20

I am using selenium webdriver to automate downloading several PDF files. I get the PDF preview window (see below), and now I would like to download the file. How can I accomplish this using Google Chrome as the browser?

Dialog Box

Riley Hun
  • 2,541
  • 5
  • 31
  • 77
  • Take a look at [this answer](https://stackoverflow.com/a/43471196/3846228)... maybe it'll help you. – dot.Py Jul 27 '17 at 11:35

6 Answers6

42

Try this code, it worked for me.

options = webdriver.ChromeOptions()
options.add_experimental_option('prefs', {
"download.default_directory": "C:/Users/XXXX/Desktop", #Change default directory for downloads
"download.prompt_for_download": False, #To auto download the file
"download.directory_upgrade": True,
"plugins.always_open_pdf_externally": True #It will not show PDF directly in chrome
})
self.driver = webdriver.Chrome(options=options)
Celius Stingher
  • 17,835
  • 6
  • 23
  • 53
Kumar
  • 421
  • 4
  • 2
  • Thanks! great answer! – Dimitris Tsarouhas Mar 13 '21 at 18:42
  • 2
    This didn't work for me until I changed the default directory to use backslash, so instead of "C:/Users/XXXX/Desktop" I use "C:\\Users\\XXXX\\Desktop". – Abang F. Dec 15 '21 at 02:45
  • 3
    What is `download.directory_upgrade` for? – Nam G VU Jul 21 '22 at 04:13
  • Ordinal0 [0x00A75230+1856048] BaseThreadInitThunk [0x76FDFA29+25] RtlGetAppContainerNamedObjectPath [0x77A37B5E+286] RtlGetAppContainerNamedObjectPath [0x77A37B2E+238] – rsc05 Oct 06 '22 at 01:17
  • Confirming this works using [Splinter](https://splinter.readthedocs.io/en/latest/elements-in-the-page.html) (based on Selenium) which doesn't do file downloads. – Liquidgenius Dec 11 '22 at 23:30
5

I found this piece of code somewhere on Stackoverflow itself and it serves the purpose for me without having to use selenium at all.

import urllib.request

response = urllib.request.urlopen(URL)    
file = open("FILENAME.pdf", 'wb')
file.write(response.read())
file.close()
Saravana
  • 59
  • 1
  • 2
  • 2
    This method will only work for non-authenticated sessions. It is not robust to websites which require a login. @Kumar's answer will work for both non-authenticated and authenticated sessions. – Liquidgenius Dec 11 '22 at 23:27
4

I did it and it worked, don't ask me how :)

options = webdriver.ChromeOptions()
options.add_experimental_option('prefs', {
#"download.default_directory": "C:/Users/517/Download", #Change default directory for downloads
#"download.prompt_for_download": False, #To auto download the file
#"download.directory_upgrade": True,
"plugins.always_open_pdf_externally": True #It will not show PDF directly in chrome 
})
driver = webdriver.Chrome(options=options)
Nick
  • 138,499
  • 22
  • 57
  • 95
user16072805
  • 49
  • 1
  • Ordinal0 [0x00A75230+1856048] BaseThreadInitThunk [0x76FDFA29+25] RtlGetAppContainerNamedObjectPath [0x77A37B5E+286] RtlGetAppContainerNamedObjectPath [0x77A37B2E+238] – rsc05 Oct 06 '22 at 01:18
3

You can download the pdf (Embeded pdf & Normal pdf) from web using selenium.

from selenium import webdriver

download_dir = "C:\\Users\\omprakashpk\\Documents" # for linux/*nix, download_dir="/usr/Public"
options = webdriver.ChromeOptions()

profile = {"plugins.plugins_list": [{"enabled": False, "name": "Chrome PDF Viewer"}], # Disable Chrome's PDF Viewer
               "download.default_directory": download_dir , "download.extensions_to_open": "applications/pdf"}
options.add_experimental_option("prefs", profile)
driver = webdriver.Chrome('C:\\chromedriver\\chromedriver_2_32.exe', chrome_options=options)  # Optional argument, if not specified will search path.

driver.get(`pdf_url`)

It will download and save the pdf in directory specified. Change the download_dir location and chrome driver location as per your convenience.

You can download chrome driver from here.

Hope it helps!

Om Prakash
  • 2,675
  • 4
  • 29
  • 50
  • this works with gui, if I add `options.add_argument('headless') ` it doesn't work. Any idea why? – jaggi Feb 14 '18 at 11:31
  • Try `add_argument("--headless")`. It works with python3. I am sure, it will work for python 2 also. – Om Prakash Feb 14 '18 at 11:37
  • I'm also using python3. it might be working for other pdf links but for AWS S3 links, it's not working. eg:`http://spark-public.s3.amazonaws.com/nlp/slides/AdvancedMaxent.pdf `. Even wget doesn't for aws links. I'm not sure how aws checks you whether you are in gui mode or not. – jaggi Feb 14 '18 at 16:00
  • it seems that 'not allowing' file downloads in headless mode is a security feature https://bugs.chromium.org/p/chromium/issues/detail?id=696481#c39 – jaggi Mar 03 '18 at 06:26
  • @ Om Prakash， have you tested your code with mode of headless chrome? Because I tested the code from your github page in headless chrome and it didn't work. – exteral Jun 23 '18 at 09:11
  • @Om Prakash, If I would like to download an xml? –  Mar 29 '19 at 12:36
-1

You can download the PDF file using Python's requests library

import requests
pdf_url = driver.current_url       # Get Current URL
response = requests.get(pdf_url)
file_name = 'filename.pdf'
with open(file_name, 'wb') as f:
   f.write(response.content)
Ravi Teja
  • 45
  • 1
  • 10
-2

In My case it worked without any code modification,Just need to disabled the Chrome pdf viewer

Here are the steps to disable it

  1. Go into Chrome Settings
  2. Scroll to the bottom click on Advanced
  3. Under Privacy And Security - Click on "Site Settings"
  4. Scroll to PDF Documents
  5. Enable "Download PDF files instead of automatically opening them in Chrome"
Umer
  • 1,098
  • 13
  • 31