Cant't download in nobrowser mode - use normal browser interaction to download

[gorcer]

Hi,

I try to download file in nobrowser mode.

1. First try

My script have a one line:

https://doghouse.ru/pricelist/100sp_full.csv

Result:
There is no file in the directory

2. Second try:

My script:

download https://doghouse.ru/pricelist/100sp_full.csv to 100sp.csv

wait 5

Result:
File exists, but it size is zero!

To run script i'm use command on my Ubuntu machine:

tagui test.tag -n

Help please.

[siowyisheng]

ERROR - for headless and visible Chrome, download file using normal webpage interaction

You should get the error above. Currently, you will need to download the file through normal webpage interaction (clicking).

[gorcer]

I'm use normal browser interaction, see - first try

It's work in normal mode, but in nobrowser mode it's doesnt work.
File doesn't download.

[siowyisheng]

For normal browser interaction, you would need a link to click on to download the file. Visiting the URL won't work.

[gorcer]

Ok, i create simple page with DOWNLOAD link https://doghouse.ru/test

My script:

https://doghouse.ru/test
click DOWNLOAD
wait 10

My command:

tagui test.tag -n

START - automation started - Wed May 27 2020 14:01:16 GMT+1000 (+10)
https://doghouse.ru/test - 

click DOWNLOAD
wait 10

https://doghouse.ru/test - 
FINISH - automation finished - 10.5s

My result - there is now file in the directory,

P.S.: it's work with normal mode ( tagui test.tag -h)

[gorcer]

Rocket - it's cool)
What should I do?

[siowyisheng]

[gorcer]

What's about problem? You solve it?

[siowyisheng]

Oh I thought it was solved already.

chrome_step('Page.setDownloadBehavior',{behavior: 'allow', downloadPath: 'c:\\Users\\userid\\Downloads'});
Try this before clicking download. I think this is the same issue as #477

[gorcer]

It's doesn't work. It's freeze on start:

gorcer@www2:~/tagui/doghouse$ tagui test.tag -n

START - automation started - Thu May 28 2020 17:08:49 GMT+1000 (+10)

My script:

chrome_step('Page.setDownloadBehavior',{behavior: 'allow', downloadPath: '/home/gorcer/tagui/doghouse/'});

https://doghouse.ru/test
click DOWNLOAD
wait 5

NOBROWSER mode

Tested on Ubuntu16 Server and Debian8 Server

[siowyisheng]

tested

https://doghouse.ru/test
click DOWNLOAD

run with tagui my_flow.tag -h

-h headless is to not show a browser.
-n nobrowser cannot download the file.

[gorcer]

Okay (((

Headless mode does't work on some my servers and if it's work it take to 3x system resource more.

[kensoh]

Hi Egor, sorry I was away the past few months - tebelorg/RPA-Python#144

I see.. Yes -headless mode runs Chrome using invisible mode, and Chrome takes up a lot of memory. However, it is better for automation because it is more accurate than the -nobrowser mode. That mode uses PhantomJS web browser which is an old browser and more likely to encounter issues during automation with modern websites.