How to get Chrome headless to download files?

[rsshilli]

How do I download files using Chrome headless? I see that:

• Chrome has recently added support for downloading headless here: https://bugs.chromium.org/p/chromium/issues/detail?id=696481
• Selenium has added support for downloading here: Download file in headless chrome SeleniumHQ/selenium#5159

I'm having a hard time piecing together how I get this working from my Chimp world.

Versions:

• Chimp: 0.50.2
• Node.js: v6.10.3
• Java: 1.8.0_121
• Operation system: Windows 10

I've tried a number of things playing with the configuration, trying things like this below, but nothing works:

chimpOptions.webdriverio.desiredCapabilities = {
  chromeOptions: {
    args : ["--no-sandbox", "--headless", "--disable-gpu", "--window-size=1050,900"],
    prefs: {
      "credentials_enable_service": false,
      "Browser.setDownloadBehavior": {behavior : "allow", downloadPath: path.join((process.env.HOME || process.env.HOMEPATH || process.env.USERPROFILE), "Downloads")},
      "profile": {
        "password_manager_enabled": false,
        "default_content_setting_values": { "automatic_downloads": 1 }
      }
    }
  },
};

My mechanism for downloading a file is to click on the download link and then use fs to wait for the file on the file system, which never shows up :(. This worked fine before adding --headless

[ybatsiun]

I am facing the same issue, were you been able to solve it?

[ivoneijr]

Same issue.

[pelly]

Any progress on this? I'm having the same issue.

[rsshilli]

I have this working, but I did it without the "--headless" mode. I was previously under the impression that the only way to make Chimp test Chrome in a docker container was the use the headless mode. I was wrong. Just remove the headless and everything works fine, even when in a docker container or running on a machine without a monitor.

[HolyWalley]

@rsshilli did you ever get an error message like this

unknown error: DevToolsActivePort file doesn't exist (Driver info: chromedriver=2.39.562737 (dba483cee6a5f15e2e2d73df16968ab10b38a2bf),platform=Linux 4.4.0-1049-aws x86_64)

when you run not headless chrome in docker-container?

[pedymaster]

@rsshilli
Iam interested in this too. I have the same error as HolyWalley when removing the headless flag

[rsshilli]

You have to pass in the full path to the file, right? It looks like from the error message that you just passed in the filename itself (DevToolsActivePort). The browser won't know where to find it.

[wuthiago]

I am also facing this problem. Anyone able to make it work? Thanks

[rsshilli]

@wuthiago I made it work. Are you using the full path for the filename?

[wuthiago]

@rsshilli were you able to download files with the "--headless" mode on?

[rsshilli]

No, I never tried that, TBH.

…

On Thu, Jul 19, 2018 at 9:34 AM, wuthiago ***@***.***> wrote: @rsshilli <https://github.com/rsshilli> were you able to download files with the "--headless" mode on? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#679 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACXzBRAODZymhQ1-bXVd62HfDqkpKaePks5uIIrngaJpZM4R8Mnm> .

-- *Ryan Shillington* CTO CherryCircle Software, Inc. 1701 Trinity Street <https://maps.google.com/?q=1701+Trinity+Street+Austin,+TX+78712&entry=gmail&source=g> Austin, TX 78712 <https://maps.google.com/?q=1701+Trinity+Street+Austin,+TX+78712&entry=gmail&source=g> Direct: 512.426.4389 www.cherrycirclesoftware.com www.qbdvision.com

[komalanandpandey]

Hi,
did any one get luck on downloading file using "headless chrome" , please help! Regards!

[pelly]

No, I gave up and used xvfb. Headless chrome had all kinds of problems for me.

…

On Thu, Jul 19, 2018 at 10:52 AM, komalanandpandey ***@***.*** > wrote: Hi, did any one get luck on downloading file using "headless chrome" , please help! Regards! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#679 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAKQmFOp-quMdn3JVk-I-0ZAsU-NMPdNks5uIJ0bgaJpZM4R8Mnm> .

[c20xh2]

@komalanandpandey @wuthiago

Here's a solution that work for me :

Setting the driver options:

options = Options()
options.add_argument('--headless')
options.add_argument('--disable-gpu')
prefs = {'download.default_directory' : out_path}
options.add_experimental_option('prefs', prefs)

download_path = './output/'

driver = webdriver.Chrome('./chromedriver', chrome_options=options)

driver.command_executor._commands["send_command"] = ("POST", '/session/$sessionId/chromium/send_command')
params = {'cmd': 'Page.setDownloadBehavior', 'params': {'behavior': 'allow', 'downloadPath': download_path}}
command_result = driver.execute("send_command", params)

Downloading a file:

driver.get(url)
time.sleep(20)

IMPORTANT : You have to call time.sleep(x) to let some time for the download to finish, if the driver close before the end of file download, solution won't work

[samhatoum]

This issue is being tracked in Chimpy
https://github.com/TheBrainFamily/chimpy/issues