Ekstrak pautan dan tarikh JavaScript tertentu daripada jadual HTML menggunakan Sup Cantik
Saya cuba menghuraikan dokumen HTML menggunakan kaedah sup dan FindALL yang cantik, tetapi saya tidak dapat mengasingkan maklumat yang saya perlukan. Saya telah melihat dokumentasi dan beberapa tutorial, mungkin kerana saya seorang pembangun junior, tetapi saya tidak dapat mengasingkan nombor dan pautan.
Ini ialah jadual HTML palsu dengan maklumat asas:
<tbody>
<tr class="results_row2">
<td align="left">
Text is here ispssgjj sgdhjksgd jhsgd sgd
</td>
<td align="left">
GHJSFAGHJSFA GAFGSH AGSHSAGJH
</td>
<td align="left">
hdjk sgdhjk fdhjk sdhjk sdghjk
</td>
<td align="center">
11/10/1964
</td>
<td align="left">
</td>
<td align="center">
5
</td>
<td align="center">
<a href="javascript:confirm_delete('informatjon I need to ignore IS HERE')">Delete</a>
<br>
<a href="javascript:PBC('information I need to grab via parse comes from here ')">LINK TITLE</a>
<br>
</td>
</tr>
</tbody>
Apabila saya menjalankan program, saya memerlukannya untuk mengekstrak yang berikut untuk setiap baris (iaitu baris): tarikh (tetapi disusun semula kepada YYMMDD, iaitu 641110) dan rentetan yang mengatakan "LINK GOES HERE" (tetapi saya perlu menggabungkannya dengan rentetan lain untuk menjadikannya pautan yang sah)
Saya tidak memerlukan sebarang maklumat tambahan seperti pautan ada di sini atau teks bercelaru (cth. Hjkhjksgd)
EDIT: Saya juga perlu log masuk ke lokasi web dengan kredibiliti yang betul (saya mempunyai kata laluan dan nama pengguna)
Semoga kod saya cukup jelas dan saya mempunyai beberapa cetakan untuk membantu saya memahami pembolehubah dsb. Saya juga terbuka kepada cara lain, saya tidak dapat memikirkan panda atau selenium yang cantik... Setakat ini saya mendapat ini:
import requests
from urllib.parse import urljoin
from bs4 import BeautifulSoup
#label the file location
file_location = r"Destination goes here"
#open the file up
with open(file_location, 'r') as f:
file = f.read()
#create a soup
soup= BeautifulSoup(file, "html.parser")
#print(f"soup is {soup}")
#find all the tags that match what we want
script = soup.findAll('td', id='center')
print('begning loop')
#this is to find the date I am going to make a separate loop to find the print certificate
#loop through the tags and check for what we want
for i in range (0, len(script)):
#these two variables are me trying to convert the tag to a variable to be used to check
scriptString = str(script[i])
scriptInt = int(script[i])
#print(f'Starting loop i is: {i}')
# Every 7th cell seems to be a number....
if((i+4)%7 == 0):
print(f'Starting IF i is: {i}')
print(f'int test is {scriptInt}')
#print(f'script is {script[i]} quote end')
#this was to find out which part of the string was a number and it's 80% accurate
#for j in range (0, len(scriptString)):
#print(f' j is {j} and string is {scriptString[j]}')
#this printed the YYMMDD
print(f'Rewritten the string is: "{scriptString[41]}{scriptString[42]}{scriptString[33]}{scriptString[34]}{scriptString[36]}{scriptString[37]}" quote end')
print("end")
Saya cuba mengeluarkan tali dari meja tetapi ia tidak kelihatan seperti int dan talinya sangat kacau. Oleh kerana kekacauan rentetan saya tidak dapat membandingkannya dengan apa yang saya mahukan. Oleh kerana terdapat berbilang tag td, saya tidak dapat mengasingkannya dengan td .
Bagi sesiapa yang cuba melakukan sesuatu yang serupa, berikut ialah beberapa kod bahasa Inggeris biasa dengan ruang letak, yang tidak akan disusun seperti sedia ada atas sebab ini... Terima kasih banyak atas bantuan jawapan! ! !
'''
To start this program you will need to go to Google Developer Tools
while you are in the website that you want to access
Go to the tab Network and right click Copy as Curl (Bash)
Copy that info to curlconverter.com to get the required code
'''
print('begin program')
#Import libraries
import requests
import os
from bs4 import BeautifulSoup
import pdfkit
import re
from datetime import datetime
'''
Get cookies by opening Google Chrome Develope Tools network tab, then sign in (or sign out and sign in if you are already signed in) and then click on login (probably the top one) and right click and then click copy (and as of this comment), it has an arrow and click on copy as curl (bash)
Curl your website using https://curlconverter.com/
it'll generate you some python code, copy and paste it below, but move the response to below
'''
# normally you will use url, but we want the welcome page to check to see if we have made a successful connection
response = requests.post(
'YOUR URL GOES HERE',
params=params,
cookies=cookies,
headers=headers,
data=data,
)
#store the file into soup and then parse it below to see if it goes to the login screen or welcome screen
soup = BeautifulSoup(response.content, 'html.parser')
#were putting a try here because soup might have been stored as a nonetype which isn't itterable so it'll also end the program if not caught
try:
#Check if it is stuck on the Login page to see if your cookies didn't work
if "LOGIN PAGE DESCRIMINATOR " in soup.find('title'):
print("ERROR cookies not getting you past the login screen you will have problems finding the PDFS")
raise UnicodeError
PDFArray = [First, second, etc.]
tempFileLocation = r"FILE LOCATION PATH GOES HERE"
totalErrors = 0
#these are the dates to search between.
lowerYear = 2020
higherYear = 2023
print('Connection & Login succesful, looping through PDFs Array\n')
#this loop goes through each of the Pagess in the array
for i in (range(0,len(PDFArray))):
#print(f'begining {PDFArray[i]}\'s loop which is {i+1} in the Array')
#declare the variables for this loop
grandTotalNumPDFs = grandTotalNumPDFs + totalNumPDFs
PDFFromArray = PDFArray[i]
#reseting these variables back to 0 for this loop, as this is now a new PDF so reset the totals etc.
totalNumPDFs = 0
previousDateObj = 0
#updating the URL and folder location to draw from and save to
url = f'Specific URL with your needed alterations to download PDFs'
folder_location = os.path.join(r"PATH TO THE FOLDER",str(PDFFromArray))
if not os.path.exists(folder_location):os.mkdir(folder_location)
#We want to pull the HTML from the website to parse
response = requests.post(
url,
params=params,
cookies=cookies,
headers=headers,
data=data,
)
#store the HTML in the soup from the response
soup = BeautifulSoup(response.content, 'html.parser')
#Let the Pages Know if the connection was succesful or not
print(f'For PDF:{PDFFromArray} it\'s {response.status_code==200} a connection was succesful and it\'s { the test you did above to see if it is stuck at the login page or not} that login was succesful')
#to ensure that we are logging in and not stuck at the login screen
if "Your login specific info for the IF statement" in soup.find('title'):
print(f"ERROR cookies not getting you past the login screen you will have problems finding the PDFS")
#if it comes up with a timberlake login then you want to stop processing to prevent useless PDFs
raise TypeError
#parse the first part by finding all the ones with the tag td and the (I cant remember what it's called) of align center
script = soup.findAll("td", align="center")
#this is to show the Pages stuff is happening etc. but not super useful and kinda cumbersome so commented out but used for debugging
#print(f"Begining the script loop for {PDFFromArray} it is {len(script)} tags long")
#loop through everything in the script to parse it
for i in script:
# We want to grab the date for later use.
try:
date_obj = datetime.strptime(i.text.strip(), "%m/%d/%Y")
month = str(date_obj.month).zfill(2) # zero padding
day = str(date_obj.day).zfill(2) # zero padding
#Catch the value error so it doesn't crash
except ValueError:
pass
#get rid of everything that's NOT a "a" tag
a_tags = i.findAll("a")
#further parsing of the "a" tag
if a_tags:
# parsing JavaScript
for a in a_tags:
pattern = r"\('(.*?)'\)"
#look for the particular stuff that has the javascript in it
match = re.search(pattern, a["href"])
if match:
#match group 1 gets the first parenthese group e.g. in the follwoing it will pick only javascript 1 javascript(1), javascript(2)
content = match.group(1)
#now we want ONLY the javascript that has our text which fortuneatly has this in it
if "The str to find only your content" in content:
#next check to see if the certificate is inside the date range we want
if(date_obj.year >= lowerYear and date_obj.year <= higherYear):
#Check to see if we need to add a suffix (e.g., -1 -2 -3 etc. to the filename and if so increase it otherwise reset to 0
if(date_obj == previousDateObj): duplicateNum = duplicateNum + 1
else: duplicateNum = 0
#save the current date to check the next one
previousDateObj = date_obj
#increase the print certificates by one
totalNumPDFs = totalNumPDFs + 1
#set the correct URL to obtain the html version of the certificate to then be converted to HTML
url = f"your URL goes here "
#if it's a duplicate date go ahead and add a dash and number otherwise just add the normal stuff
if duplicateNum > 0:
filename = f"{PDFFromArray}{str(date_obj.year)[-2:]}{month}{day}-{duplicateNum}.pdf"
duplicateNum = duplicateNum + 1
else:
filename = f"{PDFFromArray}{str(date_obj.year)[-2:]}{month}{day}.pdf"
#set the file path
file_path = os.path.join(folder_location,os.path.basename(filename))
#We've already done the cookies above so we just make a new request with the updated URL
response = requests.post(
url,
params=params,
cookies=cookies,
headers=headers,
data=data,
)
#if the response was good go ahead otherwise print an error
if(response.status_code == 200):
soup = response.content
#write the content to the tempFileLocation to be converted to a PDF in the next step
with open(tempFileLocation, 'wb') as f:
f.write(response.content)
#in case a file is open or other problem
try:
#write the file to a PDF
pdfkit.from_file(tempFileLocation, file_path)
except:
print(f'ERROR: Couldn\'t save {filename} because {file_path} is NOT VALID... or some other reason')
totalErrors = totalErrors + 1
#this isn't a critical error so just flag it and move on
pass
#else if this was something besides 200 ( a valid one) print an error
else: print(f"ERROR: STATUS CODE WAS NOT 200 for {PDFFromArray} with the url {url} when trying to write out the temporary file")
#else if this does not have a print certificate in it, then add one to the totalNumDelete
else: totalNumDelete = totalNumDelete + 1
#this is the end of the loop for the Pages, and loops to the next Pages in the array
print(f'End of searching through PDF:{PDFFromArray}\'s page, {totalNumPDFs} valid PDFs were found, between the dates of {lowerYear} - {higherYear} \n')
print(f'\nEnd of the program, program searched {len(PDFArray)} Pages and found, and downloaded {grandTotalNumPDFs} PDFs between the dates of {lowerYear} - {higherYear} and had {totalErrors} error(s) saving documents')
## These are to catch all the errors from the begining of the program
except TypeError:
print("PROGRAM END WITH ERROR: TYPE ERROR, meaning the Cookie probably is NO GOOD! or expired. and you got stuck at the login screen, and to prevent a bunch of PDFs with the login screen the program has been aborted")
except:
print('This is a general catch all ERROR no idea what went wrong here')