def interactWithPhotos(driver, interactionsTillNow, blackListedUsers,
archiveUsers):
interactionCount = 0
# Find all photos in the current page
# photolist = driver.find_elements_by_css_selector(articleClassSelector)
photolist = driver.find_elements_by_tag_name("article")
# Find like and archive button classes
optionList = webcrawl(photolist[0].get_attribute('innerHTML'),
"html.parser").contents[2].contents[0].findAll("a")
likeButtonClass = str(optionList[0].attrs["class"])
archiveButtonClass = str(optionList[2].attrs["class"])
likeButtonClassSelector = parser.concatenate2(likeButtonClass)
archiveButtonClassSelector = parser.concatenate2(archiveButtonClass)
likeButtons = driver.find_elements_by_css_selector(likeButtonClassSelector)
archiveButtons = driver.find_elements_by_css_selector(
archiveButtonClassSelector)
for i in range(0, len(photolist)):
# Return, if we reached max interactions allowed
if (interactionsTillNow +
interactionCount) == CONSTANTS.MAX_INTERACTIONS:
break
# Find the username to which the photo belongs
photoContainer = photolist[i].get_attribute('innerHTML')
usernameOfPhoto = str(
webcrawl(photoContainer, "html.parser").header.contents[1].
contents[0].contents[0].contents[0].contents[0])
# Don't interact with black listed users
if usernameOfPhoto in blackListedUsers:
utils.logMessage("User: "******" is blacklisted")
continue
shouldLike = shouldlikePhoto(photoContainer)
shouldArchive = shouldArchiveThePhoto(photoContainer, usernameOfPhoto,
archiveUsers)
if shouldLike:
likeButtons[i].click()
utils.logMessage("Liked photo of user: "******"Saved photo of user: "******"Interaction Count: " +
str(interactionCount + interactionsTillNow))
return interactionCount
def getData(meaning):
url = "https://wordsinasentence.com/" + meaning + "-in-a-sentence/"
print("Making a request to the URL: " + url)
return_val = u''
try:
# Put some headers
headers = {'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36'}
data = requests.get(url, headers=headers)
soup = webcrawl(data.text, "html.parser")
content = soup.find("div", {"class": "thecontent clearfix"})
list_of_p = content.find_all("p")
for p_tag in list_of_p:
# print(p_tag)
# print("\n\n\n")
if len(p_tag.contents) > 0:
unicode_string = p_tag.find(text=True, recursive=False)
if unicode_string is None:
continue
else:
# print("Unicode string is: " + unicode_string)
if "vocabulary videos" not in unicode_string and "YouTube channel" not in unicode_string:
return_val += unicode_string + u"\n"
return return_val
except Exception as e:
print("Exception raised in getData(" + meaning + ") method: " + str(e))
return return_val
def getStaus(name):
url = "https://www.quora.com/profile/" + name
print(url)
try:
# Put appropriate headers here later
headers = {
'User-Agent':
'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36'
}
data = requests.get(url, headers=headers)
soup = webcrawl(data.text, "html.parser")
statusTag = soup.find("span", \
class_="IdentityCredential UserCredential")
if hasattr(statusTag, "contents"):
return statusTag.contents[0]
else:
return "No status found"
except Exception as e:
print(e)
print('Extra info:' + '\nURL: ' + url)
def login(url, driver):
credentials = getCredentials(sys.argv[1])
loginPageData = openLoginPage(driver, url)
# Find the submit button on login page
submitClassName = parser.findSubmitButton(loginPageData)
submitClassName = parser.concatenate1(submitClassName)
# Find username and password
username = driver.find_element_by_name("username")
password = driver.find_element_by_name("password")
username.send_keys(credentials["username"])
password.send_keys(credentials["password"])
# Do the login
driver.find_element_by_css_selector(submitClassName).submit()
# Wait for the page to load
time.sleep(3 * CONSTANTS.WAIT_TIME)
homePageSrc = driver.page_source
homePageData = str(webcrawl(homePageSrc, "html.parser"))
return homePageData
def shouldlikePhoto(photoContainer):
# Find whether we have already liked the photo or not
likeStatusContent = str(
webcrawl(photoContainer,
"html.parser").contents[2].contents[0].contents[0])
return bool(re.search(CONSTANTS.WHITE_LIKE_BUTTON_CLASS,
likeStatusContent))
def shouldArchiveThePhoto(photoContainer, usernameOfPhoto, archiveUsers):
if usernameOfPhoto in archiveUsers:
# Find whether we have already saved the photo or not
archiveStatusContent = str(
webcrawl(photoContainer,
"html.parser").contents[2].contents[0].contents[2])
return bool(
re.search(CONSTANTS.WHITE_ARCHIVE_BUTTON_CLASS,
archiveStatusContent))
else:
return False
def watchStories(driver):
pageData = driver.find_element_by_tag_name("body").get_attribute(
'innerHTML')
storiesButtonClass = parser.findStoriesButton(
str(webcrawl(pageData, "html.parser")))
if parser.containsSpaces(storiesButtonClass):
storiesButtonClassSelector = parser.concatenate2(storiesButtonClass)
driver.find_elements_by_css_selector(
storiesButtonClassSelector).click()
else:
driver.find_element_by_class_name(storiesButtonClass).click()
def openLoginPage(driver, url):
# By default we go to the sign-up page
driver.get(url)
# Wait for the page to load
time.sleep(CONSTANTS.WAIT_TIME)
'''
On the sign-up page we need to
find the login button and click
it, so that we go to login page
'''
signUpPageSrc = driver.page_source
signUpPageData = str(webcrawl(signUpPageSrc, "html.parser"))
loginButton = parser.findLoginButton(signUpPageData)
driver.find_element_by_class_name(loginButton).click()
# Wait for the page to load
time.sleep(CONSTANTS.WAIT_TIME)
loginPageSrc = driver.page_source
loginPageData = str(webcrawl(loginPageSrc, "html.parser"))
return loginPageData
def parseQuoraQuestionFeed(filename):
# Read the HTML file
content = ""
with open(filename, "r") as f:
content = f.read()
'''
Start parsing the document for required entities:
We require the following:
(i) Question Name
(ii) Existing Answers
(iii) Number of followers
Note: This is a hardcoded version of HTML class attributes and
we need to change this to something concrete to make it error-prone
so that the code doesn't break if Quora changes its UI or HTML
skeleton. For instance, if Quora provides an API to access
topic-question-feed then we do not need to follow this approach
'''
soup = webcrawl(content, "html.parser")
questionList = soup.findAll(
"div", {"class": "QuestionFeedStory FeedStory feed_item"})
results = []
for question in questionList:
questionName = question.find('span', class_="ui_qtext_rendered_qtext")
existingNumAnswers = question.find('a',
class_="answer_count_prominent")
if existingNumAnswers.text == "No answer yet":
existingNumAnswers = 0
else:
existingNumAnswers = int(existingNumAnswers.text.split()[0])
followButton = question.find(
'div', class_="ItemComponent FollowActionItem primary_item")
if followButton == None:
followButton = question.find(
'div', class_="FollowActionItem ItemComponent primary_item")
followCountWrapper = followButton.find('span',
class_="icon_action_bar-count")
followerCount = followCountWrapper.find_all('span')[1].text
# Prepare a dictionary and keep appending to the list
result = {
"questionName": questionName.text,
"existingAnswers": existingNumAnswers,
"followers": followerCount
}
results.append(result)
return results