Request: Restore download option or equivalent?

[JimmXinu]

Can we get the download option or an equivalent back? I use it with FanFicFare to download small cover images for stories.

Environment

• FlareSolverr version: v2.0.0
• Last working FlareSolverr version: v1.2.9
• Operating system: Win10
• Are you using Docker: no
• FlareSolverr User-Agent (see log traces or / endpoint):
• Are you using a proxy or VPN? no
• Are you using Captcha Solver: no
• If using captcha solver, which one:
• URL to test this issue:

Description

download option was removed, can't download images in v2.0.0.

Example: https://www.fanfiction.net/image/6721098/180/

Logged Error Messages

N/A

Screenshots

N/A

[ghost]

I also use this for downloading the raw HTML bytes for a page (without modification) to better handle some pages that have a mix of charsets in the page (eg where some content in the page is UTF8 and some is from windows-1252. Would be very useful to see it come back,

[ngosang]

I remove this feature since is not working with Puppeter + Firefox. There are 2 options:

• Request the main page with FlareSolverr. It will bypass Cloudflare and return the User-Agent and the cf_clearence cookie. You make your own code to make the download request.
• Implement this feature in FlareSolverr using the same logic but performing the download with NodeJS HTTP client.

[JimmXinu]

I would suggest the second option is better:

1. Restoring the download option is more backward compatible.
2. Doing it in FlareSolverr is cleaner for clients, which means you have fewer users asking for help when they(we) do it wrong. :-)

[Philipp5796]

This feature was removed, because it wasn't working with Puppeter + Firefox.
Since FlareSolverr v3 now uses Chrome, is this still valid, or can this feature added again? @ngosang

[ngosang]

I will revisit this in the next release.
FlareSolverr v1 used Puppeter + Chrome and now we are using Selenium + Chrome. I don't know if it's supported yet.

[acg5159]

Just seeing if this is possible now, I see a couple of stack overflow responses which seem like you can adjust the arguments you pass to chrome. They will tell chrome to auto download files of a certain type (image/jpg, etc.) instead of attempting to load them in a browser tab.

https://stackoverflow.com/questions/32205245/need-the-chrome-equivalent-of-these-firefox-browser-profile-settings-in-seleni

Could we pass these, and then when the file is downloaded in the docker image, send the file either trough a curl response as binary data or maybe have it download to some mounted docker location?

[maliayas]

Is there at least a workaround until the feature gets implemented? I want to download jpeg files but no way.

[ilike2burnthing]

Per @acg5159's link above, you could try adding the following after L138 in utils.py, changing the file types and download directory as you want:

    options.add_argument("browser.download.folderList=2")
    options.add_argument("browser.helperApps.neverAsk.saveToDisk=image/png")
    options.add_argument("browser.download.dir=C:\\Users\\Public\\Downloads")

As my home internet is currently being a dick, that is untested.

Failing that, you could use a script to make the initial request with FlareSolverr, return the cookie, and then use the cookie for future curl requests in the script.