def getListOfFromCSV(fileName):
data = []
with open(fileName, encoding = 'utf-8', errors = 'ignore') as f:
data = f.readlines()
list = []
for member in data:
member = messageCleaner.removeSpecialCharacter(member)
list.append(member)
return list
def __gatherWebMD(forumDetails):
url = forumDetails[0]
treatment = forumDetails[1]
# WebMD is problamtic in that it provides three types of ratings. To simplify the problem we will only be pulling the satisfaction rating, as this
# seems to be the most general one
forumName = 'Webmd.com'
results = requests.get(url)
soup = BeautifulSoup(results.content, 'html.parser')
rawPosts = soup.find_all(name="div", attrs={'class': 'userPost'})
posts = []
maxRating = 5
for rawPost in rawPosts:
try:
review = rawPost.find(name="p",
attrs={
'id': re.compile('comFull.*')
}).get_text()
rating = rawPost.find(name="div",
attrs={
'class': 'catRatings lastEl clearfix'
}).get_text()
except:
print(
" Error[__gatherWebMD(url)]= Threw an error whilst looking for a review or rating"
)
continue
# Removing special characters here
review = messageCleaner.removeSpecialCharacter(review)
review = review.replace("Comment:", "")
review = review.replace("Hide Full Comment", "")
rating = re.findall('\d+', rating)
if (review and rating):
# As mentioned, webmd provide three
forumPost = post(review, __scaleRatings(rating[0], maxRating), url)
posts.append(forumPost)
webMD = forum(forumName, url, maxRating, posts, treatment)
return webMD
def __gatherDrugsCom(forumDetails):
url = forumDetails[0]
treatment = forumDetails[1]
forumName = 'Drugs.com'
results = requests.get(url)
soup = BeautifulSoup(results.content, 'html.parser')
rawPosts = soup.find_all(name="div", attrs={'class': 'boxList'})
posts = []
maxRating = 10
# Extract the reviews and the ratings. If a post does not provide a review and a rating we will
# dismiss it because it will not be useful during the learning phase
for rawPost in rawPosts:
try:
review = rawPost.find(name="div", attrs={'class': 'user-comment'})
review = review.find(name="span").get_text()
rating = rawPost.find(name="div", attrs={
'class': 'rating-score'
}).get_text()
except:
logger.error(
" Error[__gatherDrugsCom(url)] = Threw an error whilst looking for a review or rating"
)
continue
# Removing special characters here
review = messageCleaner.removeSpecialCharacter(review)
if (review and rating):
forumPost = post(review, __scaleRatings(rating, maxRating), url)
posts.append(forumPost)
drugsComForum = forum(forumName, url, maxRating, posts, treatment)
return drugsComForum
def __gatherWebMD(forumDetails):
url = forumDetails[0]
treatment = forumDetails[1]
# WebMD is problamtic in that it provides three types of ratings. To simplify the problem we will only be pulling the satisfaction rating, as this
# seems to be the most general one
forumName = 'Webmd.com'
results = requests.get(url)
soup = BeautifulSoup(results.content, 'html.parser')
rawPosts = soup.find_all(name="div", attrs = {'class' : 'userPost'})
posts = []
maxRating = 5
for rawPost in rawPosts:
try:
review = rawPost.find(name = "p" , attrs = {'id' : re.compile('comFull.*')}).get_text()
rating = rawPost.find(name = "div" , attrs = {'class' : 'catRatings lastEl clearfix'}).get_text()
except:
print(" Error[__gatherWebMD(url)]= Threw an error whilst looking for a review or rating")
continue
# Removing special characters here
review = messageCleaner.removeSpecialCharacter(review)
review = review.replace("Comment:", "")
review = review.replace("Hide Full Comment", "")
rating = re.findall('\d+', rating)
if(review and rating):
# As mentioned, webmd provide three
forumPost = post(review, __scaleRatings(rating[0], maxRating), url)
posts.append(forumPost)
webMD = forum(forumName, url, maxRating, posts, treatment)
return webMD
def __gatherDrugsCom(forumDetails):
url = forumDetails[0]
treatment = forumDetails[1]
forumName = 'Drugs.com'
results = requests.get(url)
soup = BeautifulSoup(results.content, 'html.parser')
rawPosts = soup.find_all(name = "div", attrs = {'class' : 'boxList'})
posts = []
maxRating = 10
# Extract the reviews and the ratings. If a post does not provide a review and a rating we will
# dismiss it because it will not be useful during the learning phase
for rawPost in rawPosts:
try:
review = rawPost.find(name = "div", attrs = {'class' : 'user-comment'})
review = review.find(name = "span").get_text()
rating = rawPost.find(name = "div", attrs = {'class' : 'rating-score'}).get_text()
except:
logger.error(" Error[__gatherDrugsCom(url)] = Threw an error whilst looking for a review or rating")
continue
# Removing special characters here
review = messageCleaner.removeSpecialCharacter(review)
if(review and rating):
forumPost = post(review, __scaleRatings(rating, maxRating), url)
posts.append(forumPost)
drugsComForum = forum(forumName , url, maxRating, posts, treatment)
return drugsComForum
