=headers)
#print("Reading staff.json")
#url_staff = "https://api.servicem8.com/api_1.0/Staff.json"
#staffs = requests.get(url_staff, auth=(user, passw))
scheduled_jobs = []
if jobs.status_code == 200:
print("finished reading json file")
for job in jobs.json():
scheduled_date = job['job_is_scheduled_until_stamp']
try:
parsed_date = dateutil.parser.parse(scheduled_date)
if parsed_date.year == 2016:
if parsed_date.month == 9:
cus_name = df_companies[df_companies.uuid == job['company_uuid']].iloc[0]['name'].upper()
cus_name = cus_name.replace('/', '')
scheduled_jobs.append([job['uuid'], job['generated_job_id'], cus_name])
except ValueError:
pass
print("{} jobs to fetch".format(len(scheduled_jobs)))
for job in scheduled_jobs:
print("fetch for job attachments {}".format(job))
#url = "https://api.servicem8.com/api_1.0/Attachment.json?%24filter=related_object_uuid%20eq%20{}".format(job[0])
if attachments == []:
pass
for attachment in attachments.json():
if attachment['related_object_uuid'] == job[0]:
if attachment['active'] == 1 and attachment['file_type'] != '.pdf' and attachment['attachment_source'] != 'INVOICE_SIGNOFF':
for staff in staffs.json():
if staff['uuid'] == attachment['created_by_staff_uuid']:
tech = "{}_{}".format(
staff['first'].split()[-1].strip(), staff['last'])
creation_timestamp = dateutil.parser.parse(
attachment['timestamp'])
creation_date = creation_timestamp.strftime("%d.%m.%y")
creation_time = creation_timestamp.strftime("%H_%M_%S")
path = FOLDER + "/{}/{}/SC_-O_D{}_T{}_{}{}".format(
job[2], job[1], creation_date, creation_time, tech.upper(), attachment['file_type'])
# fetch attachment
if not os.path.isfile(path):
url = "https://api.servicem8.com/api_1.0/Attachment/{}.file".format(attachment[
'uuid'])
r = requests.get(url, auth=(user, passw), stream = True)
if r.status_code == 200:
if not os.path.exists(FOLDER + "/{}/{}".format(job[2], job[1])):
os.makedirs(
FOLDER + "/{}/{}".format(job[2], job[1]))
print("writing file to path {}".format(path))
with open(path, 'wb') as f:
r.raw.decode_content = True
shutil.copyfileobj(r.raw, f)
else:
print("file already exists")
else:
print(r.text)

General idea is to use asynchronous url requests and there is a python module named grequests for that-https://github.com/kennethreitz/grequests
From Documentation:
import grequests
urls = [
'http://www.heroku.com',
'http://python-tablib.org',
'http://httpbin.org',
'http://python-requests.org',
'http://fakedomain/',
'http://kennethreitz.com'
]
#Create a set of unsent Requests:
rs = (grequests.get(u) for u in urls)
#Send them all at the same time:
grequests.map(rs)
And the resopnse
[<Response [200]>, <Response [200]>, <Response [200]>, <Response [200]>, None, <Response [200]>]

Multi-threaded Python Web Crawler Got Stuck

I'm writing a Python web crawler and I want to make it multi-threaded. Now I have finished the basic part, below is what it does:
a thread gets a url from the queue;
the thread extracts the links from the page, checks if the links exist in a pool (a set), and puts the new links to the queue and the pool;
the thread writes the url and the http response to a csv file.
But when I run the crawler, it always gets stuck eventually, not exiting properly. I have gone through the official document of Python but still have no clue.
Below is the code:
#!/usr/bin/env python
#!coding=utf-8
import requests, re, urlparse
import threading
from Queue import Queue
from bs4 import BeautifulSoup
#custom modules and files
from setting import config
class Page:
def __init__(self, url):
self.url = url
self.status = ""
self.rawdata = ""
self.error = False
r = ""
try:
r = requests.get(self.url, headers={'User-Agent': 'random spider'})
except requests.exceptions.RequestException as e:
self.status = e
self.error = True
else:
if not r.history:
self.status = r.status_code
else:
self.status = r.history[0]
self.rawdata = r
def outlinks(self):
self.outlinks = []
#links, contains URL, anchor text, nofollow
raw = self.rawdata.text.lower()
soup = BeautifulSoup(raw)
outlinks = soup.find_all('a', href=True)
for link in outlinks:
d = {"follow":"yes"}
d['url'] = urlparse.urljoin(self.url, link.get('href'))
d['anchortext'] = link.text
if link.get('rel'):
if "nofollow" in link.get('rel'):
d["follow"] = "no"
if d not in self.outlinks:
self.outlinks.append(d)
pool = Queue()
exist = set()
thread_num = 10
lock = threading.Lock()
output = open("final.csv", "a")
#the domain is the start point
domain = config["domain"]
pool.put(domain)
exist.add(domain)
def crawl():
while True:
p = Page(pool.get())
#write data to output file
lock.acquire()
output.write(p.url+" "+str(p.status)+"\n")
print "%s crawls %s" % (threading.currentThread().getName(), p.url)
lock.release()
if not p.error:
p.outlinks()
outlinks = p.outlinks
if urlparse.urlparse(p.url)[1] == urlparse.urlparse(domain)[1] :
for link in outlinks:
if link['url'] not in exist:
lock.acquire()
pool.put(link['url'])
exist.add(link['url'])
lock.release()
pool.task_done()
for i in range(thread_num):
t = threading.Thread(target = crawl)
t.start()
pool.join()
output.close()
Any help would be appreciated!
Thanks
Marcus

Your crawl function has an infinite while loop with no possible exit path.
The condition True always evaluates to True and the loop continues, as you say,
not exiting properly
Modify the crawl function's while loop to include a condition. For instance, when the number of links saved to the csv file exceeds a certain minimum number, then exit the while loop.
i.e.,
def crawl():
while len(exist) <= min_links:
...