PDF download from website (some files are with error and cannot be open)

[Ray_Pospisil]

Hi all,

I have already created a topic where BrandonB helped me a lot when teaching me how to download several PDF file from a website.

However, I am currently facing with an issue that when I download multiple PDF files simultaneously, not all of them are actually fully downloaded.

Some of the downloaded files are corrupted and cannot be opened (they only have 1 KB size).

Can anybody advise why is that? 

I attach the workflow. You can just change the output address from my rive to yours and you will be able to run the workflow yourselves.

All the best,

Radek

[grossal · Accepted solution]

Hi @Ray_Pospisil,

I think there is a simple fix for this by simply changing the configuration of the Download-Tool. I am not sure if both are needed, but I'd change the following:

1) Change the Connection-Settings

I'd downsized it to only 1 Connection but a higher timeout. This was simply trial-and-error. I first tried to only upsize the timeout but it started working when I also changed the connection to 1.

2) Add user-agent

This is probably optional the current point, but if you download more pages, the page might detect you as a bot and block you. Therefore it's a good idea to add a user-agent. I simply added the one from Google Chrome.

Result:

Let me know if this helped or if we need to dig further into it.

Best

Alex

[Ray_Pospisil · Accepted solution]

Hi Alex,

The solution seems to be working.

Thanks very much for your support.

I did not know how to use these features of the download tool.

Radek

[velisetty]

Hi, I have a similar issue and tried to configure based on the settings you mentioned but still, I get an error when I open a pdf - 

When I use a public URL, there isn't an issue but with the internal company URL, it doesn't work

ERROR - 

Can you please help?

Thanks,

Karishma

[velisetty]

@grossal - can you please help ?

[grossal]

Hi @velisetty,

sorry for the late response. I overlooked the mail during the week.

Two questions:

1) Can you show me how the download folder looks?

2) Did it ever work? I see that you used a username/password in the connection tab. How does the login in the website look/work?

Best

Alex

[velisetty]

Hello @grossal !

1) the output folder is like this -  I have tested it in multiple file locations but the pdf error is same. I even tried to download to Temporary file but the same error

2) it is basically a salesforce site. it works with SSO and whether I enter the username password or not, the pdf file is created and the download headers is HTTP/1.1 200 Connection established.

I have 1000's of URL and alteryx would be the best approach for me if it works. Can you please help ??

Thanks,

Karishma

[grossal]

Hi @velisetty,

I think the SSO is the issue. There is an idea to add SSO possibilities to the download tool:

https://community.alteryx.com/t5/Alteryx-Designer-Ideas/SSO-feature-in-Download-tool/idi-p/626121

I personally would work around it with the Python-Tool and the Selenium library - It's not too hard if you are a bit familiar with Python and obviously the Python Tool has to be allowed in your organization.

The issue here is, that I cannot describe a generic way for you. To make the login work, you need to do some basic webscrapping. I could show you how it's done in a quick 15 minutes Zooms/Teams call if you want.

Maybe someone else knows a non-Python way to do it.

Best

Alex